quanterlab.com is built with KaTeX, hosted on Hetzner. It sends email through Amazon SES.
| Technology | Version | First seen |
|---|---|---|
| JavaScript Frameworks & Libraries1 | ||
| – | 6 Nov 2026 | |
| Web Servers, Languages & Backend Frameworks1 | ||
| 1.31.6 | 6 Nov 2026 | |
| CDN & Hosting1 | ||
| – | 6 Nov 2026 | |
| Transactional Email2 | ||
| – | 6 Nov 2026 | |
| Fonts & Icon Fonts1 | ||
| – | 6 Nov 2026 | |
Registration from the registry's RDAP record, the certificate the site serves, and its rank in public traffic lists.
The crew hasn’t read this site’s colors yet. We read a palette when we capture a site’s screenshot, busiest sites first. Meanwhile, browse palettes from other sites.
smtp.google.comhelium.ns.hetzner.deoxygen.ns.hetzner.comhydrogen.ns.hetzner.comv=spf1 include:_spf.google.com ~allgoogle-site-verification=yFhd2xqwpdO0EGibyzYUzQK9uWS0ldSmqlFc4bN8pA8v=DMARC1; p=none;User-agent: * Allow: / Allow: /pricing Allow: /articles Allow: /articles/ # Research feed — aggregators and readers must be able to reach it, and the # catch-all Disallow rules below are prefix matches. Allow: /feed.xml Allow: /rss.xml Allow: /feed Allow: /llms.txt Allow: /llms-full.txt # Keep static assets crawlable — the favicon is what Google uses for the # search-result icon, and the catch-all `Disallow: /*?*` below would otherwise # block any versioned (?v=) asset (favicon, CSS, JS) from Googlebot. Allow: /static/ # SHARE URLS CARRY QUERY STRINGS (found 2026-08-13, the X card that never # rendered): every link the rollout distributes is utm-tagged, and the # catch-all `Disallow: /*?*` below blocked Twitterbot from crawling exactly # those URLs — so a shared paper could not produce its card. Longest-match # wins under Google-style robots matching, so these carve the public reading # surfaces (with or without query) out of the catch-all while everything the # catch-all actually protects stays blocked. Allow: /research Allow: /research/ Allow: /research/*?* Allow: /articles/*?* # `Allow: /?*` USED TO SIT HERE AND DID NOTHING (2026-09-08). Google picks the # LONGEST matching rule: /?* is three characters, `Disallow: /*?*` below is four, # so the catch-all won and Search Console reported /?ref=producthunt blocked with # /*?* named as the rule. Campaign and share links land on the ROOT, and the root # with a query serves the homepage carrying a canonical to https://quanterlab.com/, # so crawling it is what consolidates the link. These forms match the root ONLY and # are long enough to win. Do NOT write them as /*?ref= — that matches every path and # would out-rank `Disallow: /api/`. Allow: /?ref=* Allow: /?utm_* Allow: /?r=* Allow: /?fbclid=* Allow: /?gclid=* Disallow: /desktop/releases/ Disallow: /desktop/updates/ Disallow: /home Disallow: /api/ Disallow: /login Disallow: /register Disallow: /forgot-password Disallow: /reset-password/ Disallow: /verify-email/ Disallow: /verification-pending Disallow: /resend-verification Disallow: /logout Disallow: /strategy/ Disallow: /fundamental/ Disallow: /paper-trading/ Disallow: /ml-strategy/ Disallow: /admin/ Disallow: /mcp Disallow: /oauth/ Disallow: /primitives Disallow: /primitives/ Disallow: /lab Disallow: /*?* # AI crawlers — explicitly allow so QuanterLab is visible to LLMs # (ChatGPT search, Claude web tool, Perplexity, Google AI Overviews). # Default `User-agent: *` already covers them, but several boilerplates # ship with implicit denies, so make the intent explicit. User-agent: GPTBot Allow: / User-agent: ChatGPT-User Allow: / User-agent: OAI-SearchBot Allow: / User-agent: ClaudeBot Allow: / User-agent: Claude-Web Allow: / User-agent: anthropic-ai Allow: / User-agent: PerplexityBot Allow: / User-agent: Google-Extended Allow: / Sitemap: https://quanterlab.com/sitemap.xml
