Open-Alpha-Lab

An open quant research desk.

I take famous trading ideas — anomalies, folk strategies, things people swear by — put each through the same brutal protocol, and publish the verdict: edge or mirage.

Most are mirages. The honest write-up of why is the point.

---

Built by someone who ran the real thing — a fully systematic global-macro book scaled from sub-$100M to $9B+ in monthly traded notional — so every idea is judged on the two questions most repos skip: is the signal real? and does it survive real execution and scale?

How it works

Every idea goes through the same protocol and earns two stamps, so results are comparable:

Signal — is the effect statistically real?
Tradability — does it survive costs, capacity & scale?

Robust inference (Newey-West / Lo SEs, bootstrap CIs, White Reality Check for data-snooping), an honest alpha-vs-beta split, and a square-root market-impact capacity test — the full house style is written up in METHODOLOGY.md.

---

The map

The whole bench on one grid — every study is a numbered chip, sorted by its two stamps. Almost everything ends up bottom-right; two chips are green — and neither one predicts returns.

The counts, the mortality by family of idea, and the five lessons the bench keeps teaching are in What 210 teardowns taught us.

---

The studies

#	Study	The claim — tested to destruction	Real?	Tradable?
01	Overnight Anomaly	Do stocks really make all their money overnight?
02	Falling-Knife	Does buying the dip (−3%) beat a random day?
03	Fear-Gauge	Does buying the VIX spike pay?
04	Social-Oracle	Does following a viral crowd's stock picks pay?
05	Twin-Spread	Does textbook pairs trading still pay after everyone copied it?
06	Clockwork-Vol	Does the VIX run on a fixed-period cycle you can time?
07	Coiled-Spring	Does the "20-EMA breakout" deliver explosive +30% pops?
08	True-Strength	Is the "True" Strength Index truer than MACD/RSI?
09	Phantom-Kernel	Is Avellaneda-Stoikov's "optimal spread" built on a real order-arrival law — and does it matter to the P&L?
10	Markov-Mint	Can a Markov-chain pipeline "win every trade" on Polymarket?
11	Vanishing-Penny	How fast does a guaranteed $40M Polymarket arbitrage close?
12	Paper-Prophet	Does an ARIMA+GARCH stack forecast the SPY, or is it vol-targeting in a trenchcoat?
13	Crimson-Hour	Does a red opening hour + IB-rejection really call the close at 88%?
14	Gamma-Gospel	Does dealer gamma (GEX) call the day's character, or is it the VIX in a trenchcoat?
15	Sigma-Sleight	Does length-aware "AdaptiveRSI" beat fixed 70/30, or is the σ-transform a monotone relabel?
16	Storm-Shy	Does scaling exposure down when markets get loud actually pay — or is it just selling low?
17	Glass-Ceiling	Does a 1:1 resistance breakout have momentum to harvest, or are you just paying the spread to buy the high?
18	Dull-Roar	The low-volatility anomaly: free alpha on the modern S&P 500, or just a defensive low-beta tilt?
19	Rubber-Band	Internal Bar Strength: does a stock that closes near its low really snap back the next day — and can you still cash it?
20	Freight-Train	Time-series momentum: does riding the trend across many markets pay — and why would you hold a thin-Sharpe sleeve?
21	Fools-Gold	The "golden cross" (50/200 MA): is it a real buy signal, or a trend filter that only shines on the one index everyone quotes?
22	Crystal-Ball	An HP-filter detrending strategy backtests at Sharpe 2 — on a coin flip. Is it an edge, or is the filter peeking at the future?
23	Broken-Tether	Pairs trading: two assets drift apart, bet they snap back — but does the cointegration hold out of sample?
24	Stampede	Cross-sectional momentum: do past winners keep winning on the modern S&P 500 — and what does the crash cost?
25	Clean-Slate	Residual momentum — is the "cleaner cousin" of cross-sectional momentum actually cleaner on this tape?
26	Sand-Castle	A "mathematically optimal" stat-arb portfolio (w ∝ C⁻¹E) backtests beautifully — does inverting an estimated covariance help, or maximize the error?
27	Steamroller	The FX carry trade — borrow cheap, lend dear: a real premium, or rent paid in front of a steamroller?
