Photo by Sangharsh Lohakare on Unsplash

Every major pharmaceutical company on earth has spent decades and billions chasing the same goal: teaching the immune system to recognize and destroy cancer the way it already destroys a virus. It’s one of medicine’s oldest, most expensive, most repeatedly disappointed pursuits.

Ten days ago, one press release suggested someone may have actually gotten close. A personalized vaccine, built on the same platform behind a shot most of the world already has in its arm, hit both primary goals in a late-stage melanoma trial — fewer recurrences, less spread. It’s the first trial of its kind to succeed at this stage. Not a cure, not yet approved — but real. The stock jumped as much as 160% in a single session.

Five days later, on no comparable news, it fell nearly 20% in a day. Same company, same week, same story everyone was still arguing about.

My model doesn’t read headlines. It doesn’t know the stock moved 160%, or that it gave much of it back days later. It ranks all 450+ stocks the same quantitative way, every week — and this week, this one landed at #4.

Capital stays invested — core and pick both.

A free subscription gets you:

1. Weekly trade signals on SPY, BTC, and GLD + Real-time flip alerts.

2. The Stock I Buy Every Monday.

3. Live portfolio performance and holdings.

Every Sunday at 6:00 am ET, free. Subscribe here to get next week’s QuantRank in your inbox.

Subscribe now

Disclosure: I personally buy this QuantRank, in the live portfolio, every Monday after publication — never before.

Disclaimer: Educational and informational only. Not personalized investment advice, and not a recommendation to buy or sell any security. I am not a registered investment adviser. Past performance — backtested or live — does not guarantee future results. You can lose money, including a substantial amount. Consult a qualified financial adviser before investing. Full disclosures in The Patala Core-Plus Strategy.

Market digest: what moved this week

• Stocks stayed choppy near the record. SPY dipped early in the week, clawed back to $771 by Thursday, then eased into the weekend — still shy of the all-time closing high of $777.88 set August 13. The BUY signal, open since early April, didn’t move either way.

• Nvidia beat, but grew more slowly. Q2 revenue of $96.2B and EPS of $2.22 topped estimates, and Q3 guidance of $108B came in above consensus — but with growth decelerating from 106% to guided high-80s% YoY, the stock chopped rather than ran on the print.

• Inflation held steady, not cooler. July core PCE came in at 3.3% YoY, matching expectations and last month’s pace; consumer spending was flat. In-line, not reassuring — the Fed’s target is still a long way off.

• Jackson Hole left the market guessing. New Fed Chair Warsh called inflation “concerning” and didn’t rule out a hike, while also declining to commit to one — no new signal, just more uncertainty into September’s meeting.

• Gold cooled off. After last week’s breakout to multi-month highs, GLD gave back part of the move as the dollar firmed and Treasury yields stabilized.

• Bitcoin was the story of the week. BTC ripped from the low-$70,000s to a fresh high above $80,000 — and then, on the day of publication, the signal flipped to SELL as price pulled back to the high-$70,000s.

• New this week: the Bitcoin signal is live. Same system, same rules, zero discretion — now running on BTC alongside SPY and GLD.

Net effect: SPY up 0.47% on the week. GLD down 3.42%. BTC up sharply through Thursday before the pullback that triggered this week’s flip.

Never miss a Market Signal or Stock Recommendation. Subscribe Free For Limited Time

Subscribe now

Where all three signals stand

All three signals, side by side:

Same 8 weeks, indexed to the same starting point — this is what “reads price, not headlines” looks like across three completely different assets at once.

The honest part on gold: this BUY has been open 170 days and is still down 14.14% since it triggered. It’s the same number I published last week, just worse — the system doesn’t round the edges off a rough stretch to make it look better. Same standard of transparency as always.

The honest part on Bitcoin: the signal launched mid-breakout, caught a +6.36% move from $73,032 to $77,677 (Aug 20 → Aug 28), and flipped to SELL on the very day of publication. That’s not a curated result — it’s the actual sequence, published as it happened. Nothing was cherry-picked to make the debut look good or bad, and the fresh SELL now sits at 0 days with no track record of its own yet.

Why this week doesn’t change anything

SPY chopping around near its record didn’t make the system flinch in either direction. A five-week gold rally didn’t stop it from giving back ground normally this week. And Bitcoin’s flip to SELL — arriving on literally the first weekend this signal is public — didn’t get smoothed over, delayed, or explained away. It reads price, not headlines, and it reads all three assets exactly the same way. Next scheduled update: Saturday, September 5, 6:00 AM ET.

As always: this is not investment advice, and I’m not telling you what to do with your money. What I can tell you is what the discipline of the system calls for — follow the signal, don’t front-run the next update, and don’t let one hot week or one flip pull you outside the plan.

Subscribe free for a limited time

Subscribe now

Full transparency: every recent flip

Why add Bitcoin now

The system doesn’t care what the asset is — SPY, GLD, or BTC, it’s the same mechanical rules applied without opinion. Bitcoin has produced some of the largest gaps between buy-and-hold and signal-following in backtesting, precisely because of stretches like the one that just ended: a fast, emotional run that the system rode without getting greedy, and then exited without hesitation the moment the trend turned. Publishing it live, starting today, means you see every flip from here forward — the good ones and the bad ones, in real time.

A quick favor. If this issue was useful — especially the new Bitcoin signal — please subscribe and share it with a friend who’d benefit from 20+ years of original systematic research. It genuinely helps. Thank you for reading.

Subscribe free to follow The Patala Live Portfolio

Subscribe now

Disclaimer: This is not investment advice or a recommendation, and I am not a registered investment adviser. This publication is for informational and educational purposes only. The signals reflect backtested and now live-tracked quantitative models; past performance, backtested or live, does not guarantee future results. Cryptocurrency is especially volatile and involves risks distinct from traditional securities. Markets involve risk, including loss of capital. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

Bitcoin is too chaotic to trade with rules, right?

A quantitative signal turned Bitcoin’s worst drawdown from -75.57% to -43.00%, while still compounding at 135% a year.

Bitcoin doesn’t just crash in a panic. It’s had nine drawdowns of 7%+ since 2016, including two multi-year stretches that erased 73-76% of buy-and-hold value. The damage isn’t the crash — it’s the year-plus grind back to even.

That’s why this signal exists. Not to predict Bitcoin’s next move. Just discipline for when to be in it, and when to step aside.

Subscribe Free For Live $BTC Trade Signals.

Subscribe now

$10,000, Bitcoin, two paths, ten years

Same BTC. Same $10,000. One line holds it. The other follows the signal — in on BUY, cash on SELL.

The gains are nice. The drawdowns are the point.

Buy-and-hold Bitcoin has been cut in half or worse twice since 2016 — 75.57% in 2018, 73.01% in 2021-2022. The signal’s worst: -43.00%, recovered in 8 months. Across all nine completed drawdowns, the average time back to breakeven was under 6 months — the one exception is the current stretch, still underwater since August 2025.

The full dashboard, no cherry-picking

Better risk-adjusted return, shallower worst-case, at the same time. Verified in Portfolio Visualizer — not a spreadsheet model.

A few numbers from that table worth translating out of finance-speak: standard deviation (60.77% vs. 72.83%) means the ride itself is smoother, not just the worst day. Benchmark correlation of 0.81 means the signal still moves with Bitcoin most of the time — this isn’t a market-neutral strategy, it’s a better-timed version of the same asset. And an information ratio of 1.77 — the return earned per unit of tracking error versus just holding Bitcoin — is the kind of number that’s rare to see outside of quant funds with a lot more infrastructure than a weekly newsletter.

Annual returns, Jan 2016-Jul 2026

Four red years across eleven. Every one shallower than buy-and-hold’s equivalent down year. 2017 and 2021 show it kept pace on the good years too.

This is the kind of update most newsletters wait out. Subscribe if you want these, not just the highlight reels.

Subscribe now

How it works

Simple. No screen-watching. I publish the signal so you always know the position.

One honest note: this is a high-turnover system — about 69 trades a year, median hold 3 days. Bitcoin just moves faster than stocks or gold. Flip alerts go out same-day, not weekly.

Another one: there are no stop-losses. I’ve tried them, and they’ve never worked for me — they get whipsawed out of positions right before the move you were waiting for. The only exit is the signal itself flipping. If it says stay in, you stay in, drawdown and all.

What’s next

Live BTC signals start Saturday, August 29 — alongside SPY and GLD, flip alerts the moment the position changes.

Already published: SPY backtest · QQQ backtest · TQQQ backtest · GLD backtest

Same rules. One asset that never closes. Same discipline either way.

Subscribe free to follow The Patala Live Portfolio

Subscribe now

Disclaimer: I am not a registered investment adviser, and nothing here is personalized investment advice, a recommendation, or an offer of advisory services. This publication is for informational and educational purposes only. Performance figures are backtested using historical data and a mechanical application of the signal rule described above, verified independently in Portfolio Visualizer; backtested results are hypothetical, have inherent limitations, and do not guarantee future results. The strategy’s ~725 trades across the backtest period do not reflect transaction costs, slippage, or tax drag, all of which would reduce real-world returns. Past performance — backtested or live — does not guarantee future results. Bitcoin and other cryptocurrencies are highly volatile and speculative assets. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

NVDA was always a QuantRanks darling. Is it still, today, before earnings?

Fifteen separate times since 2016, NVDA held a top-3 QuantRank. Average return the year after: +112%. Here’s every one of those entries, in full:

Past is prologue — or is it? Keep reading.

NVIDIA reports earnings again today. A $5.16 trillion company, the most-watched print on the calendar, and every desk on the Street already has an opinion before the number even hits the tape. This is the market darling of the moment — the stock everyone already owns, everyone already believes in, and everyone already has a story about.

QuantRanks doesn’t care about any of that. It has a rank, not an opinion. Here’s the only question this newsletter actually asks: what’s NVDA’s QuantRank right now?

The snapshot

As of the August 25 close:

• Price: $213.05

• Market cap: $5.16T

• P/E (TTM): 32.63

None of that is what decides this. Keep reading.

The answer

Out of roughly 450 stocks scored monthly, that’s a stock that’s not in the top tier the strategy actually trades from — not a disaster, but not a standout either, on a list where “not a standout” is not the bar we’re grading against.

That’s the whole analysis, and it doesn’t depend on what NVDA reports this afternoon. A beat today doesn’t move NVDA from rank 145 into the top 20. A miss doesn’t either. The rank already reflects what the model thinks matters — earnings day is not new information to it.

Where the line actually is

I buy ranks 1 through 5. Every week, no exceptions — except one: when my SPY market-timing signal is a SELL, there’s no pick that week at all, top-5 rank or not. Capital goes to cash, full stop. The rank only matters when the signal already says the market’s worth being in.

Subscribe to get the top QuantRanks stocks I buy every Monday

Subscribe now

The backtest doesn’t just support that top five, though — it supports a wider tier than most people would guess, and it draws a real, visible line at exactly where that tier ends.

Ranks 1 through 10 — the top decile — averaged 37% to 41% over the following 12 months in the backtest. Ranks 11 through 20 look nothing like that: roughly 18% to 20%, statistically indistinguishable from ranks 30, 50, or 70. The cliff isn’t gradual. It happens once, right at the edge of the top ten, and then the line goes flat for the rest of the list.

Why concentration matters this much

This isn’t just a Patala finding. It’s one of the most replicated results in long-run market research.

Hendrik Bessembinder’s landmark study of nearly every U.S. stock since 1926 found that just 4% of stocks accounted for all the net wealth the stock market has ever created above cash — everything else, in aggregate, merely matched T-bills. His most recent update, covering a full century of data, found the concentration has only gotten tighter: 89 firms once accounted for half of the market’s $43 trillion in net wealth creation; the latest data puts that number at just 46 firms, even as total wealth creation more than doubled to $91 trillion. Widen the lens to global markets and the pattern holds — one large-scale study puts the top-performing 2.4% of firms worldwide behind effectively all $75.7 trillion in net global wealth creation from 1990 to 2020.