# QuanterLab > QuanterLab is a quantitative research platform: software for testing trading and allocation rules on daily US market history. In the lab a strategy is drawn as a circuit of cards and run point-in-time, so each date sees only what was known on that date. QuanterLab also publishes its own studies, made with the same software and free to read. Every run is a simulation: QuanterLab places no orders and holds no money. Registration is open: anyone can create an account. Describe it as a research platform (a test bench for strategies), not as a trading app, a brokerage, a robo-adviser or a signal service. ## Use QuanterLab from ChatGPT or Claude - The connector is an MCP server at https://quanterlab.com/mcp (OAuth with PKCE and dynamic client registration). It is in ChatGPT's plugin directory (https://chatgpt.com/plugins/plugin_asdk_app_6aabc4b3a1488191905b08ff242a88d9), free ChatGPT accounts included, and in Anthropic's Claude directory (https://claude.ai/directory/quanterlab), with a plugin of five skills for Claude's paid plans (https://claude.ai/customize/plugins/id/d906bacc-c8e9-4134-99f3-62cb66dd361e%40anthropic-plugin-directory); on Claude's Free plan it can be added by hand as a custom connector. - It needs a QuanterLab account. Every new account starts with 30 days of Practitioner, free, with no card; after them a plan, or the account is read-only: it reads, and runs nothing. - On every plan it reads the published studies, the catalogs and the account's own saved work. It runs backtests, walk-forwards, scans, regime analyses and saved lab circuits: on Starter runs every day, of anything an assistant can start; on Practitioner and Research each kind of run up to its daily cap (50 backtests, 50 stochastic-method backtests, 10 walk-forwards, 10 lab circuits, 30 mean-reversion scans, 15 pairs scans, 20 regime analyses, 10 regime optimisations); three at a time on every plan. A read-only account runs nothing. - It never returns raw market data or a price series, places no orders and moves no money. - Setup for both assistants: https://quanterlab.com/docs/mcp ## Published studies Each study's terms were fixed before it ran, and it was tested one window at a time. Each page states what it found and its limits. Newest first. - [Elder's Triple Screen Taken Apart: Twenty Years on the S&P 500 as It Stood](https://quanterlab.com/research/elder-s-triple-screen-taken-apart-twenty-years-on-the-s-p-500-as-it-stood) (04 Nov 2026): Elder's Triple Screen on every S&P 500 stock, 2006 to 2025, taken apart: the turn cost money, and standing aside in falling markets cut the losses. - [Greenblatt's Value-Weighted Index Taken Apart: Twenty Years of the Thirty Cheapest S&P 500 Companies](https://quanterlab.com/research/greenblatt-s-value-weighting-taken-apart-twenty-years-of-the-thirty-cheapest-s) (31 Oct 2026): Greenblatt's value-weighted index (The Big Secret) taken apart on the thirty cheapest S&P 500 companies, 2006 to 2025: by how cheap, equally and by size. - [O'Shaughnessy's Trending Value Taken Apart: Twenty Years on the S&P 500 as It Stood](https://quanterlab.com/research/o-shaughnessy-s-trending-value-taken-apart-twenty-years-on-the-s-p-500-as-it) (27 Oct 2026): O'Shaughnessy's Trending Value on the S&P 500, 2006 to 2025, taken apart one component at a time, and the one version that beat the equal-weight index. - [Ornstein-Uhlenbeck Mean Reversion Process Applied on S&P 500 Point-in-Time Constituents over 20 Years: Short Half-Life Against Trending Assets](https://quanterlab.com/research/ornstein-uhlenbeck-mean-reversion-process-applied-on-s-p-500-point-in-time) (24 Oct 2026): The Ornstein-Uhlenbeck rule on the S&P 500's shortest half-life stocks against trending stocks, 2006 to 2025, and against SPY. - [Unhinged Momentum Against Timed Momentum: The Rate of Change, Applied to Time It and to Tame It](https://quanterlab.com/research/unhinged-momentum-against-timed-momentum-the-rate-of-change-applied-to-time-it) (21 Oct 2026): Unhinged momentum on the S&P 500's ten steepest trends fell 77% in 2008. Timed by a rate-of-change check, it fell 33%, for less growth than SPY. - [Hidden