28	Carousel	Sector rotation: does chasing the hottest sectors beat just holding all eleven equal-weight — or is it concentration risk for nothing?
29	Hedgers-Toll	Are speculators really paid (via CFTC positioning) for taking the other side of producers' commodity hedges — or has the toll booth closed?
30	House-Edge	Lever a vol-targeted dip-buyer to beat the market — does the edge survive the retail CFD markup?
31	Trade-Winds	Cross-asset time-series momentum — the cliché closest to surviving: does our 26-year tape certify it, or only the crisis-alpha sleeve?
32	Rip-Tide	Short-term contrarian on the same 18 futures as Trade-Winds — same machine, opposite sign: does fading the move pay net of turnover?
33	Slingshot	The equity mirror of Rip-Tide — fade each stock against its peers (dollar-neutral): a real reversal, or eaten by the spread?
34	Aftershock	Post-earnings drift — the decades-old premium: does it survive on a liquid S&P 500 universe after costs?
35	Contango	Commodity carry / roll yield — long backwardation, short contango: is the contango tax harvestable, or a −83%-drawdown curve-timing book?
36	Greenback	Dollar-carry, and the carry⊕momentum combo: can diversification fix the steamroller a vol overlay couldn't?
37	Barometer	The trend in macro data (growth, inflation) as a cross-asset signal: right-sided, but big and fast enough to trade?
38	Chorus	The capstone — blend three weak, decorrelated signals (momentum + reversal + low-vol): do they add up to one tradable book?
39	Black-Box	A neural net fed crypto OHLCV scores a dazzling in-sample Sharpe: does it survive walk-forward, or is it memorising noise?
40	Paper-Tiger	A vendor's published dual-momentum backtest, mirrored back: is the timing alpha real, or does it belong to the assets it holds?
41	Hangover	The January Barometer ("as goes January, so goes the year"): a real omen, or a base-rate illusion?
42	Last-Call	The turn-of-the-month effect — real, large and famous: can you actually beat buy-and-hold with it?
43	Free-Lunch	Betting against beta — the "low-risk free lunch": is it free, or is it the leverage bill?
44	Growth-Spurt	The asset-growth effect — the highest headline Sharpe on the vendor list: does it survive look-ahead-free on large caps?
45	Vanishing-Act	The size premium (Banz 1981), finance's original factor: is it still there on the longest free proxy?
46	Bargain-Bin	The value premium (HML) on tradable value/growth ETF pairs: a dependable edge, or a regime bet?
47	Paper-Moon	The Fed Model (E/P vs the 10-year yield): a real timing rule, or a money-illusion bug?
48	Groundhog	Return seasonality — a stock repeats its calendar-month performance (Heston-Sadka): real signal, or astrology?
49	Black-Gold	Does the oil price predict the stock market (Driesprong 2008) — still, out of sample?
50	High-Water	The 52-week-high effect (George-Hwang): a distinct anomaly, or just momentum wearing a hat?
51	Blue-Chip	The quality (gross-profitability) premium — the factor that survived where size and value faded (Novy-Marx): does it, on real SEC data?
52	Smoke-Screen	The accruals anomaly (Sloan 1996) — cash-backed earnings beat accrual-heavy ones: does it replicate on real SEC data?
53	Jackpot	The lottery (MAX) effect — high single-day-gain stocks should underperform (Bali et al.): do they, on large caps?
54	Static	The idiosyncratic-volatility puzzle (high-vol → low returns, Ang et al.): does it hold on large caps?
55	Summer-Lull	"Sell in May" — the Halloween seasonal: real enough to beat buy-and-hold?
56	Tide-Table	Does the Shiller CAPE forecast long-run returns — and can it time the market, or only set expectations?
57	Yield-Trap	Do high-dividend stocks beat the market on total return?
58	Bunker	Does the min-vol ETF (USMV) deliver the low-vol anomaly — free Sharpe, or just less risk?
59	Downhill	Riding the yield curve for the term premium: real, but worth the duration risk?
60	Long-Shot	The skewness (lottery) effect in commodities — low-skew beats high-skew: does it point the right way, and pay?
