75.5% of Breakouts Fail. We Counted All 1,611GoldOANDA:XAUUSDBreakoutAlertsAsk how often breakouts fail and you'll hear 70%, 80%, "most" — confident numbers, never a dataset. So we counted ours. Every signal our engine fires is tracked from entry to exit — entry, stop and target fixed at fire time, no hindsight. We pulled every completed breakout-family signal from Feb 24 to Aug 7, 2026: 1,611 signals across 15 instruments (FX majors and crosses, gold, silver, oil, US indices). THE HEADLINE NUMBERS • 24.5% hit their target • 66.7% reversed all the way to the stop • 8.8% expired at our 72-hour evaluation deadline So 75.5% failed to follow through — and that's WITH momentum and level-quality filters already applied. Naked level breaks in the wild fail more often. Here's the part the campfire numbers always leave out: the same sample netted +337R. • Average loser: −0.97R • Average winner: +3.04R • Breakeven win rate at that ratio: ~24% • Actual win rate: 29.4% That five-point gap, compounded across 1,611 signals, is the entire economics of breakout trading. The failure rate is the price of admission. Expectancy is the product. WHAT ELSE THE DATA SAID 1. Failed breakouts die fast. Median time to stop-out: 1.8 hours. Median time to target: 6 hours. A break that's going to work starts working — one that loiters at the entry is usually telling you which column it's headed for. 2. Failure rate doesn't rank sessions — expectancy does. New York had the most failures (80.2%) AND a negative average (−0.11R per signal). Asia and the London morning failed about 78% of the time and were still the most profitable windows, because their winners ran furthest. 3. Timeframes fail alike, pay differently. M15: 74.9% fail, roughly 0R average. H1: 77.5% fail, +0.63R. H4: 69.9% fail, +0.31R. Faster charts don't fail much more often — they just pay less when they work. 4. Silver is the fakeout king: 91.1% of XAG/USD breakouts failed. Fewer than one in eleven reached target. Meanwhile the Dow failed 76.9% of the time and STILL averaged +0.95R per signal — rare winners, enormous payoffs. 5. The +1R trap. At these win rates, taking profit at +1R flips the math from +0.21R per signal to roughly −0.39R. Same signals, same failure rate — now a losing system. Letting winners reach their targets isn't a style choice; it's arithmetic. HOW TO USE IT Size every position so a full −1R stop-out is boring, because two of every three entries will come back through your level. Fire your breakouts where the winners run — sessions and timeframes above. And be suspicious of anyone selling win rate: at 3-to-1 payoffs, winning 29% of the time is a business. One firm's tracked data, one regime, filtered signals — your numbers will differ. But it's a real count, not folklore. We publish every outcome weekly, winners and losers alike. Not financial advice. Trading involves risk.