Fade Trading: 1,254 SPY Gap Ups, and How Often They Faded
Fade trading is taking a position against a move that has just happened, such as shorting a stock that opened sharply higher, on the view that the move overshot and will partly reverse. It is a short-term form of mean reversion, and its success depends on how often the move actually snaps back.
Fade trading means taking the other side of a move that just happened: selling a stock that opened sharply higher, or buying one that dropped hard. The trader is betting that the move went too far and will give some of it back. This page defines the fade on SPY’s opening gap, shows one that reversed and one that kept going, and counts how often SPY’s gaps faded from 1993 to September 2026.
How it forms
A sharp move comes first. The classic case is the open: news, earnings or overseas trading push the first print well above or below the prior close. A fade can also follow a big intraday run or a big day.
The fader takes the opposite side. Short the gap up, buy the gap down. The reasoning is that early orders overshoot and that, once they are filled, price drifts back toward where it was. It is the shortest-timeframe version of mean reversion.
The follower takes the same event the other way. Buying a gap up that keeps going is the gap and go trade. Fade and follow are opposite answers to one question, so the only honest way to choose between them is to count which one the market has been paying.
A fade has no natural stop. The move being faded is the most recent information in the market. If it keeps going, there is no earlier level inside the gap to lean against, which is why fades are usually taken small.
The definitions used here
The gap. The official open divided by the official prior close, minus one. A gap up of 0.5% or more on SPY happened on 1,254 of 8,471 sessions from 1 February 1993 to 25 September 2026.
Faded. For a gap up, the close finished below the open, so a short at the opening price was ahead at the close. For a gap down, the close finished above the open.
Filled. The session traded back to the prior close at some point. Closed through. The close finished on the far side of the prior close, erasing the whole gap and more.
The base rate. Across all 8,471 sessions, SPY closed below its open on 3,955, 46.7%, and above it on 4,431, 52.3%. A fade only has an edge if the gapped days beat those numbers.
A worked example
Wednesday 5 August 2026, SPY on 5-minute bars. The prior close was $771.33. SPY opened at $775.85, $4.52 higher, a gap up of 0.59%. The 9:35 bar made the day’s high of $776.85, $1.00 above the open.
The gap filled before lunch. The 11:15 bar traded down to $771.00, below the prior close. SPY closed at $769.79, $1.54 under the prior close and $6.06 under the open, near its low of $769.51.
The arithmetic of the fade. A short at the $775.85 open was $1.00 against the trader within ten minutes, then ahead by $6.06 per share at the close. On 100 shares that is a move of $606 in the trader’s favor, before costs, after first sitting $100 behind.
The original data
Gap ups faded less often than ordinary days. Of the 1,254 SPY gap ups of 0.5% or more, 541 closed below their open, 43.1%, against 46.7% for all sessions. The two-sided binomial test gives p = 0.012, so the gap-up days were more likely than usual to close at or above the open. The median gap up of 0.5% or more gained a further 0.13% from open to close. Every session is in the SPY gap fades file.
Fills and full reversals. 539 of the 1,254, 43.0%, traded back to the prior close the same day. 229, 18.3%, closed below it.
Bigger gaps faded less. Closing below the open: 48.2% of the 1,337 gap ups from 0.25% to 0.5%, 44.6% of the 904 from 0.5% to 1%, 39.2% of the 283 from 1% to 2% and 40.3% of the 67 at 2% or more. Same-day fills fell from 60.2% to 47.8%, 30.4% and 31.3%.
Gap downs matched the base rate. Of the 1,049 gap downs of 0.5% or more, 536 closed above their open, 51.1%, against 52.3% for all sessions (p = 0.44). 459, 43.8%, filled the same day, and 201, 19.2%, closed above the prior close.
How the fade changed after 2010
Gap ups ran further in the later half. From 1993 to 2009, 291 of 601 gap ups of 0.5% or more closed below the open, 48.4%, and 54.1% filled. From 2010 to 25 September 2026, 250 of 653 did, 38.3%, and 32.8% filled. The median open-to-close move on those days went from +0.03% to +0.20%.
Gap downs barely moved. 50.0% faded in the first period and 52.1% in the second, while same-day fills fell from 52.8% to 35.6%.
Hourly timing, recent years. SPY’s 723 full hourly sessions from 27 October 2023 held 111 gap ups of 0.5% or more, and 32 of them filled the same day. 9 filled in the first hour and 23 later, spread across the day.
Fading a big day instead of a gap
The next-day version. Some traders fade the day after a big move rather than the open. After SPY closed up 2% or more (273 days), the next session closed lower 130 times, 47.6%, against 45.7% for all days (p = 0.544). After SPY closed down 2% or more (340 days), the next session closed higher 194 times, 57.1%, against 53.6% (p = 0.211).
Both are within chance. Neither result is far enough from the everyday rate to separate from noise on these counts.
When it fails
The gap keeps going. On 21 September 2026 SPY opened at $766.25, 0.60% above the prior close of $761.69. The low of the day was $766.03, 22 cents under the open, and SPY closed at $773.50, $7.25 above the open. A short at the open was behind for the whole session.
The big gaps are the worst fades. The gaps that look most overdone, 1% and more, closed below their open about four times in ten on SPY, and fewer than a third filled the same day.
The regime changes under you. A fade rule tested on 1993 to 2009 would have met a market that faded gap ups about one time in two. Since 2010 it has been closer to four in ten.
One fund, daily bars. These counts are SPY only. A single stock gapping on its own news is a different event, and the fade rate there has to be counted separately.
Related
The gap fill page explains why a fill rate needs a definition and a deadline before it means anything. Mean reversion covers the broader idea that prices return toward an average, of which the fade is the shortest-term case. And the opening gap page covers why the first print lands away from the prior close.
Count the fade on the instrument you trade before you take one. When I look at a fade, the first question is how often that kind of move reversed on that market in recent years, and the second is where I am wrong if it keeps going.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.