Does a trading idea beat a coin flip at the same moment?
Mostly no. Five of six classic ideas pick a side no better than a coin flip at the same bar. One held in both halves of the year: a rejection of yesterday's high or low on gold, by +0.07 to +0.13 R a trade. It is a lead, not proof.
The question
An entry idea does two things: it picks a moment and it picks a side. The moment can matter on its own; the first hour moves more than lunch. To see whether the side adds anything, we compared each idea with a coin flip taken at exactly the same bars, with the same stop and target.
What we ran
- Six ideas, each read on the bar before entry: engulfing candle, VWAP cross, first opening-range break, rejection of yesterday's high or low (a wick through it that closes back inside), RSI(14) crossing 30 or 70, and EMA 9/21 cross.
- NQ, ES, gold, oil and Bitcoin futures, 5 and 15-minute bars, 1 Sept 2025 to 5 Nov 2026.
- Three exits sized on the 14-bar ATR: stop 1 and target 1, stop 1 and target 3, and a trailing stop. Results in R per trade before costs.
- Judged on the first and second half apart.
Results
| Idea | 15-min, first half | 15-min, second half | 5-min, first half | 5-min, second half |
|---|---|---|---|---|
| Engulfing candle | −0.01 | +0.04 | −0.01 | −0.01 |
| VWAP cross | 0.00 | +0.01 | 0.00 | 0.00 |
| Opening-range break | +0.03 | −0.12 | +0.07 | −0.04 |
| Rejection of yesterday's high or low | +0.06 | −0.05 | 0.00 | −0.01 |
| RSI 30/70 cross | −0.11 | −0.04 | −0.07 | +0.04 |
| EMA 9/21 cross | −0.05 | −0.08 | 0.00 | +0.01 |
| Bars and exit | First half | Second half |
|---|---|---|
| 15-min, stop 1 / target 1 | +0.08 | +0.09 |
| 5-min, stop 1 / target 1 | +0.07 | +0.13 |
| 5-min, stop 1 / target 3 | +0.13 | +0.13 |
The moment itself, a coin flip at the idea's bars against a coin flip at any time, was close to zero for every idea. The exception is the opening-range break with wide exits, which is the first-hour effect: the time adds something, the side the idea picks does not.
Gold's rejection was much stronger when it also pointed back toward VWAP: +0.31 and +0.43 R in the two halves on 15-minute bars, against about zero for the VWAP context alone. Oil showed the same pattern, weaker. On NQ and ES the idea worked in the first half and not the second.
Why it might be so
Yesterday's high and low are levels many traders watch. When price pokes through and closes back inside, the breakout traders who joined are wrong, and their exits push price back. Gold and oil tend to snap back during the day more than stock indexes do, which fits a fade working there and not on NQ or ES.
What it means when you test
Before trusting an entry, ask whether its side beats a coin flip at the same moment. For most basic ideas the moment and the context do the work, not the pattern.
Test the idea from this note
Gold: rejection of yesterday's high or low
Build it in Analyze on our market data, add or ignore filters, and see its trades, its month-by-month record and its luck check. Numbers in the app run to today, so they differ a little from this note.
Limits
About 14 months and five markets, before costs. The gold result was picked out of 180 cells, so some of it may be luck; that is why it is a lead to re-check on new months, not proof. The ideas were rebuilt from bars, not the exact app cards.
Simulated results on past market data from 2025 and 2026. Research only, not trading signals or advice. Past results do not promise future ones; see the hypothetical performance disclosure at the bottom of the page.
Get this note as a PDF
Leave your email and the PDF downloads straight away. Easy to keep, print or pass on.
Test your own ideas the same way
Analyze runs any idea on NQ, ES, gold, oil and Bitcoin with costs, both halves and a luck check, so you can see whether it beats chance before you trade it.
Past results on market data, not a forecast. No advice and no signals.
More research
We test the ideas traders talk about
Popular trading ideas tested after costs and against luck, including the ones that failed.
Does a trading filter have a cause, or is it just a statistic?
Eleven market reasons written before the test, scored on five checks per market. One passed four. One came out backwards.
What gamma levels really tell you: how far, not which way
Negative gamma widens the next hour on NQ and ES, high gamma calms it. Direction: no difference. Gold, oil and Bitcoin: nothing usable.
Gold after a sweep of yesterday's high: the money sits on Fed afternoons
Gold shorts after a sweep of yesterday's high made their money on Fed afternoons and in the first hour. A strong reason, too few days to call it proven.
Educational content, not investment advice. Futures trading involves substantial risk of loss. See the risk disclosure.