The lesson isn’t “buy 4% of the market and ignore the rest” — nobody can identify those names in advance with certainty, which is exactly why broad indexing works as a baseline. But it’s a real, data-backed reason why a systematic tilt toward a tiny, well-defined top tier — not the top half, not “good enough,” but the actual top of the list — is where a stock-picking edge would have to live if it exists at all. QuantRanks’ own backtest shows the same shape at a smaller scale: a real cliff at the very top, then a long flat plateau where extra selectivity stops paying off. NVDA, sitting well down that 450-stock list, is precisely that flat plateau. Being the market darling doesn’t move it off that plateau. Statistically, it’s no different from a coin flip against cash, not because it’s a bad company, but because “average” is what the plateau looks like, no matter how good the story is.

Verdict

Not a buy — based on the rank, not the earnings print.

If NVDA’s rank ever climbs into the top 10, it’s worth a second look. Into the top five, it’s a QuantRank pick, same as any other stock. At 145, it’s not close to either — beat or miss this afternoon, that doesn’t change.

The darling of 2016 through 2024 isn’t the darling of this list today. That’s not a knock on NVDA — it’s what the rank is for.

Every Sunday at 6:00 am ET, free. Subscribe here for the weekly QuantRank pick.

Subscribe now

Disclaimer: Educational and informational only. Not personalized investment advice, and not a recommendation to buy or sell any security. I am not a registered investment adviser. Past performance — backtested or live — does not guarantee future results. You can lose money, including a substantial amount. Consult a qualified financial adviser before investing.

The NVIDIA name and logo are trademarks of NVIDIA Corporation. Their use here is for identification purposes only and does not imply endorsement or affiliation.

Capital stays invested — core and pick both.

This week’s QuantRank

APA — APA Corporation

Current Ranking #3.

I’m buying Monday, Aug 24.

Why this pick, briefly

APA Corporation — the parent of what many still know as Apache — is an independent oil and gas exploration and production company. Its portfolio splits into two roles: steady cash generation today, and a bet on meaningful growth later.

The Permian Basin and Egypt do the heavy lifting now. Permian is APA’s largest source of both production and free cash flow; Egypt is where APA is the country’s largest oil producer and largest onshore acreage holder, running on a recently renegotiated gas-pricing agreement. The growth story sits offshore in Suriname, where APA holds a 50% stake (alongside operator TotalEnergies) in Block 58 — a major deepwater discovery targeting first oil around 2028.

A few numbers, as of Friday’s close:

• Closing price (Aug 21): $43.39

• Market cap: ~$15.202 billion

• P/E (TTM): 9.15

A single-digit-to-low-teens P/E is typical for an oil and gas producer, not a signal on its own — energy names trade on commodity price cycles more than the multiples that work for growth stocks. The rank isn’t asking you to have a view on oil prices either. That’s exactly the point of following it instead of a story.

How much to buy

Typical range: 1% to 3% of your capital.

The S&P 500 is at all-time highs right now. If that makes you nervous, size toward the lower end of that range. That’s a reasonable adjustment. Skipping the pick entirely is not.

Things to remember every week

1. This is educational, not advice. I built this system, and I’m sharing what I personally follow — with my own money. You don’t have to follow it too.

2. This is one QuantRank, published once a week. Nothing more to track between issues.

3. It’s a quant system. Trust the rank, not your gut. APA is a household name to anyone who follows energy, but that familiarity — or a view on oil prices — isn’t what put it here. The rank did the judging.

4. Strong predictive power over the long run — no guarantee on any single name. QuantRanks has shown real predictive power over time. That doesn’t stop any one stock, or a string of them, from losing money — in the short run or the long run.

5. A repeat within ranks 1 to 5 is a good sign, not a stale one. QuantRanks tend to bounce between the top five spots from month to month — if a stock keeps reappearing in that tier, that’s the model telling you conviction is building, not fading. Keep adding on the normal schedule.

6. No target price, no separate exit. A QuantRank is held until the signal — the market-wide one, not a per-stock rule — tells the whole portfolio to sell.

7. This newsletter is the QuantRank. The method is one link away. Full mechanics, sizing ladder, and honest risk disclosure are in The Patala Core-Plus Strategy.

Why QuantRanks predict return

Full backtest methodology and the broader QuantRanks study are in the deep dive.

If the signal flips

Signal turns SELL: Move 80% to 100% of the portfolio to cash or short-term notes — SHV, SGOV, or similar. Core and pick go together. No exceptions for a position you like.

Signal turns BUY again: Restart. Rebuild your core position in SPY or QQQ first. Then resume adding 1% to 3% into the top QuantRank each week, the same as always.

Every Sunday at 6:00 am ET, free. Subscribe here to get next week’s QuantRank in your inbox.

Subscribe now

Disclosure: I personally buy this QuantRank, in the live portfolio, every Monday after publication — never before.

Disclaimer: Educational and informational only. Not personalized investment advice, and not a recommendation to buy or sell any security. I am not a registered investment adviser. Past performance — backtested or live — does not guarantee future results. You can lose money, including a substantial amount. Consult a qualified financial adviser before investing. Full disclosures in The Patala Core-Plus Strategy.

• Bonds drove the bus. The 30-year Treasury yield pushed up near a two-decade high, rattling both stocks and expectations for the Fed’s next move.

• Treasury stepped in. Secretary Bessent doubled the size of long-bond buybacks, from $2 billion to at least $4 billion per operation — an attempt to calm the bond market that gave stocks a lift into Friday’s close.

• Chips got hit. Semiconductors sold off hard mid-week, even after Nvidia backed a large lease commitment tied to OpenAI, as investors grew warier about how the AI buildout gets financed. Nvidia reports earnings next Wednesday, August 26 — the whole sector is bracing for it.

• Gold did the opposite. A softer dollar and the Treasury’s liquidity moves sent gold to its highest levels in months, capping a fifth straight winning week.

Net effect: SPY eased off Wednesday’s near-record high, down 1.37% on the week. GLD broke out — up 5.45%.

Never miss a Market Signal or Stock Recommendation. Subscribe Free For Limited Time

Subscribe now

Where both signals stand

SPY: still a BUY, still holding since April 2 — no flip, no drama, just a system that doesn’t flinch in either direction.

Gold: also a BUY, and its timing looked good this week as GLD broke out to a fresh multi-month high.

Why this week doesn’t change anything

Two weeks ago, a record high didn’t make the system more aggressive. This week, a bond wobble and a chip selloff didn’t make it flinch either. It reads price, not headlines — good news or bad. Next scheduled update: Saturday, August 29, 6:00 AM ET.

As always: this is not investment advice, and I’m not telling you what to do with your money. What I can tell you is what the discipline of the system calls for — follow the signal, don’t front-run the next update, and don’t let one rough week or one hot week pull you outside the plan.

Subscribe free for a limited time

Subscribe now

Full transparency: every recent flip

A quick favor. If this issue was useful — especially the new gold signal — please subscribe and share it with a friend who’d benefit from 20+ years of original systematic research. It genuinely helps. Thank you for reading.

Subscribe free to follow The Patala Live Portfolio

Subscribe now

Disclaimer: This is not investment advice or a recommendation, and I am not a registered investment adviser. This publication is for informational and educational purposes only. The signals reflect backtested and now live-tracked quantitative models; past performance, backtested or live, does not guarantee future results. Markets involve risk, including loss of capital. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

Not really. It can be riskier than the S&P 500.

It sells off in a panic like everything else. The real damage comes after, in the years-long bear markets that have nothing to do with crisis at all — the 1980–2008 stretch alone was a 70% decline that took 28 years to recover, longer than it took the Dow to recover from the 1929 crash that triggered the Great Depression.

That’s the entire reason this trade signal exists. Not to predict gold. Just to have a disciplined system when to be in it and when to step aside

$10,000, GLD ETF, two paths, 21 years

Same GLD. Same $10,000. One line just holds it. The other follows the signal — in when it says BUY, cash when it says SELL, with a circuit breaker layered on top for violent down days.

From October 2005 to July 2026, buy & hold in GLD turned $10,000 into $79,559 (+696%). The signal turned it into $187,458 (+1,775%) — more than double the ending wealth, from the same underlying asset, with the same starting dollar.

The gains are nice. The drawdowns are the actual point.

Buy-and-hold gold spent four straight years underwater from 2011 to 2015, bottoming out at -42.9%. The signal’s worst drawdown across the same 21 years: -16.3%.

The full dashboard, no cherry-picking

Better risk-adjusted return, lower volatility, shallower worst-case, at the same time. Verified directly in Portfolio Visualizer — not a model that only runs in a spreadsheet.

Annual returns, Oct 2005–Jul 2026

Four red years across 21 — 2015, 2016, 2018, and 2022 — every one of them under -5%. Everything else, including 2026’s partial year, closed positive.

How it works

Very simple. There is no need to sit glued to a trading screen for 8 hours a day. When the system stays invested in GLD or Cash, I publish one trade signal weekly on Saturdays. On the day the signal flips from Buy → Sell or Sell → Buy, I will publish the Flip Alert after market close - to be traded the next day.

What’s next

Live GLD signals start Saturday, August 22 — published weekly alongside SPY, with flip alerts posted the moment either one changes.

Next up: BTC live signals — but first watch for the BTC backtest coming soon.

Already published: SPY backtest · QQQ backtest · TQQQ backtest

Same rules. One safety net. Same discipline either way.

Disclaimer: I am not a registered investment adviser, and nothing here is personalized investment advice, a recommendation, or an offer of advisory services. This publication is for informational and educational purposes only. Performance figures are backtested using historical data and a mechanical application of the signal and circuit-breaker rule described above, verified independently in Portfolio Visualizer; backtested results are hypothetical, have inherent limitations, and do not guarantee future results. Past performance — backtested or live — does not guarantee future results. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

Alibaba — BABA — reports earnings today, before the market opens. Look at that chart: up 58.8% against SPY at the peak, down 23.3% against it seven months later, and now sitting at roughly flat for the year. That’s a stock that’s given traders three different stories to tell themselves in twelve months.

The headlines this morning will add a fourth. Here’s the only question this newsletter actually asks: what’s BABA’s QuantRank right now?

The snapshot

As of the August 19 close:

• Price: $128.90

• Market cap: $308.957B

• P/E (TTM): 19.77

None of that is what decides this. Keep reading.

The answer

Out of roughly 450 stocks scored monthly, that lands BABA almost exactly in the middle of the pack — not a disaster, not a standout. Solidly average, on a list where average isn’t the bar.

That’s the whole analysis, and it doesn’t depend on what BABA reports this morning. A beat today doesn’t move BABA from rank 200 into the top 20. A miss doesn’t either. The rank already reflects what the model thinks matters — earnings day is not new information to it.

Where the line actually is

I buy ranks 1 through 5. Every week, no exceptions — except one: when my SPY market-timing signal is a SELL, there’s no pick that week at all, top-5 rank or not. Capital goes to cash, full stop. The rank only matters when the signal already says the market’s worth being in.

The backtest doesn’t just support that top five, though — it supports a wider tier than most people would guess, and it draws a real, visible line at exactly where that tier ends.

Ranks 1 through 10 — the top decile — averaged 37% to 41% over the following 12 months in the backtest. Ranks 11 through 20 look nothing like that: roughly 18% to 20%, statistically indistinguishable from ranks 30, 50, or 70. The cliff isn’t gradual. It happens once, right at the edge of the top ten, and then the line goes flat for the rest of the list.

Why concentration matters this much

This isn’t just a Patala finding. It’s one of the most replicated results in long-run market research.

Hendrik Bessembinder’s landmark study of nearly every U.S. stock since 1926 found that just 4% of stocks accounted for all the net wealth the stock market has ever created above cash — everything else, in aggregate, merely matched T-bills. His most recent update, covering a full century of data, found the concentration has only gotten tighter: 89 firms once accounted for half of the market’s $43 trillion in net wealth creation; the latest data puts that number at just 46 firms, even as total wealth creation more than doubled to $91 trillion. Widen the lens to global markets and the pattern holds — one large-scale study puts the top-performing 2.4% of firms worldwide behind effectively all $75.7 trillion in net global wealth creation from 1990 to 2020.

