Wyckoff sign-of-strength entries sit on the chance line

15,538 completed Wyckoff structures: sign-of-strength entries hit their targets at the rate 1/(1+k) predicts at every distance, and matched placebo entries did the same.

The claim

Wyckoff accumulation and distribution (climax, automatic rally, secondary test, spring, sign of strength) is taught as a way to catch the start of a markup or markdown, with volume confirming the effort behind the move.

How we tested it

We detected 15,538 completed structures, from climax to sign of strength, on instruments including NQ, the S&P 500, gold, Bitcoin, Ether and Solana, on timeframes from 1 minute to 4 hours with two to five years of data each. Entry was the sign-of-strength close, with the stop behind the spring or range low. Hit rates were compared with 1/(1+k) and with 77,690 placebo entries matched on New York hour, volatility and stop width.

What we found

At 1R the hit rate was 49.56% (chance 50.00%); at 1.5R, 39.90% (40.00%); at 2R, 33.24% (33.33%); at 3R, 24.43% (25.00%). No gap was significant (all |z| ≤ 1.6). The placebo landed within 0.5 points of Wyckoff everywhere; Wyckoff sat 0.5 points above it at all four distances, with z between 1.0 and 1.6, which counts as one correlated observation rather than four.

0 of 34 instrument and timeframe cells survived a multiple-testing correction, and both halves of a date split were null. Using real tick volume on NQ instead of a range-based proxy did not improve the result, so the effort-versus-result reading added nothing here. The only subgroup that came close to significance, high-effort springs, pointed the wrong way.

The classic target, the range height projected from the breakout, sat behind the entry on 22.8% of trades. Measured properly, its median distance was only 0.76R, and it was hit 54.92% of the time against a chance line of 54.89% (averaged over each trade's own distance).

What to do with it

A Wyckoff structure can be a useful way to describe a range and place a stop, but in this test it did not change the odds of reaching a target. Price the trade with 1/(1+k), and check where a projected target really sits before relying on it.

Limits

Structures came from our own rule-based detector, and the study found and fixed two bugs in an early version of it. A discretionary Wyckoff read and trading costs were not part of these hit rates.

Education only, not financial advice. These are measurements of past data, not a forecast or a signal.