SPX returned to its opening price 65% of the time by noon. Levels picked later in the morning did just as well.

Backtested on SPX one-minute data (79 sessions, June to September 2026), SPX second-by-second data (41 sessions), and SPY one-minute data (1,044 sessions, 2022 to 2026). Updated 5 Oct 2026.

Short answer: the open isn't a magnet. SPX moved at least 10 points away from its open every single day I tested, and came back to it by noon on 65% of days. That sounds like a trade. But over the same length of time, SPX came back to its 10:00 price 62% of the time and its 10:30 price 68% of the time. Price wanders back to most recent levels about that often. There's nothing special about the open.

I tested five ways of fading a move back to the open: filtering by speed, waiting for a pullback, picking a fixed time, waiting for the move to slow down, and trading what happens once price gets back. None did better than chance.

79 of 79days SPX moved 10 points from the open

65%came back to the open by noon

62% / 68%came back to fake "opens" at 10:00 and 10:30

31%came back after running 25 points or more

What is fading back to the open?

A common day-trading belief is that the opening price acts like a magnet. When the market runs away from the open in the first hour, traders sell the rally or buy the drop, targeting a return to the open. It's taught as "fair value", "the open as support and resistance", or simply "the open gets retested".

The question isn't whether price comes back to the open. It often does. The question is whether it comes back more often than it would to any other level, and whether fading the move makes money after you account for the days it doesn't come back.

The exact rules I tested

The "open" is the SPX price at the end of the first minute, because the index's printed opening price is often stale. Using the printed open gave nearly the same numbers. Results are in index points, before trading costs.

The base rate: how often SPX, after moving 10 points from the open, comes back to it by noon and by the close. The control is the same measurement from fake opens, the 10:00 and 10:30 prices, over the same length of time.
Speed filter: sell (or buy) when SPX is 10 points from the open, target the open, stop at 20 points. Do fast moves (10 points in 15 minutes or less) come back more often than slow ones?
Pullback entry: after a run of 10, 15 or 20 points, wait for price to give back 3 or 5 points, then fade it toward the open. Close by noon. With a stop at the new extreme, a 10-point stop, or none.
Fixed-time fade: at 09:49, if SPX is 16 points or more from the open, fade it. 49 combinations of target and stop, close by noon. (09:49 was chosen because it was the median time of the morning's turning point.)
Slowdown detectors: after a 16-point run, fade when the move stalls: no new extreme for 2, 3 or 5 minutes, a narrowing range, or a rejection candle. 150 combinations.
Back at the open: once price has run away and returned to the open, does it bounce off the open or go through it?

Results

How often SPX came back to the open by noon, by how far it had run first:

Ran 10+ points65%

Ran 15+ points53%

Ran 20+ points39%

Ran 25+ points31%

Ran 30+ points22%

The line marks 50%.

The base rate: 65% back by noon, 75% by the close. Fake opens at 10:00 and 10:30 came back 62% and 68% of the time. The morning's turning point was usually early, with a median of 09:48, and the median run before it was 16 points.
Speed filter: fast moves did come back more often (about 80% against 47%). But fast moves from the fake opens came back just as often, so it's a sign of a volatile day, not of the open. As a trade, speed didn't separate winners from losers, and it flipped from month to month.
Pullback entry: the pullback was usually not the start of the return. Only 11% to 24% of trades reached the open before price made a new extreme. Without a stop, they reached the open exactly as often as with no pullback at all: the trigger added no information.
Fixed-time fade: the best of the 49 combinations made 4.7 points a trade, but trading in a random direction produced a result that good 24% of the time. Moving the entry a few minutes either way erased it, and 13 of its 18 trades came in June and July.
Slowdown detectors: none found the turning point. The move had already peaked 15% to 24% of the time when they fired, against 20% for a random entry at the same delay. The best of 150 combinations was worth nothing after accounting for the search, and it faded month by month.
Back at the open: price went through the open about as often as it bounced, on SPX and on 1,044 days of SPY, whatever the size of the earlier run.

Why it failed

Prices revisit recent levels all the time. In a two-to-three-hour window, SPX comes back to most prices it has recently traded. The open is just the first of them.
The days that don't come back are the big ones. A fade wins small and often on choppy days, and loses big on trend days. The further the run, the less likely the return: after 25 points, only 31%.
Turning points are only obvious afterwards. The median turn at 09:48 is a hindsight number. Every rule that tried to spot the turn as it happened did no better than chance.
The open's price is unreliable. SPX's printed open often lags the real market for the first minute, which makes many "back to the open" charts look cleaner than the trade would have been.

Did I test it wrong?

Quite possibly. I tested the version of Fading back to the open described above, with settings I thought were fair, and it didn't work. That doesn't prove every version fails. I may have picked the wrong settings, the wrong market or timeframe, or missed a filter that experienced traders take for granted. Many strategies also depend on judgement that a set of fixed rules can't capture.

If you trade Fading back to the open and it works for you, I'd genuinely like to hear how. Tell me the chart and timeframe, what counts as a setup, any filters, where you get in, where your stop and target go, and when you get out. I'll test your version on the same data, the same way, and publish the result here, whichever way it comes out.

Send me your version

What would change my mind

Evidence that price returns to the open more often than to other levels picked the same way, over a long period and across markets, by enough to pay for the days it doesn't. That comparison against fake levels is the one that matters, and the one this idea fails.

Questions

Does the market always come back to the opening price?

Often, but not because of the open. SPX came back to its open by noon on 65% of days. Over the same length of time, it came back to its 10:00 price on 62% of days and its 10:30 price on 68%.

Is fading a move back to the open profitable?

Not in my tests. The further price ran, the less often it came back: 53% after 15 points, 31% after 25. The runs that kept going cost more than the returns paid.

Does a fast move away from the open mean it will come back?

Fast moves did return more often, but so did fast moves from fake levels. It's a sign of a volatile day, not a pull toward the open, and it didn't make the fade profitable.

When price gets back to the open, does it reverse or keep going?

It's a coin flip. On nine years of SPY and on SPX, price went through the open about as often as it bounced.

How it was tested

Data: SPX one-minute data, 5 June to 29 September 2026 (79 sessions); SPX second-by-second data, 31 July to 28 September 2026 (41 sessions); SPY one-minute data, July 2022 to October 2026 (1,044 sessions), checked against a second source.
Checks: fake opens at 10:00 and 10:30, random-direction comparisons for every parameter search, and month-by-month splits. One-minute and second-by-second results were compared where short stops made the order of highs and lows matter.
Backtest, not live trading: these results come from historical prices with trading costs included. Backtests usually flatter a strategy, so a failed backtest is a strong "no".

This page records my own testing. It isn't financial advice.

Disclosures · Privacy · [email protected]