61	Slow-Burn	Do 3× leveraged ETFs (TQQQ) decay to zero or amplify — and is there a free lunch?
62	Premium-Seller	Does a covered-call "income" ETF (QYLD) beat the index it holds?
63	Free-Fall	The short-volatility carry — selling vol earns a real premium: can you keep it through the steamroller?
64	Share-Shuffle	Do firms that issue stock underperform the ones buying it back (the net-issuance anomaly) — on large caps?
65	Scorecard	Does Piotroski's 9-point F-score (the famous fundamental-health screen) sort large-cap winners?
66	Inverted	Does an inverted yield curve forecast equity downturns — and is it a usable sell button?
67	Fed-Drift	Do stocks drift up before FOMC meetings (Lucca-Moench) — still, after everyone knows?
68	All-Weather	Does Dalio's risk-parity portfolio beat everything, or just diversify better?
69	Safe-Haven	Does gold hedge inflation and crashes — or one, or neither?
70	Digital-Gold	Is bitcoin "digital gold" — a crash-and-inflation haven, or just a high-octane risk asset?
71	Ambush	Four edges that each died to daily turnover (low IBS, turn-of-month, red close, VIX stress), gated to fire only at their rare confluence: does the overlay pay?
72	Loaded-Dice	The 5-minute SMA(5/10) crossover scalp — "a coin flip with the trend on your side": is the trend really on your side?
73	First-Light	Does the 5-minute opening-range breakout earn the returns its viral 2023 backtest claims?
74	Mind-the-Gap	Does an opening gap always get filled?
75	Knee-Jerk	Does Connors' RSI(2) oversold bounce still pay — or did publishing it wear it out?
76	Rice-Paper	Do Japanese candlestick reversals predict anything a coin doesn't?
77	Golden-Mean	Do Fibonacci levels and round numbers hold better than random price levels?
78	Crossed-Wires	Is the MACD crossover any more than a slower coin flip?
79	Sleigh-Ride	Is the Santa Claus rally real — and does a failed one warn of trouble ahead?
80	Cold-Open	Does January's direction really call the rest of the year?
81	Four-Year-Itch	Is year three of the presidential cycle really the market's best?
82	Witching-Hour	Does triple-witching expiry move the market, or just the volume?
83	Half-Life	Does Bitcoin reliably pump after each four-year halving?
84	Moon-Math	Does the Stock-to-Flow model actually predict Bitcoin's price?
85	Dr-Copper	Does the copper/gold ratio forecast the economy, or just echo it?
86	Tail-Radar	Does the CBOE SKEW index see black swans coming?
87	Center-Line	Does price really get pulled back to the intraday VWAP?
88	Dogs-of-the-Dow	Buy the ten highest-yielding Dow stocks each January — do you really beat the Dow?
89	Turn-of-the-Month	Do almost all the market's gains really land in the ~4 days around month-end?
90	Weekend	Are Mondays really negative, and does buying Tuesday beat the market?
91	Death-Cross	When the 50-day crosses below the 200-day, should you really sell?
92	Easy-Money	Vol ETPs bleed lower every day — so can you just short them and collect free money?
93	Round-Numbers	Prices pile up at whole dollars — but can you trade the magnetism?
94	Level-Pegging	Does equal-weighting the S&P always beat cap-weighting?
95	Holiday-Cheer	Do stocks really pop the day before every market holiday — and can you buy it?
96	New-Year-Pop	Do small-cap stocks really pop versus large caps every January?
97	Balancing-Act	Is the classic 60/40 portfolio the sensible default — and do bonds really cushion every crash?
98	High-Noon	Is buying at an all-time high really the riskiest moment to invest?
99	Safety-Net	Does a trailing stop-loss really protect your capital and improve returns?
100	Melting-Ice	Do 3x ETFs like TQQQ really "decay to zero", so you must never hold them overnight?
101	Slow-and-Steady	Dollar-cost averaging: safer and smarter than going all-in, or just a way to stay less invested?
102	Free-Rebalance	Is rebalancing really a free lunch that adds return on top of the risk control?
103	Turtle-Trader	Did the legendary Turtle breakout actually beat entering at random?
104	Bollinger-Reversion	Does piercing the lower Bollinger band beat just buying a random day?