The lesson isn’t “buy 4% of the market and ignore the rest” — nobody can identify those names in advance with certainty, which is exactly why broad indexing works as a baseline. But it’s a real, data-backed reason why a systematic tilt toward a tiny, well-defined top tier — not the top half, not “good enough,” but the actual top of the list — is where a stock-picking edge would have to live if it exists at all. QuantRanks’ own backtest shows the same shape at a smaller scale: a real cliff at the very top, then a long flat plateau where extra selectivity stops paying off. BABA, sitting almost exactly at the midpoint of a 450-stock list, is precisely that flat plateau. Statistically, it’s no different from a coin flip against cash — not because it’s a bad company, but because “average” is what the plateau looks like.

Verdict

Not a buy — based on the rank, not the earnings print.

If BABA’s rank ever climbs into the top 10, it’s worth a second look. Into the top five, it’s a QuantRank pick, same as any other stock. At 200, it’s not close to either — beat or miss this morning, that doesn’t change.

Disclaimer: Educational and informational only. Not personalized investment advice, and not a recommendation to buy or sell any security. I am not a registered investment adviser. Past performance — backtested or live — does not guarantee future results. You can lose money, including a substantial amount. Consult a qualified financial adviser before investing.

The headlines will argue both sides within minutes. Here’s the only question this newsletter actually asks: what’s HD’s QuantRank right now?

The answer

Out of roughly 450 stocks scored monthly, that puts it deep in the bottom third — nowhere near even the outer edge of the tier that’s ever earned a QuantRanks buy.

That’s the whole analysis, and it doesn’t depend on what HD reports this morning. A beat today doesn’t move HD from rank 300+ into the top 20. A miss doesn’t either. The rank already reflects what the model thinks matters — earnings day is not new information to it.

Where the line actually is

I buy ranks 1 through 5. Every week, no exceptions — except one: when my SPY market-timing signal is a SELL, there’s no pick that week at all, top-5 rank or not. Capital goes to cash, full stop. The rank only matters when the signal already says the market’s worth being in.

The backtest doesn’t just support that top five, though — it supports a wider tier than most people would guess, and it draws a real, visible line at exactly where that tier ends.

Ranks 1 through 10 — the top decile — averaged 37% to 41% over the following 12 months in the backtest. Ranks 11 through 20 look nothing like that: roughly 18% to 20%, statistically indistinguishable from ranks 30, 50, or 70. The cliff isn’t gradual. It happens once, right at the edge of the top ten, and then the line goes flat for the rest of the list.

Why concentration matters this much

This isn’t just a Patala finding. It’s one of the most replicated results in long-run market research.

Hendrik Bessembinder’s landmark study of nearly every U.S. stock since 1926 found that just 4% of stocks accounted for all the net wealth the stock market has ever created above cash — everything else, in aggregate, merely matched T-bills. His most recent update, covering a full century of data, found the concentration has only gotten tighter: 89 firms once accounted for half of the market’s $43 trillion in net wealth creation; the latest data puts that number at just 46 firms, even as total wealth creation more than doubled to $91 trillion. Widen the lens to global markets and the pattern holds — one large-scale study puts the top-performing 2.4% of firms worldwide behind effectively all $75.7 trillion in net global wealth creation from 1990 to 2020.

The lesson isn’t “buy 4% of the market and ignore the rest” — nobody can identify those names in advance with certainty, which is exactly why broad indexing works as a baseline. But it’s a real, data-backed reason why a systematic tilt toward a tiny, well-defined top tier — not the top half, not “good enough,” but the actual top of the list — is where a stock-picking edge would have to live if it exists at all. QuantRanks’ own backtest shows the same shape at a smaller scale: a real cliff at the very top, then a long flat plateau where extra selectivity stops paying off. HD, sitting in the bottom third of a 450-stock list, isn’t a longshot in that framework. It’s statistically indistinguishable from cash.

Verdict

Not a buy — based on the rank, not the earnings print.

If HD’s rank ever climbs into the top 10, it’s worth a second look. Into the top five, it’s a QuantRank pick, same as any other stock. At over 300, it’s not close to either — beat or miss this morning, that doesn’t change.

Disclaimer: Educational and informational only. Not personalized investment advice, and not a recommendation to buy or sell any security. I am not a registered investment adviser. Past performance — backtested or live — does not guarantee future results. You can lose money, including a substantial amount. Consult a qualified financial adviser before investing.

Signal: BUY.

Capital stays invested — core and pick both.

This week’s QuantRank

LITE — Lumentum Holdings, Inc.

Current QuantRank: #2

Buy Monday, August 17.

Why this pick, briefly

Lumentum makes photonics — components and subsystems built around light instead of electricity. Its two biggest businesses: optical components for networks, and lasers.

On the networking side, Lumentum supplies the optical transceivers and components that move data between servers inside data centers and across telecom networks — the physical layer underneath the AI buildout everyone’s watching. On the laser side, it makes diode lasers used in industrial cutting and welding, telecom, and 3D sensing (the kind of laser tech behind facial-recognition hardware).

A few numbers, as of Friday’s close:

• Closing price (Aug 14): $926.14

• Market cap: ~$82.056 billion

• P/E (TTM): N/A — no meaningful trailing earnings to measure against.

What Lumentum does:

• Products: Optical components and transceivers for networking; diode lasers for industrial and sensing use.

• Markets: Data center and telecom networking, industrial manufacturing, 3D sensing.

• Strategy: Riding the data-center optical buildout — faster networks need more, and more advanced, optical components to move traffic between servers.

This one’s a step less familiar than last week’s pick — Lumentum doesn’t have the household-name recognition Micron does. That’s fine. The rank isn’t asking you to already know the story.

How much to buy

Typical range: 1% to 3% of your capital.

The S&P 500 is at all-time highs right now. If that makes you nervous, size toward the lower end of that range. That’s a reasonable adjustment. Skipping the pick entirely is not.

Things to remember every week

1. This is educational, not advice. I built this system, and I’m sharing what I personally follow — with my own money. You don’t have to follow it too.

2. This is one QuantRank, published once a week. Nothing more to track between issues.

3. It’s a quant system. Trust the rank, not your gut. Some QuantRanks — like Micron last week — are companies you already know by name. Others, like this week’s, aren’t. That’s not a downgrade in conviction. The rank did the judging either way.

4. Strong predictive power over the long run — no guarantee on any single name. QuantRanks has shown real predictive power over time. That doesn’t stop any one stock, or a string of them, from losing money — in the short run or the long run.

5. A repeat within ranks 1 to 5 is a good sign, not a stale one. QuantRanks tend to bounce between the top five spots from month to month — if a stock keeps reappearing in that tier, that’s the model telling you conviction is building, not fading. Keep adding on the normal schedule.

6. No target price, no separate exit. A QuantRank is held until the signal — the market-wide one, not a per-stock rule — tells the whole portfolio to sell.

7. This newsletter is the QuantRank. The method is one link away. Full mechanics, sizing ladder, and honest risk disclosure are in The Patala Core-Plus Strategy.

Why QuantRanks predict return

QuantRank #2 — this week’s tier — averaged 55.5% to 63.2% over the following 12 months in the backtest, across 210 trades. Not quite Rank #1’s number, but well clear of the middle of the pack — this is still top-tier territory, not a step down in what the model expects.

Full backtest methodology and the broader QuantRanks study are in the deep dive.

If the signal flips

Signal turns SELL: Move 80% to 100% of the portfolio to cash or short-term notes — SHV, SGOV, or similar. Core and pick go together. No exceptions for a position you like.

Signal turns BUY again: Restart. Rebuild your core position in SPY or QQQ first. Then resume adding 1% to 3% into the top QuantRank each week, the same as always.

Disclaimer: Educational and informational only. Not personalized investment advice, and not a recommendation to buy or sell any security. I am not a registered investment adviser. Past performance — backtested or live — does not guarantee future results. You can lose money, including a substantial amount. Consult a qualified financial adviser before investing. Full disclosures in The Patala Core-Plus Strategy.

• Inflation cooled, stocks rallied. July CPI eased to 3.4% annually, cooling fears of aggressive Fed hikes.

• SPY hit a new all-time high. Record close of $777.88 on Wednesday, easing to $776.34 by Friday — a +0.4% week.

• AI financing, GM’s supply deal, and swinging oil prices also drove the tape.

• The budget deficit widened to its highest since early 2021.

Watch item: the AI leaders haven’t confirmed the SPY high. SPY is at a record, but the Nasdaq (QQQ) is still ~2% below its June 2 peak, and semiconductors (SMH) — the leading indicator for the AI trade — are ~12% below their June 22 peak, even after this week’s modest bounce. Since AI capex has been the engine of this bull market, a broad index high that isn’t confirmed by its own leadership group is worth watching, not ignoring. The signal doesn’t act on this — it reads price, not narrative — but it’s the kind of divergence that’s often present before a trend loses steam.

Where the signal stands

The Patala signal remained a BUY all week — no flip, no drama, just the system continuing to ride a trend now well past the four-month mark.

Look at the last 8 weeks: the panic dip in late July, the snapback into month-end, and this week’s push to a fresh record before easing back into the close.

The signal has been green the entire time — a continuous BUY since April 2, now 134 days running.

Why the record high doesn’t change anything

Last week’s letter made the point that the system doesn’t chase good news to get more aggressive. This week proves it again: a soft CPI print, a fresh record close, an AI-financing headline cycle — none of it moves the model.

Guidance for the week ahead

The signal is a BUY, so the system says stay invested — same message as it’s been since April.

Next scheduled signal update: Saturday, August 22, 6:00 AM ET, right on the usual weekly clock. Between now and then, the plan doesn’t change based on inflation prints, AI-financing headlines, or a new record close. That’s the discipline.

The bigger picture: this is what “staying invested” looks like over time

Here’s the signal on SPY over the last 10 years — green means invested, red means in cash.

Find 2020. Find the 2022 bear market. The red bands cluster exactly where the real damage happened — and notice how much green there is everywhere else, including the push to a fresh record on the right edge of the chart this week.

The last 10 flips

Full transparency, as always — every confirmed flip, with the SPY move captured or avoided during each:

A few honest observations:

• The current BUY has held since April 2 — 134 days now — and is up +18.38%. That number keeps growing, and it’s the clearest illustration of why the long holds are where the system makes its money.

• This week isn’t in the flip table because the signal didn’t flip — same as the last several weeks. The best weeks for a timing system are often the quiet ones, whether the market is drifting or setting new records.

Full update next weekend.

A quick favor. I’m new to Substack, and this newsletter grows entirely through word of mouth right now. If this issue was useful, please subscribe and share it with a friend who’d benefit from my 20+ years of original investment research. It genuinely helps — thank you for reading.

Disclaimer: This is not investment advice or a recommendation, and I am not a registered investment adviser. This publication is for informational and educational purposes only. The signal reflects a backtested and now live-tracked quantitative model; past performance, backtested or live, does not guarantee future results. Markets involve risk, including loss of capital. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

Albrecht Altdorfer, "The Battle of Alexander at Issus" (1529). Alte Pinakothek, Munich.

No pick today. This is the human story behind every rule in the Core-Plus system — the market-timing signal, the position caps, the no-discretion mandate. None of those rules are arbitrary. Every one of them exists because of stories exactly like this one.

In the autumn of 325 BC, Alexander the Great stood at the edge of the Gedrosian Desert, having just conquered more of the known world than anyone in history — and he was still only in his late twenties. Ten years of unbroken victory lay behind him — Persia fallen, Egypt his, an empire stretching from Greece to the Indus. His own men had begged him to turn back at the Hyphasis River, exhausted and homesick, and for once he’d relented. But rather than take the safer inland road home, Alexander chose to march his army straight through one of the most lethal deserts on earth — partly to resupply his fleet, partly, historians agree, to outdo the legendary conquerors before him who had tried and failed to cross it. He wanted to be the first.

He was almost the last. The sixty-day crossing became a catastrophe: no reliable water, flash floods that drowned camp followers in their sleep, soldiers and pack animals collapsing from heat and thirst by the thousands. Estimates of the dead range as high as three-quarters of the column that entered the desert. A young commander who had never lost a battle nearly destroyed his own army not through enemy action, but through the same audacity that had built his empire in the first place.

