Tickfloor

Day and month seasonality: 62 Backtests

Data last updated

Rules tied to the day of the week, hour of the day or month of the year.

8 of 62 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

RuleMarketVs benchmarkSamplep-valueCorrected qReceipt
Crypto FOMC-cycle weeks: long all 18 coins in even FOMC-cycle weeks, flat in odd weeks (Cieslak-Morse-Vissing-Jorgensen transfer) receipt
Source: c1023/acad2		crypto+5.3 pts/yr2,700 trades0.4531.000On-site receipt
Treasury auction cycle, both legs: short TLT/IEF for the 5 sessions into each 10-year auction, long for the 5 after (Lou-Yan-Zhang) receipt
Source: c1024/acad3		ETFs+2.0 pts/yr410 trades0.2991.000On-site receipt
Month-end Treasury rebalancing: long TLT/IEF/AGG over the last 3 sessions of each month (Etula et al.) receipt
Source: c1022/acad		ETFs+1.6 pts/yr446 trades0.1181.000On-site receipt
Announcement-day Treasuries: long TLT/IEF only over employment-report and FOMC sessions (Savor-Wilson) receipt
Source: c1023/acad2		ETFs+1.5 pts/yr332 trades0.3011.000On-site receipt
Tax-loss window (15 Dec - 31 Jan) deep-loser tilt, US equities receipt
Source: Tax-loss selling: Branch (1977), Roll (1983), Grinblatt & Moskowitz (2004) December loser reversal		US stocks+1.4 pts/yr2,136 rebalances0.1341.000On-site receipt
MATCHED CONTROL: same deep-loser tilt at a non-tax window (15 Jun - 31 Jul) receipt
Source: Tax-loss selling: Branch (1977), Roll (1983), Grinblatt & Moskowitz (2004) December loser reversal		US stocks+0.7 pts/yr2,136 rebalances0.2341.000On-site receipt
September effect: flat 13 US equity ETFs for the sessions of September, long otherwise receipt
Source: c1024/acad3		ETFs+0.3 pts/yr128 trades0.3981.000On-site receipt
'Sell Rosh Hashanah, buy Yom Kippur': flat 13 US equity ETFs between the two holidays, long otherwise (Frieder-Subrahmanyam) receipt
Source: c1024/acad3		ETFs+0.0 pts/yr128 trades0.4851.000On-site receipt
Daylight-saving Monday: flat QQQ/DIA on the first session after each clock change, long otherwise (Kamstra-Kramer-Levi) receipt
Source: c1023/acad2		ETFs−0.1 pts/yr61 trades1.0001.000On-site receipt
Treasury auction cycle: long TLT/IEF for the 5 sessions after each month's 10-year note auction (Lou-Yan-Zhang) receipt
Source: c1024/acad3		ETFs−0.4 pts/yr206 trades1.0001.000On-site receipt
Halloween beta rotation: highest-beta 10 large caps November to April, lowest-beta 10 May to October, monthly receipt
Source: c1031/stockxs		US stocks−0.4 pts/yr89 rebalances1.0001.000On-site receipt
Post-FOMC drift in Treasuries: hold TLT/IEF in the direction of the FOMC-day move for the next 15 sessions (Brooks-Katz-Lustig) receipt
Source: c1023/acad2		ETFs−0.5 pts/yr134 trades1.0001.000On-site receipt
Seasonal-affective bond demand: long Treasury/aggregate bond ETFs September to November (Kamstra-Kramer-Levi-Wermers) receipt
Source: c1022/acad		ETFs−0.6 pts/yr56 trades1.0001.000On-site receipt
Best Six Months with MACD timing (Hirsch) receipt
Source: c1022/retail		ETFs−2.3 pts/yr118 trades1.0001.000On-site receipt
Sector ETF same-calendar-month seasonality, top 3 receipt
Source: Heston & Sadka 2008 JFE (seasonality in the cross-section, including industries); Keloharju, Linnainmaa & Nyberg 2016 JF 'Return seasonalities'		ETFs−2.4 pts/yr102 rebalances1.0001.000On-site receipt
Calendar filter: long equity ETFs except in July to September (UTC month of the next session) receipt
Source: c1031/acad4		ETFs−3.4 pts/yr128 trades1.0001.000On-site receipt
Halloween effect by sector: cyclicals Nov-Apr, defensives May-Oct (Jacobsen-Visaltanachoti) receipt
Source: c1022/acad		ETFs−4.1 pts/yr81 trades1.0001.000On-site receipt
January barometer per sector: hold each sector/index ETF Feb-Dec only if its own January return was positive receipt
Source: c1022/acad		ETFs−4.6 pts/yr59 trades1.0001.000On-site receipt
Earnings-season market premium: hold equity ETFs from the 15th of Jan/Apr/Jul/Oct to the 14th of the next month (Savor-Wilson) receipt
Source: c1031/acad4		ETFs−5.0 pts/yr482 trades1.0001.000On-site receipt
Post-expiration week weakness: flat QQQ/DIA for the 5 sessions after the third Friday (Stivers-Sun) receipt
Source: c1022/acad		ETFs−5.8 pts/yr346 trades1.0001.000On-site receipt