105	Coppock-Curve	Can a 1962 indicator built for grief really time bear-market bottoms?
106	Supertrend	Does TradingView's most-watched ATR flip actually beat a coin?
107	Stochastic-Oscillator	Does the stochastic oversold crossover beat a coin?
108	ADX-Filter	Does "only trade when the trend is strong" (ADX > 25) add anything?
109	OBV-Divergence	Does on-balance volume really precede price?
110	Faber-Timing	Does the famous 200-day timing rule beat buy-and-hold, or just cut risk?
111	VIX-Term-Structure	Does the VIX term-structure slope time the market?
112	Move-Index	Does bond-market volatility (MOVE) warn of equity drawdowns?
113	Gold-Silver-Ratio	Does the gold/silver ratio reliably mean-revert into a trade?
114	Dollar-Smile	Does the dollar's direction forecast stocks, or just move with them?
115	Credit-Spreads	Does widening high-yield credit warn equities before they fall?
116	Power-Hour	Does the last "power hour" continue the day's move — or fade it?
117	Pi-Cycle-Top	Does Bitcoin's Pi-Cycle Top indicator really call the top?
118	Fed-Model	Does the earnings yield minus the 10-year bond yield tell you when to own stocks?
119	Real-Rate-Regime	Should you step aside when real long rates are high or rising?
120	Excess-CAPE-Yield	Does Shiller's ECY forecast the next decade's equity risk premium?
121	Magic-Formula	Does Greenblatt's quality-plus-value rank beat the S&P 500?
122	Gross-Profitability	Does Novy-Marx's gross profitability still earn a spread on large caps?
123	Altman-Z	Do financially distressed (low Altman-Z) stocks pay a premium?
124	Cash-Flow-Yield	Does buying high operating-cash-flow yield beat plain earnings yield?
125	Ichimoku-Cloud	Does the Ichimoku cloud system beat a coin?
126	Parabolic-SAR	Does Wilder's stop-and-reverse flip beat a coin once you pay the whipsaws?
127	Williams-R	Does the Williams %R oversold signal predict a bounce?
128	Keltner-Channel	Keltner breakout or Keltner reversion — is either one signal?
129	Heikin-Ashi	Do smoothed Heikin-Ashi candles filter noise into an edge?
130	Vol-Risk-Premium	Is selling the gap between implied and realized vol a free lunch?
131	Utilities-Canary	Do utilities leading the market warn of a coming drawdown?
132	Yield-Curve-Steepener	Does the curve slope tell you when to own the long bond?
133	Crypto-Seasonality	Is "Uptober" — or any month — a real edge in Bitcoin?
134	Bitcoin-Dominance	Does falling Bitcoin dominance call "alt season"?
135	FOMC-Cycle	Do stock returns really hide in the even weeks of the Fed cycle?
136	Mark-Twain	Is October really the most dangerous month to own stocks?
137	Mansfield-RS	Does Weinstein's Stage-2 relative-strength screen beat the market?
138	Random-Forest	Does a walk-forward Random Forest on price features beat a shuffled-label control?
139	AI-Powered-ETF	Can IBM Watson's AI ETF beat a dirt-cheap S&P 500 index fund?
140	Amihud-Illiquidity	Does sorting on price-impact illiquidity earn a premium, or just survivorship?
141	Turnover-Anomaly	Do low-turnover stocks deliver the Datar-Naik-Radcliffe liquidity premium?
142	Split-Drift	Do stocks drift up after a split takes effect?
143	Dividend-Capture	Can you pocket a dividend by buying just before the ex-date?
144	Permanent-Portfolio	Does Harry Browne's four-way mix really survive every regime?
145	Home-Bias	Does diversifying abroad still improve a US investor's portfolio?
146	Country-Momentum	Does rotating into the strongest-trending country ETFs pay?
147	FX-Momentum	Does currency momentum still pay after costs?
148	Lunar-Effect	Do stocks really return less around the full moon?
149	Daylight-Saving	Does the market really slump after the clocks change?
150	SAD-Effect	Do shorter autumn days really depress stock returns?
151	Stocks-For-Long-Run	Do stocks really always win over the long run?
152	Inflation-Hedge	Are stocks really an inflation hedge?
153	Net-Operating-Assets	Does a bloated balance sheet predict low returns?
154	Leverage-Anomaly	Do high-leverage firms earn the higher return theory promises?
155	Asset-Turnover	Do capital-efficient (high asset-turnover) firms beat the market?