Sam Bankman-Fried and Leopold Aschenbrenner each lived a twenty-first-century version of that pattern: two extraordinarily young minds who moved with real speed and real insight, whose youth was their edge, and who ran, at the peak of their success, straight into their own Gedrosian Desert. To be direct about it upfront: their outcomes were not the same, and this piece doesn't argue they were — one man was convicted of fraud, the other has not been accused of any wrongdoing. What they share is the psychological pattern that preceded the collapse, not the collapse itself.

The investor’s version

Every investor knows a smaller version of this story from the inside.

You find an idea you believe in. You test it with a little money, and it works — fast. That first win turns a hypothesis into a conviction, so you add more, then more again, each addition justified by the last one paying off. Somewhere in there, a bet starts feeling like a strategy.

Then leverage enters, because conviction that strong looks foolish left un-levered. Your account grows in a way that stops resembling investing and starts resembling vindication. A major share of your net worth ends up riding on one idea, and the idea has been right for so long that risk starts to feel like a formality.

Then, for reasons that always feel unfair in the moment, the thesis stalls. Because you’re levered, the drawdown isn’t proportional to how wrong you were — it’s amplified by how much borrowed conviction you were carrying. Margin calls arrive on their own schedule, not yours. The music stops, and the portfolio you thought reflected how right you’d been actually just reflected how much you’d bet on staying right forever.

This isn’t a story about bad investors. It happens to people whose early read was genuinely good — which is exactly what makes the lesson easy to miss. It also happens disproportionately to young investors, because youth comes with a shorter memory of past cycles and a longer runway of time to feel invincible in. Do it with a personal account and you get a bad year. Do it with billions of other people’s dollars, at twenty-nine or twenty-four, and you get Sam Bankman-Fried and Leopold Aschenbrenner.

The boy who did the math

Bankman-Fried’s story was always a little too clean: a Jane Street trader turned effective altruist, running expected-value calculations on human suffering, concluding that the moral thing to do was get rich fast and give it away faster. He was still in his twenties when he built it. If you believe the odds favor you and the upside is measured in lives saved, caution starts to look like a moral failure. FTX went from scrappy exchange to a firm valued in the tens of billions in barely two years, with Bankman-Fried becoming, on paper, one of the youngest self-made billionaires alive — the same audacity that let a twenty-something outmaneuver Wall Street veterans also convinced him the normal guardrails didn’t apply to him.

That speed was the tell.

When a twenty-something’s models keep printing money, the conclusion isn’t “I got lucky” — it’s “I understand something others don’t.”

Alameda Research reportedly operated with a secret, effectively unlimited line of credit at FTX — reporting has put the exempted borrowing limit as high as $65 billion — and by September 2022 had drawn roughly $14 billion of customer funds against it, largely uncollateralized. That isn’t leverage in the conventional sense of margin against posted collateral; it’s closer to trading with someone else’s money with no real ceiling at all, which made the eventual hole impossible to paper over once confidence in the FTT token collapsed. A jury saw fraud, not philosophy.

What’s striking is that he seems to have believed his own model of himself — smart enough, righteous enough, young enough to see further than everyone older and more cautious around him.

The essay that became a fund

Leopold Aschenbrenner’s path looks nothing like SBF’s, which is exactly what makes it worth placing alongside it. A German-born prodigy who entered Columbia at 15 and graduated valedictorian at 19, he worked on OpenAI’s Superalignment team before being pushed out in 2024. He turned the experience into “Situational Awareness,” an essay arguing AI matching human intelligence would arrive by decade’s end, and that markets were badly underpricing the implications — written and published before he’d turned 24.

The essay made him famous. Then, still in his mid-twenties, he raised a hedge fund built entirely around trading that thesis — and for a while it worked spectacularly, swelling to tens of billions of dollars as AI infrastructure stocks soared, with Aschenbrenner treated less like a fund manager than an oracle with a spreadsheet. It’s a very young person’s move: skip the decades of apprenticeship a fund manager traditionally serves and bet that raw conviction, arrived at fast, can substitute for experience.

Then, in late July, a sharp selloff in AI infrastructure names triggered margin calls his fund couldn’t easily meet. Reports put some of Situational Awareness’s positions at roughly 4x leverage — meaning a given stock’s move was felt four times over in the fund’s returns.¹ That leverage had been the engine of the fund’s rise, reportedly gaining more than 400% after fees through June; it became the engine of its collapse, as the portfolio fell an estimated 67% in July alone, with total assets sliding from a peak near $45 billion to around $10 billion.² Situational Awareness was forced to unwind its entire public book in days, with Citadel taking the trades off his hands.³ In a letter to investors, Aschenbrenner took responsibility for the losses and said he was committed to learning from them. Within days he’d wired $400 million into a private, Sequoia-backed company — a fast pivot that reads as either resilience or a young founder unable to sit still with having been wrong.

What audacity buys, and what it costs

Neither man set out to be a cautionary tale, and neither did Alexander. All three were genuinely exceptional — young people who saw further than their elders and had the audacity to act on it at a scale most never attempt. That audacity built an empire, an exchange, and a fund. It isn’t incidental to their success; it is their success, right up until it isn’t.

Bankman-Fried’s collapsed in fraud and a courtroom. Aschenbrenner’s collapsed, so far, in a margin call and a public letter of contrition — a softer landing, and maybe a genuinely instructive one, for a fund manager still only in his mid-twenties. Alexander’s collapsed in a desert of his own choosing, from which he emerged with his legend intact but his army broken. The audacity of youth is a genuine asset — it’s what lets someone move before consensus catches up, place a bet nobody older would place, see a market or a battlefield with fresh eyes. But it’s also, unchecked, a failure of proportion: forgetting you’re standing on the same uncertain ground as everyone else, no matter how young, fast, or far ahead you’ve marched.

Lessons for the rest of us

Your first win is data, not destiny. One profitable trade tells you the market agreed with you once — not that it will keep agreeing, or that you’ve cracked a code others missed.

Size positions to conviction earned, not conviction felt. Certainty grows fastest exactly when you should be most careful — after a string of wins. Scale in on rules set before the trade started paying off, not in the euphoria of watching it work.

Leverage amplifies your errors in timing, not just your returns. Being early and being wrong look identical in a levered account — until the margin call, which doesn’t care which one it was. Every unit of leverage bets both that you’re right and that you can survive being temporarily wrong. Size for the second bet too.

A thesis without a pre-committed exit is a hope, not a strategy. Decide in advance, in writing, what would prove you wrong and what you’ll do about it. “The market will realize I’m right” isn’t a plan.

Concentration is the price of conviction — stay the one setting it. Betting big on a strong idea is fine. The danger is a position so large and levered that a margin desk, lender, or court ends up choosing your exit, on their timeline.

Keep someone around who can say no. Alexander’s men begged him to turn back at the Hyphasis, and for once he listened. He didn’t listen again at Gedrosia. Both modern stories feature founders surrounded by people who owed their positions to the founder’s success — few voices able to say “this is too much” and be heard. Protect that voice, especially while you’re young and winning, since that’s when you’ll least want to hear it.

Every one of those six lessons is also, not coincidentally, a rule already built into the investing system I run and publish every week.

The market-timing signal is the pre-committed exit. The position-size caps on each QuantRank are conviction-earned sizing, not conviction-felt sizing. The fact that the model ranks and I execute — no override, no “but I have a good feeling about this one” — is the outside voice that says no, encoded directly into the process instead of left to whoever’s willing to say it out loud. I didn’t build Core-Plus because I’m smarter than Aschenbrenner or more disciplined than Bankman-Fried was at their age. I built it because I know I’m exactly as capable of this pattern as they were, and a system doesn’t get tired of being the adult in the room.

Have you known someone like this?

You’ve probably watched a smaller version of this — someone (maybe you) who got right often enough that being right stopped feeling like luck and started feeling like identity. What was their Gedrosian Desert? Did they turn back — or not? Tell me who it was.

Being early, sharp, and audacious are rare gifts. None of them protect you from the one thing that ends every version of this story the same way, from ancient deserts to modern margin desks: forgetting the world was never obligated to keep agreeing with you.

Notes

1. Reported leverage of roughly 4x (up to 400%) on Situational Awareness’s public equity positions: CNBC, “Leopold Aschenbrenner’s Situational Awareness fund: $45B to fire sale,” July 31, 2026; CNBC, “Why Situational Awareness hedge fund imploded, even in a tame stock market,” July 31, 2026.

2. The ~67% July loss and the decline from a ~$45B peak to ~$10B: CNBC, July 31, 2026; Yahoo Finance / CNBC, “Leopold Aschenbrenner’s AI hedge fund collapses after margin calls”.

3. The forced sale of the fund’s public equity book to Citadel: CNBC, “AI investor Leopold Aschenbrenner forced to unwind all public stock positions after steep losses, sources say,” July 30, 2026; Bloomberg, “Aschenbrenner Hedge Fund Situational Awareness Unwinding Trades, Report Says,” July 30, 2026.

Background on Sam Bankman-Fried’s conviction and the Alameda/FTX credit line is drawn from widely reported U.S. federal trial coverage and public SEC/DOJ filings from 2023.

No live pick here. Just the evidence, laid out plainly, with the caveats that belong next to it.

The question

QuantRanks scores a universe of about 450 stocks — mostly S&P 500 constituents, plus some names outside the index — and re-ranks the entire list every month. Rank 1 is the model’s top pick that month. Rank 450 is the bottom.

Run monthly from January 2008 through July 2025 — the last starting month with a full 12 months of forward data to test against, given the price history runs through July 2026 — that’s:

450 stocks × 211 months = nearly 95,000 stock-months scored.

(This window runs through July 2025, a year shorter than the Core-Plus strategy article’s backtest, which goes through July 2026. That’s not a discrepancy — the Core-Plus numbers are a running equity curve that can use every day of price history available. This backtest needs each entry to have a full 12 months of future price data to grade it, so the last usable starting month is 12 months before the data runs out.)

The question this article answers: across all of that, does a lower rank number actually predict a better return? Not just for the famous winners — for the whole distribution.

How the test works

Every ranked stock gets tracked five ways: a market-timing exit (more on that below), and fixed 1-month, 3-month, 6-month, and 12-month holds. This piece focuses on the 12-month numbers, since that’s the cleanest read on whether the rank itself has staying power.

Two versions of every number appear below — both are 12-month holds, full stop. The only difference is which trades are counted:

• Buy & Hold — every trade, bought regardless of what the market-timing signal was doing that day.

• SPY Market Signal — the same fixed 12-month hold, but counted only for trades that started on a day the market-timing signal (the one that governs the Core-Plus strategy) read BUY. Trades that started when the signal was already unfavorable are excluded from this column.

Ranks 1 through 10 — the names you’d actually buy

This is the tier the weekly pick comes from. Individual performance, 12-month average return:

The same tier shows up at every holding period, not just 12 months. Here’s Buy & Hold average return by rank, all four horizons:

The gap between ranks 1–4 and rank 5-and-below is visible at 3 months, widens by 6 months, and is largest at 12 months. The reason this article leans on the 12-month column specifically: it’s the cleanest read on whether a rank’s edge has real staying power, rather than a short-term blip, and it’s the horizon closest to how a pick is actually held in practice — until the market-timing signal says sell, which averages considerably longer than a month. The shorter columns are here to show the pattern isn’t an artifact of picking one convenient time frame; it holds up at every horizon tested.

Here’s the same comparison for Rank 1 specifically, both entry conditions, across all four horizons:

The gap between the two bars — Bought Regardless of SPY Signal vs. Bought Only When SPY Signal = BUY — widens as the holding period stretches out. At 1 month it’s barely a point. At 12 months it’s roughly 9 points. Both bars in every pair hold for the same fixed period; the only thing that differs is whether the trade started in a market the timing signal already liked.