Announcement-day beta: high-beta on employment-report and FOMC days, low-beta otherwise (Savor-Wilson) receipt
Source: c1022/acad		US stocks−6.5 pts/yr1,873 rebalances1.0001.000On-site receipt
First-half-of-month effect: long QQQ/DIA for the first 9 sessions of each month (Ariel) receipt
Source: c1022/acad		ETFs−6.5 pts/yr346 trades1.0001.000On-site receipt
Sell in May and go away (Halloween effect) receipt
Source: Bouman & Jacobsen 2002 AER; Jacobsen & Zhang 2020 working paper		ETFs−6.6 pts/yr21 trades1.0001.000On-site receipt
Full test: Sell in May
Sell in May and go away (Halloween effect) (wide US stocks) receipt
Source: Bouman & Jacobsen 2002 AER; Jacobsen & Zhang 2020 working paper		US stocks−6.6 pts/yr1,536 trades1.0001.000On-site receipt
Pre-holiday beta rotation: high-beta quintile on the session before each NYSE holiday, low-beta quintile otherwise (Ariel; Savor-Wilson shape) receipt
Source: c1023/acad2		US stocks−7.3 pts/yr1,872 rebalances1.0001.000On-site receipt
Seasonal reversal: long names with the lowest average return in the OTHER calendar months (Keloharju-Linnainmaa-Nyberg) receipt
Source: c1022/acad		US stocks−7.9 pts/yr66 rebalances1.0001.000On-site receipt
Turn-of-month beta rotation: high beta over month-end days, low beta otherwise (Etula et al. 'Dash for cash') receipt
Source: c1022/acad		US stocks−8.5 pts/yr1,873 rebalances1.0001.000On-site receipt
Month-of-year crypto seasonality (Uptober) - rob1 neighbouring months (wide US stocks) receipt
Source: Retail/crypto-media seasonality claims (e.g. CoinGlass monthly return tables)		US stocks−9.1 pts/yr1,535 trades1.0001.000On-site receipt
Autumn effect in gold: long GLD/SLV only during September and November (Baur) receipt
Source: c1022/acad		ETFs−9.5 pts/yr58 trades1.0001.000On-site receipt
Lunar cycle: long QQQ/DIA in the week around the new moon, flat around the full moon (Dichev-Janes) receipt
Source: c1022/acad		ETFs−10.2 pts/yr355 trades1.0001.000On-site receipt
Matched control: turn-of-month drift with NO overlapping FOMC announcement receipt
Source: none published (Tickfloor research-desk test)		ETFs−10.6 pts/yr176 trades1.0001.000On-site receipt
Turn-of-month drift conditional on an overlapping FOMC announcement receipt
Source: Interaction of Lakonishok & Smidt 1988 (TOM) and Lucca & Moench 2015 (pre-FOMC drift)		ETFs−10.9 pts/yr76 trades1.0001.000On-site receipt
Employment-report day premium: long QQQ/DIA into the monthly jobs report only (Savor-Wilson) receipt
Source: c1022/acad		ETFs−12.1 pts/yr338 trades1.0001.000On-site receipt
Monthly options expiry week drift (US equities, QQQ proxy) receipt
Source: Stivers & Sun 2010 J. Banking & Finance; CBOE expiry-week studies		ETFs−12.3 pts/yr242 trades1.0001.000On-site receipt
Santa Claus rally rob1 (last 6 + first 3 days, +25% window) (wide US stocks) receipt
Source: Stock Trader's Almanac (Yale Hirsch)		US stocks−12.5 pts/yr1,535 trades1.0001.000On-site receipt
Friday-to-Monday persistence: take Monday in the direction of Friday's return, 13 equity ETFs (Cross; Abraham-Ikenberry) receipt
Source: c1031/acad4		ETFs−12.8 pts/yr6,064 trades1.0001.000On-site receipt
FOMC announcement drift receipt
Source: Lucca & Moench 2015 J. Finance		ETFs−12.9 pts/yr73 trades1.0001.000On-site receipt
January effect (small-cap outperformance) - QQQ proxy (wide US stocks) receipt
Source: Rozeff & Kinney 1976 J. Financial Economics; Keim 1983 J. Financial Economics		US stocks−13.1 pts/yr1,535 trades1.0001.000On-site receipt
Full test: The January Effect
January effect (small-cap outperformance) - QQQ proxy receipt
Source: Rozeff & Kinney 1976 J. Financial Economics; Keim 1983 J. Financial Economics		ETFs−13.1 pts/yr21 trades1.0001.000On-site receipt
Full test: The January Effect
Santa Claus rally (last 5 + first 2 days of year) receipt
Source: Stock Trader's Almanac (Yale Hirsch)		ETFs−13.3 pts/yr21 trades1.0001.000On-site receipt
Full test: Santa Claus Rally
Pre-holiday drift (off-by-one fixed) (wide US stocks) receipt
Source: Ariel 1990 J. Finance; Lakonishok & Smidt 1988		US stocks−13.3 pts/yr12,284 trades1.0001.000On-site receipt