156	Martingale	Does averaging down ("martingale") beat just buying and holding?
157	Kelly-Sizing	Does Kelly-optimal position sizing beat fixed sizing in practice?
158	Super-Bowl	Can a football game predict the stock market?
159	Presidential-Party	Do stocks really do better under Democrats — or did the crashes just land on one party's watch?
160	Skyscraper-Curse	Does the world's next record skyscraper signal a crash?
161	Year-Ending-Five	Are years ending in 5 really the best for stocks?
162	Rosh-Hashanah	Should you really "sell Rosh Hashanah, buy Yom Kippur"?
163	Friday-13th	Is Friday the 13th unlucky for the stock market?
164	Mercury-Retrograde	Does Mercury retrograde really wreck the market?
165	Chinese-Zodiac	Is the Year of the Dragon lucky for stocks?
166	First-Five-Days	Do January's first five days really forecast the whole year?
167	Hindenburg-Omen	Does the Hindenburg Omen actually warn of crashes?
168	Advance-Decline	Does a fading advance-decline line call the top?
169	Fluent-Tickers	Do stocks with pronounceable tickers outperform?
170	Alphabetical-Bias	Do early-alphabet tickers get an edge?
171	Naive-1-Over-N	Does "optimal" portfolio math beat a naive equal split?
172	Hundred-Minus-Age	Is "100 minus your age in stocks" smart, or just less invested?
173	Four-Percent-Rule	Does the 4% retirement withdrawal rule actually survive history?
174	Bitcoin-Rainbow	Does Bitcoin's Rainbow Chart actually time the cycle?
175	Crypto-Weekend	Is there a tradable Bitcoin weekend effect?
176	Hot-Hand	After a winning streak, is the market "hot" — or "due" for a reversal?
177	Megacap-Concentration	Does just buying the biggest stocks beat the index?
178	CCI	Does Lambert's CCI oscillator time daily equities?
179	Aroon	Does Chande's Aroon(25) crossover point the right way?
180	TRIX	Does triple-smoothing an EMA filter false signals, or just delay the good ones?
181	Ultimate-Oscillator	Does Williams' three-period oscillator beat a coin at the extremes?
182	Vortex-Indicator	Does the VI+/VI- crossover call the daily trend?
183	Fisher-Transform	Does Ehlers' Fisher Transform sharpen turning points into an edge?
184	Williams-Fractals	Do Bill Williams' fractals mark tradable swing points?
185	Chande-Momentum	Does the Chande Momentum Oscillator beat a coin?
186	Morning-Star	Does the morning-star candle pattern call a reversal?
187	Three-Soldiers	Do three white soldiers (or black crows) predict what comes next?
188	Head-Shoulders	Does the head-and-shoulders pattern actually forecast the drop?
189	Double-Top	Do double tops and bottoms call reversals?
190	NR7	Does the narrowest range in 7 days precede a tradable breakout?
191	Leap-Year	Are leap years good or bad for stocks?
192	Tax-Day	Is there a tax-day effect around April 15?
193	Window-Dressing	Do funds pump winners into quarter-end — and can you trade it?
194	Turkey	Do stocks really feast around Thanksgiving?
195	Monthly-OpEx	Does monthly options-expiration week move the market?
196	Long-Term-Reversal	Do five-year losers really beat five-year winners (De Bondt-Thaler)?
197	Dividend-Payout-Ratio	Does a higher payout ratio really predict higher growth (Arnott-Asness)?
198	Cash-Holdings	Do cash-rich firms earn higher returns?
199	Sales-Growth	Do fast-growing "glamour" stocks underperform (Lakonishok-Shleifer-Vishny)?
200	ROE-Quality	Does buying high-ROE quality beat the market?
201	Dividend-Growth	Do dividend growers outperform?
202	Fifty-Two-Week-Low	Does buying stocks near their 52-week low pay?
203	Golden-Butterfly	Does the Golden Butterfly beat plain stocks — and its simpler parent?
204	Talmud-Portfolio	Does a 2,000-year-old "thirds" rule still hold up?
205	Three-Fund	Is the Bogleheads three-fund portfolio actually better than 100% stocks?
206	Dividend-Aristocrats	Do the Dividend Aristocrats beat the market?
207	REITs-Diversifier	Are REITs a real diversifier, or just leveraged stocks?
208	Gold-Miners	Are gold miners really leveraged gold?
209	ETH-BTC-Ratio	Can you ride the ETH/BTC rotation?
210	Crypto-Trend	Does a 200-day timing rule tame Bitcoin's crashes?