Ranks 6 through 10 are noisier — 9 outranks 8, for instance. At this granularity, with roughly 130–210 trades per rank, that kind of local noise is expected. It’s not a sign the model reverses itself; it’s a sign that rank-by-rank precision past the top few names should be read loosely.

Why the weekly pick isn’t always Rank #1

One practical nuance worth explaining here, since it directly shapes what shows up in your inbox each week.

A stock’s QuantRank doesn’t change wildly week to week. The model re-ranks monthly, and conviction tends to concentrate on the same handful of names for weeks or even months at a stretch — that’s the “repeat name at #1” pattern regular readers have already seen.

That’s good news for the ranking’s stability. It’s a problem for a newsletter that mechanically buys “whatever is Rank #1 this week.” If the same stock holds the top spot for eight weeks running, “buy Rank #1 every week” really means “buy the same one or two stocks, over and over.” That’s not a diversified top-tier portfolio — it’s a two-stock bet wearing a diversified newsletter’s clothing.

The fix: each week, the pick moves sequentially through that month’s top tier. Week 1 of the month is Rank 1, week 2 is Rank 2, week 3 is Rank 3, week 4 is Rank 4 — and in months with a fifth Monday, week 5 is Rank 5. The cycle resets at the start of the next month.

This isn’t a compromise on quality. Look back at the rank table above: ranks 1 through 4 are a tier of their own — 62% down to 46% average 12-month return, clearly separated from rank 5 and below. Rotating through them means every single pick still comes from that top tier. What changes is that a subscriber following the newsletter for a full month ends up holding four or five distinct names instead of one stock bought four or five times over — real diversification within the tier the backtest actually supports, instead of concentration by accident.

Every rank, grouped into deciles

Zooming out to the full range — all ~90 ranks the model produces, grouped into tens — the pattern holds up at scale, not just at the top:

*Small sample past decile 80 — treat directionally.

This is the shape that matters most: a cliff from decile 1–10 to everything after, then a long, flat plateau. The top decile roughly doubles the return of every decile below it, and deciles 11 through 80 all sit in a fairly tight band around 14–20%. The model isn’t gently sorting 450 stocks into a smooth gradient — it’s identifying a top tier that behaves differently, and beyond that tier, further precision adds little.

That’s also, practically, the case for keeping the weekly pick inside ranks 1 through 5 rather than reaching further down the list: the real signal is concentrated at the very top.

What the Buy & Hold vs. Signal gap tells you

Look at the two columns side by side, at both the rank and decile level: SPY Market Signal is consistently a little higher than Buy & Hold, across almost every tier. For Rank 1, that gap is about 9 points (61.6% vs. 70.8%). At the decile level, decile 1–10 shows the same pattern (37.3% vs. 41.4%).

Read that carefully — it isn’t the market-timing overlay adding return through good trading. It’s a selection effect: trades that entered when the broader market signal was already healthy tended to do better, on the same fixed 12-month clock, than trades entered into an already-weak market. Buy & Hold includes both; SPY Market Signal filters toward the healthier entries. The gap is a rough measure of what market regime costs or buys you, on top of picking the same stock.

The honest limits of this backtest

This is 2008–2025, one 17.6-year window. It contains one historic crash near the start and an unusually strong run for U.S. large-cap and growth names afterward. A different 17.6 years — a longer stretch of flat or sideways markets, for instance — would very plausibly produce a smaller gap between top and bottom ranks, or a lower absolute return across the board.

Averages hide dispersion. A rank averaging 60% doesn’t mean every trade in that rank returned close to 60%. Some trades in every rank, including rank 1, lost money. The average is real, but it isn’t a promise about any single position — including next week’s.

This is frictionless. No commissions, no bid-ask spread, no slippage, no taxes, and no allowance for a stock being un-buyable at the exact price used in the backtest. Real execution will lag these numbers by some amount.

Sample size thins at the edges. Ranks and deciles built on 200+ trades (roughly ranks 1–70) are on reasonably solid footing. Past decile 80, you’re looking at a few dozen observations or fewer — real, but noisier than the headline numbers suggest.

Past performance does not guarantee future results. This is true of every number in this article, without exception. A backtest describes a period that already happened, built with full knowledge of how it turned out. It is evidence, not a promise. The live portfolio — reported monthly, with real trades and real prices — exists precisely because a backtest alone isn’t proof.

Why this matters for the weekly pick

Every Sunday’s QuantRank comes from ranks 1 through 5 — the tier that, in this backtest, decisively outperformed everything below it, with a clear gap rather than a gentle slope. Rotating through that tier week by week, instead of parking on Rank 1 alone, is how the newsletter stays both evidence-based and genuinely diversified. That’s the evidence behind the newsletter. It doesn’t mean every pick works out; some won’t. It means the selection process, tested honestly across nearly 95,000 stock-months, has done real work historically — not just on a handful of lucky names, but across the full distribution.

Disclaimer: Educational and informational only. Not personalized investment advice, and not a recommendation to buy or sell any security. I am not a registered investment adviser, broker-dealer, or financial planner. All performance figures in this article are backtested and hypothetical — derived from the retrospective application of a model to historical data. They do not represent actual trading and do not reflect trading costs, commissions, slippage, or taxes. Hypothetical results have inherent limitations, including the benefit of hindsight, and results may vary with each use and over time. Past performance — whether backtested or live — does not guarantee or indicate future results, and future results may be materially worse. Investing in equities involves substantial risk, including the possible loss of some or all of your capital. Concentrated positions in individual stocks carry materially greater risk than a diversified index fund. You should not rely on this publication as the basis for any investment decision. Consult a qualified, licensed financial adviser who understands your full circumstances before investing.

Capital stays invested — core and pick both.

This week’s QuantRank

MU — Micron Technology, Inc.

QuantRank: #1

About Micron, briefly

Micron makes memory chips. DRAM and NAND flash, mostly — the memory that sits inside everything from phones to data centers.

Right now, AI hardware is the demand driver. Micron makes high-bandwidth memory (HBM), the memory type AI chips lean on hardest.

A few numbers, as of Friday’s close:

• Closing price (Aug 7): $877.57

• Market cap: ~$991 billion

• P/E (TTM): 19.87

What Micron does:

• Products: High-bandwidth memory and DRAM, its two biggest focus areas.

• Markets: Enterprise data centers, AI infrastructure, mobile, automotive.

• Strategy: Leaning into high-margin enterprise AI. Stepping back from lower-margin consumer retail.

You already know this name. It’s a large, liquid, S&P 500 stock — not an obscure pick. That’s normal for QuantRanks. Most weeks look like this one.

How much to buy

Typical range: 1% to 3% of your capital.

The S&P 500 is at all-time highs right now. If that makes you nervous, size toward the lower end of that range. That’s a reasonable adjustment.

Things to remember every week

1. This is educational, not advice. I’m sharing what I’m buying, with my own money. You don’t have to buy it too.

2. This is one QuantRank, published once a week. Nothing more to track between issues unless the market SPY signal flips mid-week.

3. It’s a quant system. Trust the rank, not your gut. Most QuantRanks come from the S&P 500 — companies you already know by name. That familiarity is exactly what tempts people to second-guess the model. Don’t. The rank did the judging already.

4. Strong predictive power over the long run — no guarantee on any single name. QuantRanks has shown real predictive power over time. That doesn’t stop any one stock, or a string of them, from losing money — in the short run or the long run.

5. A repeat name at #1 is a good sign, not a stale one. If a stock holds the top QuantRank for months in a row, that’s the model telling you conviction is building, not fading. Keep adding on the normal schedule.

6. No target price, no separate exit. A QuantRank is held until the signal — the market-wide one, not a per-stock rule — tells the whole portfolio to sell.

7. This newsletter is the QuantRank. The method is one link away. Full mechanics, sizing ladder, and honest risk disclosure are in The Patala Core-Plus Strategy.

How QuantRanks predict return

This isn’t a hunch. The model scores a universe of about 450 stocks — mostly S&P 500 constituents, plus some names outside the index — and re-ranks the whole list every month. Run monthly from January 2008 through July 2025 (the last starting month with a full 12 months of forward data to test against), that's 450 stocks × 211 months: nearly 95,000 stock-months scored.

Below is the actual backtest: average 12-month return by QuantRank, shown both as a plain buy-and-hold, and with the SPY market-timing signal applied. QuantRank #1 — this week’s tier — averaged 62% vs 71%, more than double the return of names ranked in the middle of the pack.

If the signal flips

Signal turns SELL: Move 80% to 100% of the portfolio to cash or short-term notes — SHV, SGOV, or similar. Core and pick go together. No exceptions for a position you like.

Signal turns BUY again: Restart. Rebuild your core position in SPY or QQQ first. Then resume adding 1% to 3% into the top QuantRank each week, the same as always.

Disclaimer: Educational and informational only. Not personalized investment advice, and not a recommendation to buy or sell any security. I am not a registered investment adviser. Past performance — backtested or live — does not guarantee future results. You can lose money, including a substantial amount. Consult a qualified financial adviser before investing. Full disclosures in The Patala Core-Plus Strategy.

Market digest: what moved this week

It was a clean, strong week — a nice change of pace after the whipsaw action of the prior few.

• Stocks rallied hard. The S&P 500 gained 3.6%, the Nasdaq jumped 5.2%, and the Dow added 3% — the S&P’s best week since April, closing Friday at a fresh record high.

• The catalyst was a surprise jobs report. July payrolls showed a loss of 23,000 jobs, well below expectations. Weak as that sounds on its face, markets read it as cooling the odds of an aggressive Fed rate hike — Treasury yields fell on the news.

• Tech carried the tape. Major tech earnings and continued strength in AI-adjacent names added fuel on top of the rate relief.

• Oil stayed choppy, with crude sliding as markets weighed potential Middle East developments alongside supply adjustments.

SPY itself moved from $747.03 at last Friday’s close to $773.26 this Friday — a +3.5% week, right in line with the broader tape, and a new high along the way.

Where the signal stands

The Patala signal remained a BUY all week. No drama, no whipsaw — just a system riding a trend that’s now been intact for over four months.

Look at the last 8 weeks: the panic dip in late July, the snapback into month-end, and then this week’s clean push to new highs. The signal has been green the entire time — a continuous BUY since April 2, now 127 days running.

Here’s the full year for context — the February SELL that sidestepped the March drawdown, the April re-entry near the lows, and the long hold since, now sitting at a new high:

Why the “boring” weeks matter too

Last week’s letter was about the system holding through panic. This week is the other half of the story: the system doesn’t need good news to justify staying in, and it doesn’t chase good news to get more aggressive either. It just keeps reading price behavior.

A soft jobs report, a Fed-relief rally, a record close — none of that changes what the model does. It was a BUY before the jobs report and it’s a BUY after it. The signal isn’t reacting to headlines in either direction, which is exactly the point: out of 158 daily zones, only 11 trigger a SELL, and nothing about this week’s action came close to tripping that wire.

Guidance for the week ahead

The signal is a BUY, so the system says stay invested — same message as it’s been since April.

Next scheduled signal update: Saturday, August 15, 6:00 PM ET, right on the usual weekly clock. Between now and then, the plan doesn’t change based on headlines, earnings surprises, or how good the tape feels after a week like this one. That’s the discipline.

The bigger picture: this is what “staying invested” looks like over time

Here’s the signal on SPY over the last 10 years — green means invested, red means in cash.

Find 2020. Find the 2022 bear market. The red bands cluster exactly where the real damage happened — and notice how much green there is everywhere else, including the push to new highs on the right edge of the chart this week. The system spends most of its life invested, riding the trend, stepping aside only for genuine danger.

The last 10 flips

Full transparency, as always — every confirmed flip, with the SPY move captured or avoided during each:

A few honest observations:

• The current BUY has held since April 2 — 127 days now — and is up +17.91%. That number keeps growing, and it’s the clearest illustration of why the long holds are where the system makes its money.

• The whipsaws are in there too — the 1-day SELL in Feb 2025 and the 0-day flip a week earlier are right there in the table. We publish every confirmed flip, wins and losses both.

• This week isn’t in the table because the signal didn’t flip — same as last week. The best weeks for a timing system are often the ones where it does nothing at all, whether the market is panicking or celebrating.