Late-October oversold entry: buy on the first bar from 28 Oct when 250-bar return is -20% or worse, 8-bar hold receipt
Source: c1031/acad4		US stocks−13.3 pts/yr177 trades1.0001.000On-site receipt
Pre-holiday drift receipt
Source: Ariel 1990 J. Finance; Lakonishok & Smidt 1988		ETFs−13.7 pts/yr73 trades1.0001.000On-site receipt
FOMC announcement drift (wide US stocks) receipt
Source: Lucca & Moench 2015 J. Finance		US stocks−15.0 pts/yr11,598 trades1.0001.000On-site receipt
Month-of-year crypto seasonality (Uptober) receipt
Source: Retail/crypto-media seasonality claims (e.g. CoinGlass monthly return tables)		crypto−15.4 pts/yr15 trades1.0001.000On-site receipt
Monthly options expiry week drift (US equities, QQQ proxy) (wide US stocks) receipt
Source: Stivers & Sun 2010 J. Banking & Finance; CBOE expiry-week studies		US stocks−15.7 pts/yr17,268 trades1.0001.000On-site receipt
Momentum in quarter-ending months: 12-1 winners held only in March, June, September and December (Sias) receipt
Source: c1023/acad2		US stocks−16.7 pts/yr31 rebalances1.0001.000On-site receipt
Same-weekday return seasonality: long the names with the best average return on the coming weekday over the past year (Keloharju et al.) receipt
Source: c1022/acad		US stocks−18.3 pts/yr1,833 rebalances1.0001.000On-site receipt
Day-of-week seasonality (Monday effect) receipt
Source: French 1980 J. Financial Economics; Steeley 2001 replication		ETFs−18.6 pts/yr2,074 trades1.0001.000On-site receipt
Crypto first-half-of-month: long all 18 coins from the month-end close through the 15th, flat in the second half (Ariel transfer; Kaiser) receipt
Source: c1024/acad3		crypto−19.3 pts/yr1,301 trades1.0001.000On-site receipt
Day-of-week seasonality (Monday effect): DIA proxy (wide US stocks) receipt
Source: French 1980 J. Financial Economics; Steeley 2001 replication		US stocks−24.6 pts/yr148,993 trades1.0001.000On-site receipt
Day-of-week lottery tilt: short high-vol/long low-vol into Monday, reverse into Friday (Birru) receipt
Source: c1022/acad		US stocks−26.2 pts/yr834 rebalances1.0001.000On-site receipt
Crypto same-weekday seasonality: each day long the 6 coins with the best average return on the coming weekday over the past year (Keloharju-Linnainmaa-Nyberg transfer) receipt
Source: c1023/acad2		crypto−38.2 pts/yr2,278 rebalances1.0001.000On-site receipt
Post-quarterly-futures-expiry long receipt
Source: c1022/crypto		crypto−38.6 pts/yr425 trades1.0001.000On-site receipt
Crypto on US employment-report day: long all 18 for the UTC session of the jobs report (Corbet et al.) receipt
Source: c1022/acad		crypto−55.2 pts/yr1,294 trades1.0001.000On-site receipt
Crypto pre-holiday drift: long all 18 coins over the UTC day before each NYSE holiday (Ariel transfer; Qadan-Aharon-Eichel) receipt
Source: c1023/acad2		crypto−59.4 pts/yr1,035 trades1.0001.000On-site receipt
Crypto year-end holiday drift: long all 18 from the Dec 24 close to the Jan 2 close (Kaiser) receipt
Source: c1022/acad		crypto−59.7 pts/yr114 trades1.0001.000On-site receipt
CME expiry drag: short crypto for the 3 sessions into the last Friday of each month receipt
Source: c1022/acad		crypto−76.2 pts/yr1,292 trades1.0001.000On-site receipt
Crypto weekday seasonality (long Sundays only) receipt
Source: retail folklore of a crypto 'weekend pump'/Sunday effect (UTC calendar day)		crypto−79.0 pts/yr5,612 trades1.0001.000On-site receipt
Crypto pre-FOMC drift: long the 8 hourly coins from 19:00 UTC the day before an FOMC decision to 18:00 UTC on the decision day (Lucca-Moench; Pyo-Lee) receipt
Source: c1023/acad2		crypto−80.9 pts/yr425 trades1.0001.000On-site receipt
Crypto hour-of-day seasonality receipt
Source: Baur, Dimpfl, Kuck 2018 Finance Research Letters (BTC intraday patterns)		crypto−143.4 pts/yr11,464 trades1.0001.000On-site receipt
Same-hour return seasonality: trade the next hour in the sign of its mean return at the same hour over the last 20 days when t>=1.5, hourly crypto (Heston-Korajczyk-Sadka transfer) receipt
Source: c1031/acad4		crypto−172.6 pts/yr55,644 trades1.0001.000On-site receipt
  • Crypto FOMC-cycle weeks: long all 18 coins in even FOMC-cycle weeks, flat in odd weeks (Cieslak-Morse-Vissing-Jorgensen transfer)