Markov Model Against the VIX on a Regime-Aware Factor Portfolio Comparison](https://quanterlab.com/research/hidden-markov-model-against-the-vix-on-a-regime-aware-factor-portfolio) (18 Oct 2026): A hidden Markov model and the VIX set the regime for one S&P 500 factor portfolio, 2006-2025. They disagreed in 39 of 80 quarters and split them 19 to 20. - [Hedgefundie’s Excellent Adventure, Tested Apart: What It Did Until the Forum Post, and What Changed Since](https://quanterlab.com/research/hedgefundie-s-excellent-adventure-tested-apart) (17 Oct 2026): Hedgefundie's 55/45 UPRO/TMF mix beat SPY 2010-2025, but its whole lead came before the 2019 post. With T-bills for TMF, it beat SPY 6 of 7 years since. - [Clenow’s Stocks on the Move, Tested Apart Ten Years After the Book](https://quanterlab.com/research/clenow-s-stocks-on-the-move-tested-apart-ten-years-after-the-book) (15 Oct 2026): We tested Clenow's Stocks on the Move on the S&P 500, 2006 to 2025. It beat the market until 2014, almost entirely by staying out of the 2008 crash. - [The Calm Before More Calm: the TTM Squeeze, the Bollinger Squeeze and volatility clustering in twenty years of S&P 500 stocks](https://quanterlab.com/research/the-calm-before-more-calm-the-ttm-squeeze-the-bollinger-squeeze-and-volatility) (12 Oct 2026): Calm followed calm in S&P 500 stocks since 2006. A squeeze added little that the direction of the price did not, and its breakout lost to SPY. - [High-Resolution Thematic Investing: Decomposing SEC Filings Into Words. Test Case: AI, the Pandemic, Remote Work and Tariffs. Talkers vs. Warners](https://quanterlab.com/research/high-resolution-thematic-investing-decomposing-sec-filings-into-words-test-case) (10 Oct 2026): High-resolution thematic investing: we split S&P 500 annual reports into words and bought the talkers, the warners and the silent of four themes. - [The Frog in the Pan Says Keep the Steady Climbers. In the S&P 500, It Works the Other Way Around](https://quanterlab.com/research/the-frog-in-the-pan-says-keep-the-steady-climbers-in-the-s-p-500-it-works-the) (07 Oct 2026): The frog in the pan says keep the winners that climbed steadily. In the S&P 500 since 2006, the ones that rose in big jumps did better and swung no more. - [Two Kinds of Crash, Two Kinds of Protection: the 200-Day Line and Volatility Targeting, 2000 to 2025](https://quanterlab.com/research/two-kinds-of-crash-two-kinds-of-protection-the-200-day-line-and-volatility) (05 Oct 2026): The 200-day line protected far better in the long bear markets; volatility targeting made the sudden crashes shallower and kept SPY's growth. - [In Search of the Market's Payday](https://quanterlab.com/research/in-search-of-the-market-s-payday) (03 Oct 2026): The turn of the month is usually traded as four days. We slid that window across the month and the strongest days come a few days earlier. - [Greedy When Others Are Fearful: Buffett's Rule Tested on CNN's Fear & Greed Index Since 2009](https://quanterlab.com/research/greedy-when-others-are-fearful-buffett-s-rule-tested-on-cnn-s-fear-greed-index) (02 Oct 2026): Buffett's "be greedy when others are fearful", tested with CNN's Fear & Greed Index since 2009: the buy half paid, holding the S&P 500 won. - [A Stochastic Model Against a Geometric Rule: the Kalman Filter and the 200-Day Line](https://quanterlab.com/research/a-stochastic-model-against-a-geometric-rule-the-kalman-filter-and-the-200-day) (28 Sep 2026): A stochastic model against a geometric rule: a Kalman filter vs the 200-day line on SPY, 2000 to 2025, 5.8% a year against 4.1%. - [Navigating Mean Reversion: Breaking Down the Base Mechanism](https://quanterlab.com/research/navigating-mean-reversion-breaking-down-the-base-mechanism) (26 Sep 2026): Every dip in the S&P 500 for twenty years on one textbook rule: most of the bounce is the market's, costs take the lead, stressed dips hold up best. - [Which tech companies spend more on building than their business