Calendar-event drift: 10 Backtests
Data last updated
Rules tied to a dated event, such as a funding reset, a holiday or the turn of the month.
0 of 10 strategies finished ahead of the benchmark on the raw number, and 0 had an unadjusted p-value under 0.05. None cleared the correction for trying 1,774 tests in total (a q-value below 0.10), so no result here is a confirmed edge. Scored 2026-10-31, using data up to 2025-03-15.
Strategies here tested on crypto, ETFs, US stocks. See the full family index, the full method and every result, or download the raw registry (JSON).
Results
| Rule | Market | Vs benchmark | Sample | p-value | Corrected q | Receipt |
|---|---|---|---|---|---|---|
| QQQ in even FOMC-cycle weeks, AGG in odd weeks receipt Source: Cieslak, Morse & Vissing-Jorgensen 2019 JF 'Stock returns over the FOMC cycle' (the equity premium is earned in even weeks 0, 2, 4, 6 of the cycle) | ETFs | −2.0 pts/yr | 5,081 rebalances | 1.000 | 1.000 | On-site receipt |
| End-of-quarter window dressing bounce receipt Source: Lakonishok, Shleifer, Thaler, Vishny 1991 J. Finance ('window dressing') | ETFs | −5.6 pts/yr | 141 rebalances | 1.000 | 1.000 | On-site receipt |
| High beta in even FOMC-cycle weeks, low beta in odd weeks (wide US stocks) receipt Source: Cieslak, Morse & Vissing-Jorgensen 2019 JF 'Stock returns over the FOMC cycle'. Cost: about 50 full book switches a year, roughly 5%/yr at 0.05%/side | US stocks | −7.7 pts/yr | 2,136 rebalances | 1.000 | 1.000 | On-site receipt |
| Turn-of-month equity drift (wide US stocks) receipt Source: Lakonishok & Smidt 1988 RFS; McConnell & Xu 2008 FAJ | US stocks | −10.1 pts/yr | 17,570 trades | 1.000 | 1.000 | On-site receipt |
| Turn-of-month equity drift receipt Source: Lakonishok & Smidt 1988 RFS; McConnell & Xu 2008 FAJ | ETFs | −10.4 pts/yr | 243 trades | 1.000 | 1.000 | On-site receipt Full test: Turn-of-the-Month Effect |
| Quadruple-witching-day reversal receipt Source: Stoll & Whaley 1987 JFQA (expiration-day effects); popular retail folklore 2020s | ETFs | −13.5 pts/yr | 79 trades | 1.000 | 1.000 | On-site receipt |
| Lunar New Year Asia-driven crypto seasonality receipt Source: Crypto trade-press/exchange volume commentary (CryptoQuant/Kaiko notes) | crypto | −26.6 pts/yr | 133 trades | 1.000 | 1.000 | On-site receipt |
| Crypto turn-of-month drift receipt Source: Lakonishok & Smidt 1988 RFS turn-of-month, tested here for the crypto-specific absence of the equity mechanism | crypto | −44.8 pts/yr | 1,308 trades | 1.000 | 1.000 | On-site receipt |
| Crypto perpetual funding-reset time drift receipt Source: Exchange/trader community commentary on funding-time microstructure | crypto | −53.5 pts/yr | 10 trades | 1.000 | 1.000 | On-site receipt |
| Crypto weekend low-volume drift receipt Source: Kaiser 2019 Financial Markets and Portfolio Mgmt; Aharon & Qadan 2019 J. Behavioral Finance | crypto | −56.9 pts/yr | 5,612 trades | 1.000 | 1.000 | On-site receipt Full test: Crypto Weekend Effect |
- QQQ in even FOMC-cycle weeks, AGG in odd weeks
ETFs · −2.0 pts/yr · 5,081 rebalances · corrected q 1.000
Source, p-value and result file
- Source
- Cieslak, Morse & Vissing-Jorgensen 2019 JF 'Stock returns over the FOMC cycle' (the equity premium is earned in even weeks 0, 2, 4, 6 of the cycle)
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- End-of-quarter window dressing bounce
ETFs · −5.6 pts/yr · 141 rebalances · corrected q 1.000
Source, p-value and result file
- Source
- Lakonishok, Shleifer, Thaler, Vishny 1991 J. Finance ('window dressing')
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- High beta in even FOMC-cycle weeks, low beta in odd weeks (wide US stocks)
US stocks · −7.7 pts/yr · 2,136 rebalances · corrected q 1.000
Source, p-value and result file
- Source
- Cieslak, Morse & Vissing-Jorgensen 2019 JF 'Stock returns over the FOMC cycle'. Cost: about 50 full book switches a year, roughly 5%/yr at 0.05%/side
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Turn-of-month equity drift (wide US stocks)
US stocks · −10.1 pts/yr · 17,570 trades · corrected q 1.000
Source, p-value and result file
- Source
- Lakonishok & Smidt 1988 RFS; McConnell & Xu 2008 FAJ
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Turn-of-month equity drift
ETFs · −10.4 pts/yr · 243 trades · corrected q 1.000
Full test: Turn-of-the-Month Effect
Source, p-value and result file
- Source
- Lakonishok & Smidt 1988 RFS; McConnell & Xu 2008 FAJ
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Quadruple-witching-day reversal
ETFs · −13.5 pts/yr · 79 trades · corrected q 1.000
Source, p-value and result file
- Source
- Stoll & Whaley 1987 JFQA (expiration-day effects); popular retail folklore 2020s
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Lunar New Year Asia-driven crypto seasonality
crypto · −26.6 pts/yr · 133 trades · corrected q 1.000
Source, p-value and result file
- Source
- Crypto trade-press/exchange volume commentary (CryptoQuant/Kaiko notes)
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Crypto turn-of-month drift
crypto · −44.8 pts/yr · 1,308 trades · corrected q 1.000
Source, p-value and result file
- Source
- Lakonishok & Smidt 1988 RFS turn-of-month, tested here for the crypto-specific absence of the equity mechanism
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Crypto perpetual funding-reset time drift
crypto · −53.5 pts/yr · 10 trades · corrected q 1.000
Source, p-value and result file
- Source
- Exchange/trader community commentary on funding-time microstructure
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
- Crypto weekend low-volume drift
crypto · −56.9 pts/yr · 5,612 trades · corrected q 1.000
Full test: Crypto Weekend Effect
Source, p-value and result file
- Source
- Kaiser 2019 Financial Markets and Portfolio Mgmt; Aharon & Qadan 2019 J. Behavioral Finance
- p-value
- 1.000
- Corrected q
- 1.000
- Receipt
- On-site receipt
How to read these results
"Vs benchmark" is the strategy's annualized return minus an equal-weight benchmark, rebalanced daily, on the same assets and days. Strategies are charged trading costs; the benchmark pays none. Ahead is not a confirmed edge: the p-value is the unadjusted chance of a gap this large by luck, and the corrected q allows for how many strategies were tried, because testing hundreds of ideas throws up lucky winners. Every strategy saw only data up to the same cutoff date. The receipt is the frozen result file for that rule.
"Sample" is trades for rules that enter and exit a position, and rebalances for portfolio rules. A rebalance is a scheduled date on which a portfolio rule re-ranks and held at least one position. It is the sample size for portfolio rules. A holdings change is a rebalance where what the portfolio held, or its weights, differed from the previous one. Only these cost money to trade. The two are different counts and are never swapped. Definitions.