    crypto · +5.3 pts/yr · 2,700 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    0.453
    Corrected q
    1.000
  • Treasury auction cycle, both legs: short TLT/IEF for the 5 sessions into each 10-year auction, long for the 5 after (Lou-Yan-Zhang)

    ETFs · +2.0 pts/yr · 410 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1024/acad3
    p-value
    0.299
    Corrected q
    1.000
  • Month-end Treasury rebalancing: long TLT/IEF/AGG over the last 3 sessions of each month (Etula et al.)

    ETFs · +1.6 pts/yr · 446 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    0.118
    Corrected q
    1.000
  • Announcement-day Treasuries: long TLT/IEF only over employment-report and FOMC sessions (Savor-Wilson)

    ETFs · +1.5 pts/yr · 332 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    0.301
    Corrected q
    1.000
  • Tax-loss window (15 Dec - 31 Jan) deep-loser tilt, US equities

    US stocks · +1.4 pts/yr · 2,136 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    Tax-loss selling: Branch (1977), Roll (1983), Grinblatt & Moskowitz (2004) December loser reversal
    p-value
    0.134
    Corrected q
    1.000
  • MATCHED CONTROL: same deep-loser tilt at a non-tax window (15 Jun - 31 Jul)

    US stocks · +0.7 pts/yr · 2,136 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    Tax-loss selling: Branch (1977), Roll (1983), Grinblatt & Moskowitz (2004) December loser reversal
    p-value
    0.234
    Corrected q
    1.000
  • September effect: flat 13 US equity ETFs for the sessions of September, long otherwise

    ETFs · +0.3 pts/yr · 128 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1024/acad3
    p-value
    0.398
    Corrected q
    1.000
  • 'Sell Rosh Hashanah, buy Yom Kippur': flat 13 US equity ETFs between the two holidays, long otherwise (Frieder-Subrahmanyam)

    ETFs · +0.0 pts/yr · 128 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1024/acad3
    p-value
    0.485
    Corrected q
    1.000
  • Daylight-saving Monday: flat QQQ/DIA on the first session after each clock change, long otherwise (Kamstra-Kramer-Levi)