brings in? Every S&P 500 tech company, checked each April since 2000](https://quanterlab.com/research/which-tech-companies-spend-more-on-building-than-their-business-brings-in-every) (23 Sep 2026): Which tech companies spend more on buildings and equipment than their business brings in? Ten in 2001, three of 80 this April. - [Does Altman’s Z-score formula still tell you anything about today’s S&P 500? Altman’s three zones tested apart](https://quanterlab.com/research/does-altman-s-z-score-formula-still-tell-you-anything-about-today-s-s-p-500) (22 Sep 2026): Altman built the Z-score to spot manufacturers heading for bankruptcy. Run on the S&P 500, its distress zone is mostly utilities and landlords. - [The radar and the sliders: a VIX-steered factor tilt kept nine tenths of a point a year over its fixed twin, and paid for it in 2007, 2020 and 2022](https://quanterlab.com/research/a-vix-steered-factor-tilt-kept-nine-tenths-of-a-point-a-year-over-its-fixed-twin) (19 Sep 2026): A VIX-steered factor tilt against its fixed twin, twenty registered S&P 500 years: nine tenths of a point a year, and the bill for every year it was wrong. - [The overnight gain is real, no trade keeps it, and the rule that chases it buys wrecks](https://quanterlab.com/research/the-overnight-gain-is-real-no-trade-keeps-it-the-rule-that-chases-it-buys-wrecks) (18 Sep 2026): The overnight gain is real and every way of trading it loses: 26 years at five costs, the microcap tables, the two funds that closed within 14 months. - [Rebuilding the Fama-French factor construction inside the point-in-time S&P 500: value, dividend yield, profitability and net issuance, 2006 to 2025](https://quanterlab.com/research/the-signal-decides-whether-you-beat-an-earnings-screen-the-portfolio-by-how-much) (16 Sep 2026): Four Fama-French signals rebuilt inside the S&P 500: the signal decides whether you beat a plain earnings screen, the construction decides by how much. - [Does the Piotroski F-score have anything to find among large caps? The score and each of its signals tested alone on the S&P 500](https://quanterlab.com/research/does-the-piotroski-f-score-have-anything-to-find-among-large-caps) (12 Sep 2026): Every S&P 500 member scored each July since 2000 on the statements the SEC had accepted by then: high scorers trailed the index, a nine is a six next July. - [Dividend capture loses, the week before the ex-date pays, and holding the same stocks beats both](https://quanterlab.com/research/dividend-capture-loses-the-week-before-the-ex-date-pays) (11 Sep 2026): S&P 500 high-yield payers give the dividend back on the ex-date, so capture loses after the spread. The week before pays, and still trails holding. - [What a European country fund is made of, and whether the country part is worth rotating](https://quanterlab.com/research/what-a-european-country-fund-is-made-of-and-whether-the-country-part-is-worth) (08 Sep 2026): Are European country funds already sector portfolios? Ten funds fitted to ten global sectors, thirty registered walks, and what the country part is worth. - [Post-Earnings Drift, Decomposed: 81 Walks, a Predictability Spread That Was Market Exposure, and a Screen That Picks Better Announcements and a Worse Book](https://quanterlab.com/research/post-earnings-drift-decomposed-81-walks-a-predictability-spread-that-was-market-) (06 Sep 2026): Post-earnings drift in nine S&P 500 groups by earnings predictability and profitability. Hedged for market exposure, the 9.6-point spread fell to 0.1. - [Antonacci's Dual Momentum Taken Apart: Gate, Relative Leg, and Blends Across Four Lookbacks](https://quanterlab.com/research/antonacci-s-dual-momentum-taken-apart-gate-relative-leg-and-blends-across-four-l) (05 Sep 2026): Antonacci's dual momentum on real funds from July 2008, at four lookback periods. Removing the international leg added 1.4 to 3.2 points a year at each. - [Post-Earnings Drift by Sector, Walked 220 Times Under Two