Click any study for the full teardown — two narrative notebooks (one for the curious, one for the quant), reproducible code, and every number behind the two stamps.

In the queue: the weekend effect, the payday anomaly, paired switching, a VIX term-structure follow-up. Suggestions welcome via issues.

---

Run it

python -m venv .venv
# Windows:  .venv\Scripts\Activate.ps1   |   *nix:  source .venv/bin/activate
pip install -r requirements.txt

pytest -q                                                            # the engine's test-suite
python studies/01-overnight-anomaly/examples/run_synthetic_demo.py   # offline, no network

Then open studies/01-overnight-anomaly/ — start with the notebook for the curious, or read the working paper.

To verify the published numbers byte-for-byte (data caches, fingerprints, release bundle), see docs/reproducibility.md.

The engine — quantlab/ (a small, tested, reusable toolkit that powers every study)

Module	Role
decompose.py	Exact overnight/intraday/close-close return decomposition + Sharpe summary.
data.py	Yahoo fetch + parquet cache; split/total-return/raw adjustment modes.
diagnostics.py	Critique layer (offline): compounding, split-artefact injector/detector, synthetic markets.
backtest.py	Cost-aware backtest, break-even cost, cost sweep.
stats.py	Bootstrap Sharpe CIs, alpha-vs-beta (gap-risk) decomposition.
analytics.py	HAC & Lo (2002) inference, calendar-time normalization, rolling-Sharpe decay, market-impact capacity.
universe.py	Firm-level cross-section across an index (S&P 500 breadth).
simulate.py	Adversarial steelman of a strategy/manipulator P&L vs capital.
bayes.py	Bayesian hypothesis posteriors + White (2000) Reality Check.
plots.py	Decomposition / grid plots.
repro.py	Reproducibility stamp: pin an as-of date + content fingerprint so headline numbers reproduce.
brokers/	Swappable BrokerBase + MT5 template (dry_run=True).

Open-Alpha-Lab/
├── quantlab/        # the reusable research engine
├── tests/           # deterministic test-suite (CI on 3.10–3.12)
├── studies/         # one folder per study: notebooks, code, data, docs
└── pyproject.toml · CITATION.cff · LICENSE

A CITATION.cff is provided — use GitHub's "Cite this repository" button.

---

Built by Guillain d'Erceville — production systems, trading & market-data plumbing, and a habit of publishing the dead-ends, not just the wins.

Not investment advice. Research & education only. See LICENSE.