    ETFs · −0.1 pts/yr · 61 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    1.000
    Corrected q
    1.000
  • Treasury auction cycle: long TLT/IEF for the 5 sessions after each month's 10-year note auction (Lou-Yan-Zhang)

    ETFs · −0.4 pts/yr · 206 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1024/acad3
    p-value
    1.000
    Corrected q
    1.000
  • Halloween beta rotation: highest-beta 10 large caps November to April, lowest-beta 10 May to October, monthly

    US stocks · −0.4 pts/yr · 89 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1031/stockxs
    p-value
    1.000
    Corrected q
    1.000
  • Post-FOMC drift in Treasuries: hold TLT/IEF in the direction of the FOMC-day move for the next 15 sessions (Brooks-Katz-Lustig)

    ETFs · −0.5 pts/yr · 134 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    1.000
    Corrected q
    1.000
  • Seasonal-affective bond demand: long Treasury/aggregate bond ETFs September to November (Kamstra-Kramer-Levi-Wermers)

    ETFs · −0.6 pts/yr · 56 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Best Six Months with MACD timing (Hirsch)

    ETFs · −2.3 pts/yr · 118 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/retail
    p-value
    1.000
    Corrected q
    1.000
  • Sector ETF same-calendar-month seasonality, top 3

    ETFs · −2.4 pts/yr · 102 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    Heston & Sadka 2008 JFE (seasonality in the cross-section, including industries); Keloharju, Linnainmaa & Nyberg 2016 JF 'Return seasonalities'
    p-value
    1.000
    Corrected q
    1.000
  • Calendar filter: long equity ETFs except in July to September (UTC month of the next session)

    ETFs · −3.4 pts/yr · 128 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1031/acad4
    p-value
    1.000
    Corrected q
    1.000
  • Halloween effect by sector: cyclicals Nov-Apr, defensives May-Oct (Jacobsen-Visaltanachoti)

    ETFs · −4.1 pts/yr · 81 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • January barometer per sector: hold each sector/index ETF Feb-Dec only if its own January return was positive

    ETFs · −4.6 pts/yr · 59 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Earnings-season market premium: hold equity ETFs from the 15th of Jan/Apr/Jul/Oct to the 14th of the next month (Savor-Wilson)

    ETFs · −5.0 pts/yr · 482 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1031/acad4
    p-value
    1.000
    Corrected q
    1.000
  • Post-expiration week weakness: flat QQQ/DIA for the 5 sessions after the third Friday (Stivers-Sun)

    ETFs · −5.8 pts/yr · 346 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Announcement-day beta: high-beta on employment-report and FOMC days, low-beta otherwise (Savor-Wilson)

    US stocks · −6.5 pts/yr · 1,873 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • First-half-of-month effect: long QQQ/DIA for the first 9 sessions of each month (Ariel)

    ETFs · −6.5 pts/yr · 346 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Sell in May and go away (Halloween effect)

    ETFs · −6.6 pts/yr · 21 trades · corrected q 1.000

    Full test: Sell in May

    Source, p-value and result file
    Source
    Bouman & Jacobsen 2002 AER; Jacobsen & Zhang 2020 working paper
    p-value
    1.000
    Corrected q
    1.000
  • Sell in May and go away (Halloween effect) (wide US stocks)

    US stocks · −6.6 pts/yr · 1,536 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Bouman & Jacobsen 2002 AER; Jacobsen & Zhang 2020 working paper
    p-value
    1.000
    Corrected q
    1.000
  • Pre-holiday beta rotation: high-beta quintile on the session before each NYSE holiday, low-beta quintile otherwise (Ariel; Savor-Wilson shape)

    US stocks · −7.3 pts/yr · 1,872 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    1.000
    Corrected q
    1.000
  • Seasonal reversal: long names with the lowest average return in the OTHER calendar months (Keloharju-Linnainmaa-Nyberg)

    US stocks · −7.9 pts/yr · 66 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Turn-of-month beta rotation: high beta over month-end days, low beta otherwise (Etula et al. 'Dash for cash')

    US stocks · −8.5 pts/yr · 1,873 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Month-of-year crypto seasonality (Uptober) - rob1 neighbouring months (wide US stocks)