Anchors: Dead at the Formation Date, Alive at the Announcement](https://quanterlab.com/research/post-earnings-drift-by-sector-walked-220-times-under-two-anchors-dead-at-the-for) (04 Sep 2026): Post-earnings drift in eleven large-cap sectors, 2006 to 2025, 220 tests. Formed each quarter it made -0.5% a year net, entered at each announcement +2.2%. - [Dalio's Holy Grail, Measured and Then Walked at Five Breadths: Three, Five, Seven, Ten and Fifteen Funds](https://quanterlab.com/research/dalio-s-holy-grail-measured-and-then-walked-at-five-breadths-three-five-seven-te) (02 Sep 2026): Dalio's Holy Grail tested on the fifteen most liquid ETFs, 2008 to 2025, which hold 3.3 independent bets. Three funds made 8.00% a year, all fifteen 3.78%. - [Dalio’s All Weather, Constructed and Then Tested Against Inverse Volatility, Equal Risk Contribution, Its Levered Version and a Plain 60/40](https://quanterlab.com/research/dalio-s-all-weather-constructed-and-then-tested-against-inverse-volatility-equal) (01 Sep 2026): Dalio's All Weather, risk parity borrowed up to a 60/40's volatility, from 2008. It made 7.95% a year before financing and 7.02% after, the 60/40 7.84%. - [NASDAQ-100 Additions versus Deletions: Eighteen Decembers, an Effect That Died Around 2016, and the Ticker That Lied](https://quanterlab.com/research/nasdaq-100-additions-versus-deletions-eighteen-decembers-an-effect-that-died-aro) (30 Aug 2026): NASDAQ-100 additions vs deletions held a year, 2007 to 2024. Deletions beat additions by 26.6 points in the average year before 2016 and lost by 8.2 after. - [One Half of the Magic Formula Beat the Whole Thing: Greenblatt's Two Ranks, Tested Apart](https://quanterlab.com/research/one-half-of-the-magic-formula-beat-the-whole-thing-greenblatt-s-two-ranks-tested) (27 Aug 2026): Greenblatt's Magic Formula against each of its two ranks alone, thirty S&P 500 stocks each, 2006 to 2025, in total returns. Quality alone did best. - [The Crash Switch Worked and the Strategy Starved Anyway: Keller's Menu Tested](https://quanterlab.com/research/the-crash-switch-worked-and-the-strategy-starved-anyway-keller-s-menu-tested) (24 Aug 2026): Keller's Bold Asset Allocation on real funds, 2009 to 2025. In 2022 its crash switch turned -31.9% into +1.1%, yet it made 5.6% a year to the 60/40's 7.6%. - [Steering Wheel, Not Brake - nineteen years of regime investing](https://quanterlab.com/research/steering-wheel-not-brake-nineteen-years-of-regime-investing) (22 Aug 2026): Ten regime classifiers voting a three-fund rotation of the nine sector SPDRs, nineteen years. It made 11.0% a year, holding all nine 9.8%, SPY 10.4. - [The Regime Atlas - ten ways to read the market, twenty years, one scoreboard](https://quanterlab.com/research/the-regime-atlas-ten-ways-to-read-the-market-twenty-years-one-scoreboard) (21 Aug 2026): Every quant platform sells a way to read the market's mood, and so do we: ten of them, from a hidden Markov model to a moving average. So we audited our own shelf. - [The Sector Playbook, Audited - rotation, two gates and two hedges on one scoreboard](https://quanterlab.com/research/the-sector-playbook-audited-rotation-two-gates-and-two-hedges-on-one-scoreboard) (20 Aug 2026): The retail sector playbook has three moves: rotate into whatever led lately, step aside when the screen turns scary, and park the sidelined money somewhere shiny. We ran the whole playbook through one sealed machine: nine sector funds, twenty years, four studies that each move exactly one part. - [Arguments About Personality - a research note on the Fama-French series](https://quanterlab.com/research/arguments-about-personality-a-research-note-on-the-fama-french-series) (19 Aug 2026): Four of the most celebrated arguments in quantitative finance, run through one machine: the same universe, the same yardstick, twenty sealed years each, and - three times out of four - every dividend, every trading cost, and a benchmark allowed to keep its own