The rules
Rank stocks each month by their return over the past six months. Buy the top tenth (winners) and sell short the bottom tenth (losers), equally weighted. Hold each monthly cohort for six months, so six cohorts overlap at any time (Jegadeesh and Titman 1993, NYSE and AMEX stocks, 1965-1989). A variant skips one week between ranking and holding. The 2001 follow-up adds Nasdaq and drops stocks under $5 and the smallest tenth by NYSE size.
- Universe
- US common stocks via the Market Screener: the 30 with the highest six-month return
- Bars
- daily
- Position size
- 3.2% of equity per stock, at most 30 held and at most two new a day. Published: equal weight within each tenth (decile).
- Rebalance
- Daily check, buying up to two a day when a slot is free; each stock held 126 trading days. Published: monthly ranking, six-month holds.
What the research found
In Jegadeesh and Titman's (1993) six-month strategy on NYSE and AMEX stocks, 1965-1989, before costs, the winner decile earned 1.74% a month and the loser decile 0.79%, a spread of 0.95% (t = 3.07; a t-statistic of about 2 or more is the usual bar for a result unlikely to be chance). Against a market index weighted by company value, the winners' alpha (return beyond what market moves explain) was 0.70% a month (t = 3.24) and the losers' -0.30% (not significant); against an equally weighted index both were significant. After a 0.5% cost per trade, buying winners and selling losers short returned 9.29% a year, risk-adjusted. Their follow-up drops stocks under $5: the working paper (NYSE and AMEX only) found a spread of 1.01% a month in 1990-1997, after the first paper's sample, against 1.11% in 1965-1989; the 2001 article, which adds Nasdaq and drops the smallest NYSE size decile, found 1.39% a month in 1990-1998 against 1.17% in 1965-1989, and 0.88% (t = 2.59) in 1990-1998 among stocks bigger than the median NYSE stock. Both versions found reversals, significant in the fourth and fifth years after the ranking, that by month 60 had undone virtually all of the first year's gain, though far less after 1981. Asness, Frazzini, Israel and Moskowitz (2014) report that the winners in the standard US momentum portfolio beat the market by 5.2% a year over 1991-2013, before costs, and cite a study of AQR's live trades showing momentum survives real trading costs; Lesmond, Schill and Zhou (2004) find instead that standard momentum strategies trade costly stocks so often that trading costs prevent a profit. Novy-Marx (2012) finds ranking on recent performance earns positive but smaller returns than ranking on performance 12 to seven months back, especially among the largest, most liquid stocks. Geczy and Samonov (2016) find momentum profits in 1801-1926 US prices too; their 2013 working paper counts seven negative rolling ten-year periods before 1927, besides the losing ten-year stretches after the Great Depression and in 2002-2012. Daniel and Moskowitz (2016) find momentum crashes cluster after market falls with high volatility. McLean and Pontiff (2016) find published predictors, momentum among them, earn 58% less after publication on average; Jensen, Kelly and Pedersen (2023) find that momentum, as a family of related signals, replicates.
Results
| Measure | Orbital test | SPY, same days |
|---|---|---|
| Window | Mar 1, 2017 to Oct 30, 2026 | Mar 1, 2017 to Oct 30, 2026 |
| Total return | −56.60% | +267.14% |
| CAGR | −8.27% | +14.40% |
| Sharpe | −0.02 | 0.60 |
| Discounted Sharpe | −0.31 | — |
| Max drawdown | −84.5% | −33.7% |
| Win rate | 39.6% | — |
| Profit factor | 0.92 | — |
| Trades | 591 | — |
Hypothetical: a backtest with modeled costs, not traded with real money.
Hypothetical growth of $10,000
R02 Six-Month Stock Momentum · −56.6% vs SPY +267.1%
Rebuild it in Orbital
Paste this into Orbital Chat. The Strategy Developer builds the workflow and asks before it runs any backtest.
On daily bars, use a Market Screener universe of US stocks with the type set to Common stock only, priced at $5 or more, with 20-day average volume of at least 500,000 shares and Volatility 1M (average daily range) of at least 0.75%, sorted by Perf 6M (highest first), top 30. Buy each stock in the screen with 3.2% of equity at the next open, in the screen's sort order, skipping any already held; hold at most 30 positions and open at most 2 new ones a day. Sell each position 126 trading days after entry (a time stop of 126 daily bars), with no other exit, even if it leaves the screen. Long only; uninvested capital stays in cash. Backtest from March 1, 2017 to October 30, 2026 with $100,000.
Is it real?
- Strategies tested together
- 7
- Beta to SPY
- 1.16
- Alpha vs SPY
- −13.5% a year
- Costs paid
- $1,600
- Data
- Daily bars, 2,432 sessions
- Fills
- Next open at market, sized at the fill
Read this before you trust it
Long only: it buys from the 30 stocks with the highest six-month return, a far more concentrated holding than the paper's top tenth of NYSE and AMEX stocks, and skips the short side. The screen ranks every day, with no skip week, but no holding is sold to make room: at two buys a day the book takes at least 15 trading days to fill, then buys only as holdings reach 126 trading days or stop trading, so its buying comes in bursts about six months apart, not a sixth of the portfolio each month as in the paper, which also rebalances to equal weights monthly. It ranks on the price change over the last 126 trading days, with dividends left out; the paper ranks on past six-month returns. The screen cannot select on company size or exchange, so US common stocks priced at $5 or more (the 2001 study's own price floor) with a 500,000-share average volume floor stand in for NYSE and AMEX (1993) and for the 2001 study's size screen. The 0.75% volatility floor is an addition: a takeover target whose deal price jump puts it in the top 30 is kept out once it has traded pinned near the deal price for most of a month, but can still pass in the first weeks after the announcement. Orders fill at the next open with 0.05% slippage (the assumed trading cost).
Sources