    US stocks · −9.1 pts/yr · 1,535 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Retail/crypto-media seasonality claims (e.g. CoinGlass monthly return tables)
    p-value
    1.000
    Corrected q
    1.000
  • Autumn effect in gold: long GLD/SLV only during September and November (Baur)

    ETFs · −9.5 pts/yr · 58 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Lunar cycle: long QQQ/DIA in the week around the new moon, flat around the full moon (Dichev-Janes)

    ETFs · −10.2 pts/yr · 355 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Matched control: turn-of-month drift with NO overlapping FOMC announcement

    ETFs · −10.6 pts/yr · 176 trades · corrected q 1.000

    Source, p-value and result file
    Source
    none published (Tickfloor research-desk test)
    p-value
    1.000
    Corrected q
    1.000
  • Turn-of-month drift conditional on an overlapping FOMC announcement

    ETFs · −10.9 pts/yr · 76 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Interaction of Lakonishok & Smidt 1988 (TOM) and Lucca & Moench 2015 (pre-FOMC drift)
    p-value
    1.000
    Corrected q
    1.000
  • Employment-report day premium: long QQQ/DIA into the monthly jobs report only (Savor-Wilson)

    ETFs · −12.1 pts/yr · 338 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Monthly options expiry week drift (US equities, QQQ proxy)

    ETFs · −12.3 pts/yr · 242 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Stivers & Sun 2010 J. Banking & Finance; CBOE expiry-week studies
    p-value
    1.000
    Corrected q
    1.000
  • Santa Claus rally rob1 (last 6 + first 3 days, +25% window) (wide US stocks)

    US stocks · −12.5 pts/yr · 1,535 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Stock Trader's Almanac (Yale Hirsch)
    p-value
    1.000
    Corrected q
    1.000
  • Friday-to-Monday persistence: take Monday in the direction of Friday's return, 13 equity ETFs (Cross; Abraham-Ikenberry)

    ETFs · −12.8 pts/yr · 6,064 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1031/acad4
    p-value
    1.000
    Corrected q
    1.000
  • FOMC announcement drift

    ETFs · −12.9 pts/yr · 73 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Lucca & Moench 2015 J. Finance
    p-value
    1.000
    Corrected q
    1.000
  • January effect (small-cap outperformance) - QQQ proxy (wide US stocks)

    US stocks · −13.1 pts/yr · 1,535 trades · corrected q 1.000

    Full test: The January Effect

    Source, p-value and result file
    Source
    Rozeff & Kinney 1976 J. Financial Economics; Keim 1983 J. Financial Economics
    p-value
    1.000
    Corrected q
    1.000
  • January effect (small-cap outperformance) - QQQ proxy

    ETFs · −13.1 pts/yr · 21 trades · corrected q 1.000

    Full test: The January Effect

    Source, p-value and result file
    Source
    Rozeff & Kinney 1976 J. Financial Economics; Keim 1983 J. Financial Economics
    p-value
    1.000
    Corrected q
    1.000
  • Santa Claus rally (last 5 + first 2 days of year)

    ETFs · −13.3 pts/yr · 21 trades · corrected q 1.000

    Full test: Santa Claus Rally

    Source, p-value and result file
    Source
    Stock Trader's Almanac (Yale Hirsch)
    p-value
    1.000
    Corrected q
    1.000
  • Pre-holiday drift (off-by-one fixed) (wide US stocks)

    US stocks · −13.3 pts/yr · 12,284 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Ariel 1990 J. Finance; Lakonishok & Smidt 1988
    p-value
    1.000
    Corrected q
    1.000
  • Late-October oversold entry: buy on the first bar from 28 Oct when 250-bar return is -20% or worse, 8-bar hold

    US stocks · −13.3 pts/yr · 177 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1031/acad4
    p-value
    1.000
    Corrected q
    1.000
  • Pre-holiday drift

    ETFs · −13.7 pts/yr · 73 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Ariel 1990 J. Finance; Lakonishok & Smidt 1988
    p-value
    1.000
    Corrected q
    1.000
  • FOMC announcement drift (wide US stocks)

    US stocks · −15.0 pts/yr · 11,598 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Lucca & Moench 2015 J. Finance
    p-value
    1.000
    Corrected q
    1.000
  • Month-of-year crypto seasonality (Uptober)

    crypto · −15.4 pts/yr · 15 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Retail/crypto-media seasonality claims (e.g. CoinGlass monthly return tables)
    p-value
    1.000
    Corrected q
    1.000
  • Monthly options expiry week drift (US equities, QQQ proxy) (wide US stocks)

    US stocks · −15.7 pts/yr · 17,268 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Stivers & Sun 2010 J. Banking & Finance; CBOE expiry-week studies
    p-value
    1.000
    Corrected q
    1.000
  • Momentum in quarter-ending months: 12-1 winners held only in March, June, September and December (Sias)

    US stocks · −16.7 pts/yr · 31 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    1.000
    Corrected q
    1.000
  • Same-weekday return seasonality: long the names with the best average return on the coming weekday over the past year (Keloharju et al.)