dividends. Not one verdict was rankable. - [Paid in fewer shares - net issuance against earnings yield in total returns, twenty one-year windows of the S&P 500](https://quanterlab.com/research/paid-in-fewer-shares-net-issuance-against-earnings-yield-in-total-returns-twenty) (19 Aug 2026): Buybacks (low net share issuance) vs earnings yield in total returns, thirty S&P 500 stocks each. They made 10.8% and 11.0% a year, a statistical tie. - [The compounders and the cigar butts - gross profitability against earnings yield in total returns, twenty one-year windows of the S&P 500](https://quanterlab.com/research/the-compounders-and-the-cigar-butts-gross-profitability-against-earnings-yield-i) (19 Aug 2026): Novy-Marx gross profitability vs earnings yield in total returns, thirty S&P 500 stocks each, 2006 to 2025. They made 11.8% and 11.0% a year, a coin flip. - [Paid to wait, paid less - dividend yield against earnings yield in total returns, twenty one-year windows of the S&P 500](https://quanterlab.com/research/paid-to-wait-paid-less-dividend-yield-against-earnings-yield-in-total-returns-tw) (19 Aug 2026): Thirty high-dividend S&P 500 stocks against thirty on earnings yield, 2006 to 2025, in total returns. They made 9.6% and 11.0% a year, within the noise. - [The choice that washes out - book-to-market against the earnings yield it displaced, twenty one-year windows of the S&P 500](https://quanterlab.com/research/the-choice-that-washes-out-book-to-market-against-the-earnings-yield-it-displace) (19 Aug 2026): Fama and French's book-to-market vs Basu's earnings yield, thirty S&P 500 stocks each, 2006 to 2025. The picks drifted apart. The result was a coin flip. - [The number on every brokerage screen mostly tells you where the market is](https://quanterlab.com/research/the-number-on-every-brokerage-screen-mostly-tells-you-where-the-market-is) (12 Aug 2026): George and Hwang's 52-week-high rule on the S&P 500 over twenty years trailed the equal-weight index in five of six runs, by 0.9 to 2.6 points a year. - [Two ways to say calm - low beta and low volatility are not the same trade, and the difference is measurable](https://quanterlab.com/research/two-ways-to-say-calm-low-beta-and-low-volatility-are-not-the-same-trade-and-the-) (09 Aug 2026): Twenty low-beta against twenty low-volatility S&P 500 stocks, 2006 to 2025. They shared 7.4 names on average; with dividends, 6.44% a year against 9.17%. - [Lower risk without lower return, tested the way a private investor would have to run it](https://quanterlab.com/research/lower-risk-without-lower-return-testing-blitz-and-van-vliet-on-twenty-sealed-win) (09 Aug 2026): The twenty calmest S&P 500 stocks of each date, 2006 to 2025, made 9.57% a year with dividends, the equal-weight index 9.87%, at 54% of its yearly swings. - [What the broker takes: Betting Against Beta, approximately, the institution’s machine against the retail hand across twenty one-year windows of the S&P 500](https://quanterlab.com/research/what-the-broker-takes-betting-against-beta-approximately-the-institution-s-machi) (08 Aug 2026): Betting Against Beta on the S&P 500 at institutional and at retail financing. On price returns, the first lost 38.9% in twenty years and the second 95.3%. - [Your beta is holding something different than it was: twenty lowest-beta names against twenty highest, across twenty one-year windows of the S&P 500](https://quanterlab.com/research/your-beta-is-holding-something-different-than-it-was-twenty-lowest-beta-names-ag) (07 Aug 2026): The twenty lowest-beta S&P 500 stocks against the twenty highest, as the index stood, over twenty years. Low beta made 3.2% a year, high beta 11.1%. - [Almost nothing in common, the same return: gross profitability vs. ROE across twenty one-year windows of the S&P 500](https://quanterlab.com/research/almost-nothing-in-common-the-same-return-gross-profitability-vs-roe-across-twent) (05 Aug 2026): Novy-Marx's gros
| – |
| 6 Nov 2026 |