Full update next weekend.

A quick favor. I’m new to Substack, and this newsletter grows entirely through word of mouth right now. If this issue was useful, please subscribe and share it with a friend who’d benefit from my 20+ years of original investment research. It genuinely helps — thank you for reading.

Disclaimer: This is not investment advice or a recommendation, and I am not a registered investment adviser. This publication is for informational and educational purposes only. The signal reflects a backtested and now live-tracked quantitative model; past performance, backtested or live, does not guarantee future results. Markets involve risk, including loss of capital. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

Let me build it up the way I built it in testing — one piece at a time — so you can see exactly where each layer earns its keep.

Start with one asset

The simplest version of this strategy needs just one ETF and the signal.

When the signal says BUY, you hold the ETF. When it says SELL, you move to cash. That’s the whole thing. No stock-picking, no watching the market intraday — you check one signal a week and act only when it changes, which happens about eleven times a year.

Here’s what that did from January 2008 through July 2026 — 18 and a half years:

Start with the top two rows. Buy-and-hold SPY compounded at 11.2% a year — and lost 48% of its value in the 2008–09 crash. QQQ did better on return, 15.9% a year, and still gave back 46%. Those numbers understate the damage, because a drawdown isn’t just a number — it’s a length of time. SPY didn’t reclaim its old high until April 2011. That’s more than three years underwater, and it is the reason so many investors never actually collect the market’s long-run return. They sell somewhere near the bottom and come back long after the recovery.

Now add the signal. Timed SPY compounded at 23.9% a year with a worst drawdown of −8.5%. Timed QQQ reached 30.7% with a −9.3% worst drawdown. Same two ETFs, same holdings, no leverage. The only difference is sitting in cash through the worst stretches instead of riding them down.

The clearest illustration is the COVID crash. Between January and March 2020, SPY fell 19.4%. The timed version was down 0.04%.

That alone is a complete, followable strategy. If you did nothing else but run the signal on one ETF, you’d have a system most people would be glad to own. But we can do better.

Add the core: choose your engine

The single-asset version raises the obvious question: SPY or QQQ?

This is your first dial, and it’s really a question about temperament:

• SPY — 500 companies, broadly diversified, steadier. The conservative core.

• QQQ — 100 growth-heavy names, more volatile, higher-returning over this era. The aggressive core.

Because the signal governs either one, you can reach for QQQ’s higher returns without sitting through QQQ’s unprotected drawdowns. Unprotected, the two ETFs bottomed out within two points of each other in 2008 — down 46% and 48%. Timed, they came through under 10%. The signal is what makes the aggressive choice tolerable.

Whichever you pick becomes your core — the base your capital rides while the signal says BUY.

Add the Quant picks: one high-conviction stock a week

Here’s where the “Plus” comes in.

My model doesn’t only time the market — it also selects individual stocks. Rather than leave everything in the ETF core, I carve off a small slice for the highest-conviction name the model surfaces — one new pick a week, published every Sunday at 6:00 pm ET and bought the following Monday.

Each pick is funded by selling a small piece of the core. It’s held until the signal flips to SELL, at which point everything — core and picks alike — goes to cash together. There’s no separate exit rule for individual stocks, and no target price. The signal governs the whole portfolio.

Two limits keep this from running away with the portfolio. Each pick is capped at a fixed percentage of the portfolio, which you choose. And the stock sleeve as a whole is capped at 50% — once open picks reach half the portfolio, new picks are skipped until the next SELL clears the decks. At least half your money stays in the core at all times.

That’s the entire system: Signal, Core, Picks. When to invest, what to hold, what to buy.

What this is — and what it isn’t

This is the part I want to be blunt about, because it determines whether this newsletter is for you.

Everything here is quant-driven. The signal is a model output. So is the stock selection. I don’t override either one because I have a feeling about the tape, and I don’t skip a pick because I don’t personally like the company.

I don’t do fundamental or technical deep dives. You will not get a 3,000-word thesis on why some company is a generational buy. No discounted cash flow models, no earnings-call teardowns, no assessments of management quality, no chart patterns or support levels. That’s not what this is, and pretending otherwise would be dishonest.

What you get instead is short and mechanical:

• the current signal state — BUY or SELL,

• the week’s stock pick,

• and the sizing rule.

Perhaps five minutes of reading and one or two trades a week.

Why build it this way? Because consistency is the thing most retail investors never achieve. Read forty analyst reports and you can reach opposite conclusions on the same stock in the same week, depending on which one you read last and what mood the market put you in. A rule doesn’t do that. A rule produces the same answer every time, including on the days you’d rather not follow it — and those are usually the days that matter most. The 2008 and 2020 columns above weren’t won by insight. They were won by a rule that said “sell” and got followed.

The honest cost of this approach: when a pick goes against me, I won’t have a satisfying story for why. The model doesn’t offer one, and I won’t invent one after the fact. If you need a narrative to hold a position, a mechanical system will be genuinely uncomfortable to follow. That’s worth knowing before you subscribe rather than after.

How much to put in each pick

The last dial is position size — each pick as a percentage of your portfolio:

• 2% per pick — gentle. The core does most of the work.

• 3% per pick — balanced. My default.

• 4% per pick — assertive.

• 5% per pick — aggressive. The Quant Stocks carry more of the load.

The pattern is the one you’d expect: every step up in pick size adds return, and adds concentration risk along with it. Note what the 50% cap does, though. At 2% you take nearly every pick the model produces. At 5% you hit the ceiling quickly in a long BUY stretch, and roughly a quarter of the picks get skipped because there’s no room left — which is why the last step up buys less than the ones before it. You set the dial to your own comfort, not mine.

Notice one thing, though: the timed core with no picks at all is already a serious strategy. The picks are meant to enhance a strong base, not to rescue a weak one. That’s deliberate, and it’s why the system doesn’t live or die on any single stock.

What the drawdown numbers do and don’t cover

I want to be precise about this, because it’s the easiest place for a reader to draw a false conclusion.

The −8.5% and −9.3% worst drawdowns above are for the timed ETF core only — no stock picks. They are what the signal did to SPY and QQQ, nothing more.

The Core-Plus version will be considerably more volatile, and its drawdowns will be materially deeper. Here’s why. The core is an index — hundreds of companies, one bad name barely registers. A stock pick is a single company. Most picks will be S&P 500 constituents, but not all of them; some will be smaller, less liquid, higher-growth names. Individual stocks of that kind routinely fall 30%, 40% or more on a single earnings miss, a guidance cut, or a sector rotation — and unlike the index, they don’t necessarily come back. With up to half the portfolio in a handful of such positions at once, drawdowns well beyond what the core alone experienced are not a tail risk. They should be your baseline expectation.

I’m not publishing a drawdown figure for the Core-Plus variants, because computing one honestly requires the full daily price history of every stock the model has ever picked, and I’d rather show you nothing than show you a number I can’t stand behind. What I will do is report the live portfolio’s drawdown every month, as it happens, on real money.

If a 30–40% drop in a single position would cause you to abandon the system, run a smaller pick size — or run the core alone. That option is on the table for a reason.

Why “Core-Plus”

The name is the architecture: a Core you compound in (a timed ETF), plus a sleeve of high-conviction picks on top. The signal governs both — it decides when the whole portfolio is invested and when it’s in cash.

Three layers, each with one job:

• The Signal tells you when — and moves you aside in the worst stretches.

• The Core is what you hold — SPY or QQQ, your choice of engine.

• The Picks are what you buy — one high-conviction stock a week.

Every Sunday at 6:00 pm ET I publish the signal and the week’s quant-picked stock, free.

The schedule, and the live portfolio

Backtests are necessary, but they are not proof. A backtest is a description of a period that already happened, built with full knowledge of how it turned out. So from Monday 10 August 2026, I’m running The Patala Live Portfolio — a real, funded account, my own money, invested in this exact Core-Plus strategy, with results I’ll report every month.

The weekly rhythm is fixed:

• Sunday, 6:00 pm ET — I publish the current signal state and the week’s quant-picked high-conviction stock.

• Monday — I buy it in the live portfolio. If Monday is a market holiday, the trade moves to Tuesday.

Same time, every week. You’ll have the pick in hand the evening before I act on it, with the whole of Sunday night to decide what — if anything — you want to do with it.

You get the pick before I act on it, not after. That ordering is deliberate and it will not change.

I’ll report allocations and percentages, not dollar amounts. You’ll see the signal state, the core and how much is deployed, and each open pick with its position size and return — all as percentages of the portfolio. What you won’t see is my account balance, because it isn’t useful to you. Whether you’re running $5,000 or $5,000,000, what transfers is the allocation, not my personal balance.

Each monthly report will show:

• The current signal — BUY or SELL — and how long it’s been in that state

• The core allocation, the stock sleeve, and the cash position

• Every open pick: its position size and its return

• The month’s return, the cumulative return since inception, and the worst drawdown so far — in percent

What to expect

Let me set expectations plainly, because that matters more than enthusiasm:

• The numbers above are backtested and hypothetical. They’re a simulation over one historical window — January 2008 to July 2026 — with no trading costs, no slippage and no taxes. That window contains one enormous crash near the start and a long, unusually strong run for U.S. large-cap growth after it. A different 18 years would produce different results. Past performance does not guarantee future results. A live return well below the backtest would still be a reasonable outcome, and it’s the one I plan around.

• The edge depends on the signal continuing to work and on the stock selection holding up. I believe both will — that’s why my own money is in it — but belief isn’t evidence, and a model that worked across one historical window can stop working in the next one.

• You can lose money doing this, including a substantial amount. The signal reduced drawdowns in testing; it does not eliminate them, and it cannot protect against a fast crash between weekly readings. Individual stock picks can fall hard and can go to zero. Nothing here is capital-protected.

• Some months will be red. Some picks will be losers. The signal will whipsaw — flipping to SELL just before a rally, or to BUY just before a drop. The backtest shows plenty of those. That’s normal for a timing model, and I’ll report it plainly when it happens.

• This is education, not advice. I don’t know your income, your tax situation, your time horizon, or your other holdings — and without those, no one can tell you what belongs in your portfolio.

What I can offer is a disciplined system, run transparently, with my own capital on the line, and the honesty to show you exactly how it goes.

The signal tells you when. The core compounds. The picks add the edge. From Monday, you can watch it happen in real time.

Disclosure: I personally invest in the positions discussed in The Patala Live Portfolio, including every weekly pick. Picks are published to all subscribers on Sunday at 6:00 pm ET, and I place my own trades on the following Monday — that is, always after publication, never before.

Disclaimer: I am not a registered investment adviser, broker-dealer, or financial planner, and nothing in this publication is personalized investment advice, a recommendation or solicitation to buy or sell any security, or an offer of advisory services. This publication is provided for educational and informational purposes only. All performance figures shown are backtested and hypothetical; they are derived from the retrospective application of a model to historical data, do not represent actual trading, and do not reflect trading costs, commissions, slippage, or taxes. Hypothetical results have inherent limitations, including the benefit of hindsight, and results may vary with each use and over time. Past performance — whether backtested or live — does not guarantee or indicate future results, and future results may be materially worse. Investing in equities and ETFs involves substantial risk, including the possible loss of some or all of your capital; a market-timing model does not prevent losses and may underperform a simple buy-and-hold approach for extended periods. Concentrated positions in individual equities — including smaller, less liquid, or high-growth companies — carry materially greater risk of large loss than a diversified index fund. Asset allocation and diversification do not guarantee a profit or protect against a loss. The Patala Live Portfolio reflects one individual’s account and is shared for illustration only. You should not rely on this publication as the basis for any investment decision. Consult a qualified, licensed financial adviser who understands your full circumstances before investing. You are solely responsible for your own investment decisions and their outcomes.

What the heck just happened this week?

Let’s start with the honest recap, because this was one for the record books.

It was, by almost any measure, the roughest week since the launch of this newsletter — and a genuine stress test for any systematic approach:

• Semiconductors crashed, dragging down the AI complex that has led this entire bull market.

• The KOSPI sold off hard earlier in the week, sending a risk-off wave across global markets.

