Fair Value Gap vs Breaker Block
Michael Whitman
·
Trading since 2017, full-time since 2020
·
Published 2026-09-08
·
Updated
·
Reviewed
A fair value gap is the empty strip between the wicks of the first and third bars of a fast three-bar run, marked before price returns. A breaker block is an area that gave way once and later stopped price arriving from the opposite side. A gap that fails and flips is the gap's own breaker-style version.
The breaker block idea rests on one claim: an area that failed and then stopped price from the other side is a stronger area than before. A fair value gap can go through the same cycle, so the claim can be measured on gaps with a rule nobody has to judge. On the gold fund GLD the flipped area was no stronger.
What each one is
The gap is the space two outer wicks leave in a run of three bars. In a bullish run, the first bar’s high never reaches the third bar’s low. The fair value gap page states the rule, applied here without changes.
A breaker block is a level that broke one way and then held when price returned from the far side. The candle usually marked is the one in front of the break. Breaker block covers the version built on order blocks.
A failed gap is the gap’s own breaker. When a bullish gap is closed through to the downside and price later rallies back into it, the band is being tested from below, exactly the breaker situation. Traders often call that an inverse fair value gap.
Where they differ
What has to happen first. A gap needs nothing before it but three candles, whereas a breaker needs a level, a break of it, and a return, so it can only be marked after two events.
What the area is. The gap is the untraded space between two candles, while a breaker is usually a whole candle’s range, wider than the gap and dependent on which candle is chosen.
Which side price comes from. A fresh gap is tested from the side it formed on; either a breaker or an inverse gap is tested from the opposite side instead, after the original idea has failed.
How often each is available. Gaps print far more often than breakers, which need a failure first, and on GLD failure was common: most gaps were closed through within a month.
Which one to use: Fair Value Gap or Breaker Block?
Pick the gap when the area must be fixed the moment it forms, rather than after a judgment about which candle counts.
Use the breaker idea when a level you already marked has failed and you want a written plan for the retest, but give it the same size and stop rules as any other area.
When a gap has been closed through, retire it as a fresh area. Whether you then trade it from the other side should rest on your own record, because on GLD the flip changed nothing measurable.
Did the flipped gap hold better?
The test. Take each gap that still has 20 later sessions in the file and locate the first session trading back into the band from the side it formed on, and record its close: back on the original side (held), inside the band, or beyond the far edge (through). Separately, for every gap that closed beyond its far edge within 20 sessions, find the first session within the next 20 that trades back into the band from the far side, and score that close the same way.
Fresh gaps: 2,113 of 2,464 were re-entered (85.75%). On that first return 854 held (40.42%), 736 closed inside (34.83%) and 523 closed through (24.75%).
Failed gaps: 1,291 of 1,561 were retested from the far side (82.70%). That retest held 514 times (39.81%), closed inside 456 (35.32%) and closed through 321 (24.86%). Held against held, neither side stood out: the two rates are within chance of each other (Fisher exact test, p 0.75).
One gold gap that flipped
The gap. On 31 August 2026 GLD’s low was $404.90. On 2 September its high was $403.33. The band from $403.33 to $404.90 was a bearish gap.
The failure. On 3 September GLD closed at $410.22, above the band, so the bearish gap was closed through on the very next session.
The retest. On 4 September the low of $403.96 came back into the band from above and the close of $406.77 was above it: the flipped band held, a textbook inverse gap. The next session, 8 September, closed at $399.72, below the whole band. One held retest settled nothing.
The original data
The sample. GLD daily bars from Yahoo Finance, 18 November 2004 to 2 October 2026, daily high, low and close only. 2,472 gaps: 1,392 bullish and 1,080 bearish, of which 2,464 had 20 sessions after them.
The headline figure. 1,562 of those 2,464 gaps (63.39%) saw a close beyond the far edge within 20 sessions; retested from the far side, 39.81% held.
The file. Every gap that was closed through, the date of the close through, the first retest from the far side and its close are in the GLD flipped gap file.
What is left out. No order-block breaker is drawn here; the test uses the gap’s own band so that no candle has to be chosen. What price did after the retest was left unmeasured.
When it fails
The failure is treating a failed area as a promoted one. A breaker feels stronger because it carries a story: trapped traders who will defend the old price. On GLD the gap’s failed version held no more often than the fresh one, so the story added confidence without adding results. The cost appears in sizing, where extra conviction becomes extra shares at a level no better than any other.
The second failure is marking the breaker after the retest held. That is hindsight.
A third is retesting forever. A level touched five times is not the same idea.
A fourth is choosing the candle afterwards. Then the band moves to fit.
And a fifth is skipping the stop because the area is “confirmed”. About a quarter closed through.
Related
Fair value gap holds the definition used for every band above. Breaker block explains the failed-level idea in its order-block form. And break of structure covers the break a breaker depends on.
Do not upgrade a level because it failed once and came back. On gold, a gap that was broken and retested from the other side held about as often as a fresh one. Size a breaker-style trade like any other area trade, with the stop written before the retest.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.