    US stocks · −18.3 pts/yr · 1,833 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Day-of-week seasonality (Monday effect)

    ETFs · −18.6 pts/yr · 2,074 trades · corrected q 1.000

    Source, p-value and result file
    Source
    French 1980 J. Financial Economics; Steeley 2001 replication
    p-value
    1.000
    Corrected q
    1.000
  • Crypto first-half-of-month: long all 18 coins from the month-end close through the 15th, flat in the second half (Ariel transfer; Kaiser)

    crypto · −19.3 pts/yr · 1,301 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1024/acad3
    p-value
    1.000
    Corrected q
    1.000
  • Day-of-week seasonality (Monday effect): DIA proxy (wide US stocks)

    US stocks · −24.6 pts/yr · 148,993 trades · corrected q 1.000

    Source, p-value and result file
    Source
    French 1980 J. Financial Economics; Steeley 2001 replication
    p-value
    1.000
    Corrected q
    1.000
  • Day-of-week lottery tilt: short high-vol/long low-vol into Monday, reverse into Friday (Birru)

    US stocks · −26.2 pts/yr · 834 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Crypto same-weekday seasonality: each day long the 6 coins with the best average return on the coming weekday over the past year (Keloharju-Linnainmaa-Nyberg transfer)

    crypto · −38.2 pts/yr · 2,278 rebalances · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    1.000
    Corrected q
    1.000
  • Post-quarterly-futures-expiry long

    crypto · −38.6 pts/yr · 425 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/crypto
    p-value
    1.000
    Corrected q
    1.000
  • Crypto on US employment-report day: long all 18 for the UTC session of the jobs report (Corbet et al.)

    crypto · −55.2 pts/yr · 1,294 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Crypto pre-holiday drift: long all 18 coins over the UTC day before each NYSE holiday (Ariel transfer; Qadan-Aharon-Eichel)

    crypto · −59.4 pts/yr · 1,035 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    1.000
    Corrected q
    1.000
  • Crypto year-end holiday drift: long all 18 from the Dec 24 close to the Jan 2 close (Kaiser)

    crypto · −59.7 pts/yr · 114 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • CME expiry drag: short crypto for the 3 sessions into the last Friday of each month

    crypto · −76.2 pts/yr · 1,292 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1022/acad
    p-value
    1.000
    Corrected q
    1.000
  • Crypto weekday seasonality (long Sundays only)

    crypto · −79.0 pts/yr · 5,612 trades · corrected q 1.000

    Source, p-value and result file
    Source
    retail folklore of a crypto 'weekend pump'/Sunday effect (UTC calendar day)
    p-value
    1.000
    Corrected q
    1.000
  • Crypto pre-FOMC drift: long the 8 hourly coins from 19:00 UTC the day before an FOMC decision to 18:00 UTC on the decision day (Lucca-Moench; Pyo-Lee)

    crypto · −80.9 pts/yr · 425 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1023/acad2
    p-value
    1.000
    Corrected q
    1.000
  • Crypto hour-of-day seasonality

    crypto · −143.4 pts/yr · 11,464 trades · corrected q 1.000

    Source, p-value and result file
    Source
    Baur, Dimpfl, Kuck 2018 Finance Research Letters (BTC intraday patterns)
    p-value
    1.000
    Corrected q
    1.000
  • Same-hour return seasonality: trade the next hour in the sign of its mean return at the same hour over the last 20 days when t>=1.5, hourly crypto (Heston-Korajczyk-Sadka transfer)

    crypto · −172.6 pts/yr · 55,644 trades · corrected q 1.000

    Source, p-value and result file
    Source
    c1031/acad4
    p-value
    1.000
    Corrected q
    1.000
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.

Related families