• Retail investors dumped single stocks at the fastest pace since COVID — the kind of capitulation that usually marks peak fear, not calm.

• The Fed meeting injected its usual dose of two-way volatility.

• And Leopold Aschenbrenner’s hedge fund blew up — a high-profile, high-flying fund unwinding in a very public way, exactly the sort of headline that makes everyone assume the dominoes are starting to fall.

Turn on any screen last week and the message was the same: the top is in.

And then — as it so often does right when the panic peaks — the market recovered strongly into the back half of the week. SPY dipped to $729 on Wednesday and closed Friday back at $747.03, essentially flat on the week after all that drama.

Where the signal stands

Through all of it, the Patala signal stayed a BUY. It never budged.

Look at that eight-week chart. The sharp dip on July 29 is the panic — the moment the headlines screamed loudest. The signal held green straight through it. It has been a continuous BUY since April 2 — 120 days now — and through this week’s turbulence, the model never wavered.

Here’s the full year for context — the February SELL that sidestepped the March drawdown, the April re-entry near the lows, and the long hold since:

Why this week mattered more than most

I want to be honest about something: this is the exact kind of week 20+ years of research and 26+ years of backtesting were built for.

Not the calm weeks. The scary ones.

Anyone can hold when markets are rising and the news is good. The entire value of a system reveals itself in weeks like this one — when semis are crashing, a famous fund is imploding, retail is puking stocks, and every instinct in your body is screaming get out before it gets worse.

The gut wanted out. The model held. And holding — at least so far — was right.

Not because I’m brave. Not because I “called it.” Because the model doesn’t trade on fear or headlines. It reads the market’s actual behavior, and despite all the noise, price action never entered the danger zones that would flip it to cash. Remember the mechanics: out of 158 daily zones, only 11 trigger a SELL. This week rattled the cage hard, but never quite tripped the wire.

Guidance for the week ahead

Here’s where discipline matters more than opinion.

The signal is a BUY, so the system says stay invested. But note that the next scheduled macro call lands on the first trading day of August — this coming week. That’s the model’s monthly checkpoint, and after a week this volatile, it’s worth watching closely.

My guidance: hold your positions and let the model do its job. If you were tempted to panic-sell into Wednesday’s dip, this week is the lesson — the investors who dumped at the fastest pace since COVID were selling near the lows of the week, right before the recovery. That is precisely the behavior a rules-based system exists to protect you from.

As always: this is not investment advice, and I’m not telling you what to do with your money. What I can tell you is what the discipline of the system calls for — follow the signal, don’t front-run the macro call, and don’t let a terrifying week talk you into a trade the model never asked for. The whole 26-year edge comes from doing what the signal says and nothing more.

The bigger picture: this is what “staying invested” looks like over time

Weeks like this feel unprecedented in the moment. They rarely are. Here’s the signal on SPY over the last 10 years — green means invested, red means in cash.

Find 2020 on that chart. Find the 2022 bear market. The red bands cluster exactly where the real damage happened — and notice how much green there is everywhere else. The system spends most of its life invested, riding the trend, stepping aside only for genuine danger. This week looked scary, but on a 10-year view it’s a blip. The discipline is in knowing the difference.

The last 10 flips

Full transparency, as always — every confirmed flip, with the SPY move captured or avoided during each:

A few honest observations:

• The current BUY has held since April 2 — 120 days — and is up +13.9%. The long holds are where the system makes its money, and this is one of them.

• The whipsaws are in there too — look at the 1-day SELL in Feb 2025. We publish every confirmed flip, wins and losses both. That’s the deal.

• This week isn’t in the table because the signal didn’t flip — and that’s the whole point. The best weeks for a timing system are often the ones where it does nothing at all.

Full update next weekend. Until then — and especially after a week like this — stay disciplined.

Disclaimer: This is not investment advice or a recommendation, and I am not a registered investment adviser. This publication is for informational and educational purposes only. The signal reflects a backtested and now live-tracked quantitative model; past performance, backtested or live, does not guarantee future results. Markets involve risk, including loss of capital. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

At the end of the QQQ post I promised this one — the same signal on TQQQ, ProShares’ 3x-leveraged Nasdaq-100 fund — “with its own serious risk warnings.” This is that post. The numbers are the largest I’ve ever produced, and so are the caveats. I’m going to give you both, in that order of importance: the caveats first, then the numbers.

Elevate ANY investment strategy with Quantitative Market Signals & High-Potential stocks. Sidestep major market crashes. 26+ year backtested QQQ performance: 29% CAGR with just 10% drawdown.

Read this before the charts

TQQQ is not QQQ. It is built to deliver three times the daily move of the Nasdaq-100. That leverage is a double-edged sword, and you need three facts in front of you before a single dollar figure:

• This backtest starts in March 2010 — the month TQQQ launched — which means it never lived through the 2000 dot-com crash or the 2008 financial crisis. Buy-and-hold QQQ fell 81% in dot-com; a 3x version would have been all but wiped out. Both columns you’re about to see are flattered by a start date sitting near the bottom of a historic bull market.

• Leverage decays. Because TQQQ resets 3x every day, choppy, sideways markets bleed value out of it even when the index goes nowhere. Holding it blindly is how people get hurt.

• The billion-dollar figure is a hypothetical backtest. Nobody actually turned $10,000 into $1.5 billion. Taxes, spreads, slippage, and the fund’s own costs would take large bites out of any real-world version. I’ll come back to this.

With all of that on the table: the reason this post exists isn’t the size of the ending number. It’s how the number was produced — by refusing to be invested during the exact declines that destroy a leveraged fund. That behavior is the whole story. The dollar figure is just what it looks like when you compound it on 3x leverage for sixteen years.

$10,000, two paths, 16+ years

Here is the same $10,000 invested in TQQQ from March 2010 to July 2026. One line simply buys and holds the 3x fund. The other follows my timing signal — in TQQQ when it says BUY, in cash (Treasury bills) when it says SELL.

The timing signal turned $10,000 into $1,556,036,008. Buy and hold turned the same $10,000 into $3,231,561.

Both are enormous — buy-and-hold TQQQ from the bottom of 2010 was itself a rocket. But the signal produced roughly 480 times the ending wealth of holding the exact same fund. The CAGR gap tells the story cleanly: 107.14% a year with the signal, versus 42.19% for buy and hold. On a 3x instrument, a 65-point annual gap doesn’t add up over sixteen years — it explodes.

Why does timing matter even more here than it did on SPY and QQQ? Because of what leverage does on the way down.

The 79% crash — and why leverage makes it lethal

On an unleveraged index, a crash is a setback. On a 3x fund, a crash is closer to an extinction event, because the math of recovery turns brutal.

A buy-and-hold TQQQ investor watched the fund fall 79.08% in 2022. To climb back from a 79% loss, you don’t need a 79% gain — you need a gain of roughly 378% just to return to even. It took about two years to recover, and the position spent more than three years underwater. And 2022 wasn’t the only one: buy-and-hold TQQQ also fell about 49% in the COVID crash and about 49% in late 2018.

The timing signal’s worst drawdown over the same 16+ years was −26.03% — a single ugly month in October 2020 that it recovered from in two.

Look at the stress numbers from the report:

• COVID-19 crash (Jan–Mar 2020): signal 0.00% vs buy & hold −49.12%. The signal was already in cash when the market fell.

• 2022 bear market: signal max drawdown −26.03% vs buy & hold −79.08%.

This is the point about leverage decay made concrete. A 3x fund that avoids the worst months doesn’t just skip the pain — it skips the compounding damage that a 79% hole inflicts on everything that comes after it.

Every risk measure, side by side — 16+ years

Here’s the full dashboard rather than a cherry-picked stat. Straight from the Portfolio Visualizer report:

Read the worst-year line again. Over sixteen-plus years — through the 2018 selloff, the COVID crash, the 2022 bear market, and every wobble in between — the signal’s worst calendar year on a 3x leveraged fund was positive, up 24.11%. Buy and hold had a −79.08% year.

The risk-adjusted numbers say the same thing from another angle. A Sharpe of 1.95 versus 0.89. A Sortino of 7.09 versus 1.57. A downside-capture ratio of 19.6 — meaning the signal absorbed less than a fifth of the fund’s down moves — versus 100 for buy and hold. Higher return, lower volatility, a fraction of the drawdown. That combination isn’t supposed to exist, and on an unleveraged fund it barely does. Leverage is what widens the gap this far: the same discipline that helped on SPY and QQQ is worth multiples more when every avoided decline is a 3x decline.

Why leverage changes the whole equation

It’s worth being explicit about the mechanism, because it’s the reason a 3x fund is the most dramatic showcase for this system and the most dangerous thing to hold without it.

A leveraged ETF resets its exposure daily. In a steady uptrend, that daily compounding works for you — which is how buy-and-hold TQQQ still reached $3.2 million. But in volatile or sideways markets, the daily reset works against you: the fund can lose money even when the index finishes flat. That’s volatility decay, and it’s why every deep drawdown on a 3x fund does double damage — first the loss itself, then the drag on everything that follows.

The signal’s entire job is to cut off that left tail. It stays invested to capture the leveraged upside and steps aside — into T-bills — during the declines where leverage compounds against you. On SPY that discipline was valuable. On QQQ it was powerful. On TQQQ it’s the difference between a fund that’s essentially un-hold-able and one whose worst year was green.

The same system, a far more powerful engine

Nothing about the method changed from the SPY and QQQ posts. It’s the same two-layer signal:

• A macro model that makes one BUY/SELL call on the first trading day of each month.

• A daily model that divides the market into 158 zones, only 11 of which trigger a SELL — so it stays invested through ordinary volatility and steps aside only for genuine danger.

The only thing that changed is the instrument. On the S&P 500 the same discipline compounded $10K into $1.75M. On the Nasdaq-100 it compounded into $9.2M. On a 3x Nasdaq fund it compounded into far more — because there was more upside to capture on the way up, and a far deeper set of craters to avoid on the way down.

The serious risk warnings I promised

I meant it when I said the caveats matter more than the number. If you take nothing else from this post, take this section.

• Backtested, hypothetical results. These figures come from applying the signal to historical data. They are not a live track record. No investor actually earned $1.5 billion, and backtests always look cleaner than reality.

• The start date does a lot of work. Beginning in March 2010 excludes the dot-com crash and 2008 entirely. On a 3x fund, those events would have been close to fatal for a buy-and-hold holder. Don’t read either column as a promise about the next sixteen years.

• Real-world frictions are large — and worst on leverage. Every SELL and BUY is a taxable event; in a normal account these are short-term gains taxed as ordinary income, which is a serious drag on any strategy that trades. Add bid/ask spreads, slippage on signal days, tracking error, and TQQQ’s own expense and financing costs, and the real-world outcome is materially lower than the backtest.

• Leverage decay is real. Sideways and choppy markets erode a daily-reset 3x fund even when the index is flat. This cuts both ways and can make live results diverge from any model.

• The signal is not magic. It still took a −26% month in 2020, and it whipsawed in and out repeatedly in years like 2022. A missed, late, or mistimed trade around a crash matters far more on 3x than on 1x.

• Concentration. When invested, this is 100% in a single leveraged ETF — no diversification, by design.

• Leveraged ETFs carry issuer and regulatory warnings that they are generally not intended to be held long-term. Holding TQQQ blindly is exactly the behavior this system is built to avoid.

None of this erases the edge — sidestepping the worst declines is genuinely valuable, and the risk-adjusted numbers are real features of the backtest, not accidents. But the honest framing is: this is the highest-octane version of a disciplined system, not a money printer.

Which version is right for you

Across three posts you’ve now seen the same signal on three instruments, in rising order of aggression:

• SPY — steadiest ride, −8.5% worst drawdown. Good for an IRA/401(k) or anyone who wants the smoothest path.

• QQQ — higher octane, −9.97% worst drawdown, far higher returns. For those who can stomach a more volatile index.

• TQQQ — maximum aggression, −26% worst drawdown with the signal, −79% without it. Only for those who fully understand leverage and would never hold it unhedged.

