Gaps
This strategy buys stocks the morning after a genuine overnight surprise: a gap up of ≥10% at the open on ≥3× normal volume, closing in the upper half of the day's range — and only when both the stock and the S&P 500 are already above their 200-day moving averages.
Entry: at that day's close, sized at 16.7% of equity per position (max 6 concurrent; same-day ties fill largest-gap-first).
Manage: a hard stop 20% below entry; an early cut at bar 10 if the trade is down >10% and never traded >3% above entry; a pyramid add-on at bar 30 (2× size) if the trade is already up ≥10%. (These are two different 10% triggers — bar 10's is a downside cut, bar 30's is an upside add-on.)
Exit: otherwise, time-based — close of bar 80 (roughly 4 calendar months).
The edge is post-event drift: the market under-prices a genuine overnight news event, and the same size of move without a gap has no edge at all. This makes it a rare signal by design — roughly 9 trades per year across the whole universe — and every mechanism tried to cap or protect the (large, late-arriving) right tail made results worse.
Gaps — chart
Signals
All historical signals, newest first — click a row to load that ticker into the chart above
| Ticker | Status | Entry Date | Gap % | RVol | Return % | Exit Date | Exit Reason | Bars Held / Remaining |
|---|---|---|---|---|---|---|---|---|
| Loading… | ||||||||
Backtest Results
Price-return, unlevered, fill-at-close, no fees/slippage
| CAGR | Max Drawdown | MAR | Sharpe | Volatility | Total Return | |
|---|---|---|---|---|---|---|
| Loading… | ||||||
Trade Statistics
Over trades actually taken (the signals table above includes skipped signals too)
Loading…
| # Trades | Trades / Yr | Mean | Median | Hit Rate | Profit Factor | Avg Win | Avg Loss | % Stopped Out | Avg Bars Held | Best | Worst |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Loading… | |||||||||||
Signals by year