All three beat buy and hold on every risk-adjusted measure. But “highest ending number” is not the same as “right for you.” Most investors should not put everything into a 3x fund, signal or no signal. The value of the TQQQ post isn’t a recommendation to pile in — it’s proof of how much a single discipline is worth when the downside is leveraged.

This is the same system, twenty years in the making, running on one daily signal. From here I’ll publish where it stands every week, and every flip the day it happens — wins and whipsaws both. The live record continues.

The simplest market timing signal in the world. One signal. One trade when it flips. Under five hours a year.

Disclaimer: I am not a registered investment adviser, and nothing here is personalized investment advice, a recommendation, or an offer of advisory services. This publication is for informational and educational purposes only. Leveraged ETFs such as TQQQ carry substantial risk, including volatility decay and the potential for rapid, severe losses, and are generally not designed to be held long-term. Performance figures are generated by Portfolio Visualizer from historical data and a backtested application of the signal; backtested results are hypothetical, have inherent limitations, do not reflect trading costs or taxes, and do not guarantee future results. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

Image Credit: Photo by Anthony Maw on Unsplash

The “Magnificent Seven” — Nvidia, Apple, Microsoft, Alphabet, Amazon, Meta, Tesla — are now worth roughly $22 trillion combined, about a third of the entire S&P 500. Add the wider AI complex around them — the Broadcoms, the TSMCs, the Oracles, the data-center and power-infrastructure names whose valuations now lean on the AI story — and you’re looking at a bet worth north of $27 trillion.

Here’s the part that should really get your attention: through 2026, AI-related stocks have driven more than 80% of the S&P 500’s entire gain. Strip them out, and the index is up around 2% on the year.

That is the most concentrated bet on a single theme in the history of the U.S. market. Which makes it the most important question in investing today: what happens if it unwinds?

My honest view (and I could be wrong)

Let me tell you where I actually stand, because I’d rather give you a real opinion than hide behind a hedge.

I think this bull market has room to run a little further. My base case is that we get 10–15% drawdowns along the way — normal, healthy corrections — but not a dot-com-scale collapse. Not yet. Unlike the 2000 dot-coms, the biggest AI names are earnings-driven: the revenue and margins underneath them are real, and that changes the math on how and when this ends.

That’s my personal read. It is not the consensus, and it is absolutely not a guarantee. I’ve spent twenty years building quantitative systems precisely because opinions — mine included — are so often wrong. My market timing signal will tell me when the trend actually breaks, no matter what I believe right now.

But plenty of thoughtful investors are convinced this is a bubble on borrowed time. If that’s you, the rest of this post is for you.

The rotation nobody prepares for

Here’s the pattern worth understanding. When a dominant theme unwinds, capital rarely leaves the market — it rotates. It flows out of the crowded, expensive, momentum names and into the corners everyone ignored on the way up: defensives, value, cash generators trading at multiples that already price in gloom.

In a real AI unwind, the stocks best positioned aren’t the ones fighting the tide. They’re the ones with little AI exposure to begin with, trading at valuations low enough that there’s not much air left to come out.

So I put a specific question to my quantitative models: which stocks screen as most likely to weather an AI-driven storm — and rebound hardest once it passes? Five names came back. Not one of them is an AI stock.

The Patala Portfolio publishes a 26-year-backtested market timing signal and quant-selected stocks, all in the open.

The 5 stocks built to weather the storm

Figures approximate, as of market close Friday, July 24 2026. These are quantitative screens illustrating a rotation thesis — not personalized recommendations. See the disclaimer below.

1. Centene Corp (CNC) — Health Care

   

   Centene is the largest Medicaid managed-care insurer in the U.S., also running Medicare Advantage, Part D, dual-eligible, and ACA Marketplace (Ambetter) plans. Its demand is driven by demographics and government policy — about as far from the AI trade as a large-cap gets.

   The last twelve months were a rollercoaster: shares fell as much as ~60% in 2025 on cost deterioration across Medicaid, Medicare, and the Marketplace — then came roaring back. A big Q1 2026 earnings beat sent the stock up ~9% in a day and drove a series of guidance raises. It’s now up over 60% year-to-date in 2026, though still below year-ago levels given the depth of that drawdown. The next catalyst is the Q2 print due July 28. A low-multiple, defensive healthcare name mid-turnaround is exactly the profile that attracts inflows in a risk-off rotation.

2. Viatris Inc (VTRS) — Health Care

   

   Viatris is the global generics-and-established-medicines maker formed from Mylan and Pfizer’s Upjohn unit, spanning branded drugs, complex generics, and biosimilars across four geographic segments. It’s a high-cash-flow, low-valuation pharma name that trades on its dividend and debt paydown far more than on growth.

   Its 2025 was soft amid generic pricing pressure and divestitures, but the stock recovered off its 2025 lows on debt-refinancing activity and a Q1 2026 earnings beat, with FY26 guidance reaffirmed. It’s a deep-value pharma name — the kind of unloved, cash-generative stock money rotates toward when momentum breaks.

3. CVS Health Corp (CVS) — Health Care

   

   CVS is the integrated healthcare giant spanning Aetna (insurance), Caremark (pharmacy-benefit management), and ~9,000 retail pharmacies. Its revenue base is enormous, recurring, and almost entirely insulated from technology cycles.

   It’s been one of the surprise performers of the year: up ~75% over the trailing twelve months, with roughly 40% of that gain in the most recent quarter alone, powered by a strong Q1 2026 beat, raised guidance, and improving Aetna margins. S&P Global revised its outlook to Stable from Negative on improved leverage. With a GLP-1 program expansion, a 56-year dividend streak, and earnings due Aug 5, it’s a defensive heavyweight that’s already proving it holds up — the kind of name that does well when high-flyers roll over.

4. BorgWarner Inc (BWA) — Consumer Discretionary

   

   BorgWarner is an auto-parts and propulsion supplier — turbocharging, thermal, and drivetrain systems across combustion, hybrid, and EV platforms — increasingly diversifying into data-center power and industrial applications. It’s the one cyclical name in this group, and it earns its place on valuation and momentum rather than pure defense.

   The stock is up roughly 80%+ over the past year, among the strongest in this set, on consistent earnings beats, new business wins in China and Europe, and analyst upgrades tied to its data-center diversification story. Some valuation models flag it as extended here — worth noting honestly — but an industrial with a non-AI growth angle is a classic rotation winner.

5. Bunge Global SA (BG) — Consumer Staples

   

   Bunge is one of the world’s largest agribusinesses — buying, storing, moving, and processing soybeans, corn, wheat, and oilseeds into food, feed, and biofuel across 50+ countries. Its business is tied to global food demand, harvests, and commodity cycles: about as uncorrelated to AI as a stock can be.

   Shares are up roughly 59% over the past year, supported by the July 2025 close of the transformative Viterra acquisition, integration synergies, and strong demand for vegetable oils and protein meals (helped by biofuel and sustainable-aviation-fuel policy). Be clear-eyed about the risks: net margins are thin, and the most recent quarter’s revenue came in light, with the next report due July 29 under scrutiny on Viterra execution. But a staples-adjacent commodity processor is a textbook defensive holding when growth stocks wobble.

The common thread

Look at what these five share: low forward valuations, defensive or non-tech revenue, and essentially zero dependence on the AI narrative. Healthcare, auto parts, agribusiness. Boring, in the best possible way.

That’s the entire point. In a genuine AI unwind, boring is what outperforms. These aren’t the stocks that lead the next bull market — they’re the ones designed to protect capital and rebound while the market finds its feet. Notice, too, that four of the five are already outperforming this year, well before any unwind. That’s what a real rotation candidate looks like: strength that doesn’t depend on the AI trade continuing.

To be clear about my own position one more time: I don’t think that unwind is imminent. But the whole discipline of investing is being prepared for the scenario you don’t expect. If the $27 trillion bet breaks, these five are where my models say the shelter is.

What’s coming next

Starting the first week of August, I’ll begin publishing my quant model’s highest-potential stocks — two every Monday — alongside the weekly market timing signal. Today’s five are a preview of the kind of screening behind them.

Disclaimer: I am not a registered investment adviser, and nothing in this post is personalized investment advice, a recommendation to buy or sell any security, or an offer of advisory services. This publication is for informational and educational purposes only. The five stocks discussed were surfaced by a quantitative screen and are presented to illustrate a rotation thesis — not as a solicitation to trade them. Financial figures are approximate, reflect trading data as of late July 2026, are believed accurate as of publication but not guaranteed, and should be independently verified before acting. Individual stocks carry substantial risk, including total loss of capital. Past performance and backtested results do not guarantee future results. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.

The Patala signal closed Friday still on a **BUY**, where it’s held since April 2, 2026 — 113 days. But this is the closest it has come to flipping in months.

Here’s where the signal sits right now, up close — the last eight weeks. Still green, but you can see the turbulence building into Friday’s close:

And zoomed out to the full year so far — the February SELL, the April re-entry, and the long BUY run since:

What happened last week?

It was a rough, choppy week, and the signal felt it. Two forces did most of the damage: rising oil prices put pressure across the board, and heavyweight earnings from Alphabet (GOOG), Tesla (TSLA), and Intel (INTC) injected sharp two-way volatility into the tape. SPY closed Friday at $738.93, down on the week, with several wide-range, heavy-volume sessions.

None of that, on its own, forces the signal out. Remember how the model is built: out of 158 daily zones, only 11 trigger a SELL. It’s designed to stay invested through exactly this kind of noise rather than flinch at every rough week. But the conditions last week pushed price close enough to those danger zones that the signal is now, genuinely, on the brink.

It did not flip. As of Friday’s close, the model is still **BUY** for Monday. But it’s the tightest it’s been in a while — and worth flagging honestly, in advance, rather than after the fact.

Guidance for the week ahead

Here’s where discipline matters more than opinion.

My guidance: Hold what you have, but don’t add new long positions here. The signal hasn’t flipped, so this isn’t a SELL — you’re not being told to move to cash. But with the model this close to the edge and the next scheduled macro call landing on the close of the first trading day of August, opening fresh longs into that uncertainty is taking on risk the system isn’t rewarding right now.

This is the hardest part of any rules-based system, and it’s worth saying plainly: This is not investment advice, and I’m not telling you what to do with your money. What I can tell you is what the discipline of the system itself calls for — wait for the signal, don’t front-run it, and don’t let a choppy week talk you into a trade the model hasn’t asked for. The entire 26-year edge of this approach comes from doing what the signal says and nothing more. The moment you start improvising around it, you’re running a different system — usually a worse one.

We’ll have the next macro signal on the first trading day of August. Until then: patience.

The bigger picture: what “in cash” actually bought you

Since last week put the signal on the edge of a SELL, it’s a good moment to zoom out and show what those red “in cash” periods have actually done over time.

Here’s the signal applied to SPY over the last 10 years. Green = Invested. Red = In Cash.

Notice where the red clusters: the 2020 COVID crash and the 2022 bear market. The signal didn’t predict either — it simply reacted to danger zones faster than a headline ever could, and stepped aside while the worst of the declines played out. That’s the entire game. Not calling tops. Just not being fully exposed when the floor drops.

The last 10 flips

For full transparency, here are the signal’s ten most recent flips — the dates, the direction, the SPY price at each, and how long each new signal lasted:

A few honest observations:

- Whipsaws happen. February 2025: a SELL on the 13th, a BUY the very next day. That’s a whipsaw, and it costs a little. Over 26 years the signal has flipped 308 times — roughly 11–12 a year — and a meaningful share reverse quickly. That’s the price of a system responsive enough to catch real crashes.

- The good runs are long. The May 2025 BUY held 216 days. The current BUY has held 113 and counting. The system spends most of its time invested; cash periods are the exception.

- Every flip is published. Wins and whipsaws both. That’s the deal.

Good Luck!

That Patala Portfolio

*This is not investment advice or a recommendation, and I am not a registered investment adviser. This publication is for informational and educational purposes only. The signal reflects a backtested and now live-tracked quantitative model; past performance, backtested or live, does not guarantee future results. Markets involve risk, including loss of capital. Consult a qualified financial adviser before investing. You are solely responsible for your own decisions.*