Fair Value Gap vs Order Block
A fair value gap is a mechanical three-candle pattern you can scan for. An order block requires a structural break to have happened first and a judgement about which candle preceded it, which makes it harder to define and much harder to test.
Both mark a zone that price left quickly and may return to. They come from the same framework and are frequently used together. The difference that matters is not what they claim — it is that one has a definition a computer can apply and the other does not.
What each one is
A fair value gap is three consecutive candles where the first and third do not overlap. Two comparisons, no judgement, the same answer for everybody who looks. Fair value gap sets out the definition.
An order block is the last opposing candle before a move that broke structure. It requires a structural level to have been marked first, a break of it, and then a decision about which candle counts. Order block covers the full version.
Both belong to the same framework, and both are described as places institutional orders remain. That claim is an interpretation rather than something the chart shows, and it applies equally to each.
Where they differ
Definability. The gap is arithmetic you can run over a decade of data in an afternoon. The block needs a prior level, a break, and a choice, none of which a scan can make for you.
What has to happen first. A gap exists the moment three candles print. A block does not exist until price has broken a level you marked earlier, which makes it a later and rarer signal.
Zone size. The gap is the untouched range between two candles, often narrow. The block is a whole candle’s high to low, usually wider — which changes the stop distance and therefore the position size.
Frequency. Gaps appear constantly on a low timeframe, which is why a size filter is essential. Blocks are rarer because the structural break has to happen too.
Where they agree
Both mark a zone and both are spent once price trades through them. Keeping traded-through zones on the chart fills it within a week either way.
Both need to be marked before price returns. A zone identified after the reaction is a line drawn at a place already known to matter, and neither pattern survives that.
Both require a reaction rather than an arrival. Price reaching a zone is arithmetic; price being rejected there is behaviour, and only the second is evidence.
And both carry the same untestable story. Whether unfilled institutional orders sit at either level is not visible in public price data, for the gap or for the block.
Which one to use
Use the fair value gap when you want something you can test. Write the two comparisons, add a minimum size, run it over your own instrument’s history, and count what happened. That answer exists for the gap and does not exist for the block.
Use the order block when you have already marked a structural break and want the candle that preceded it. In that sequence the judgement is constrained by the level you drew earlier, which is what makes it usable rather than arbitrary.
Use the gap first when you are starting out, because a rule you can verify teaches you more in a month than a rule you cannot. When the gap’s record on your instrument is known, the block becomes a question you can ask against that baseline rather than instead of one.
What neither can do
Neither contains a stop. Both mark an area; where the idea is wrong is a structural question, and that distance sets the size on either pattern.
Neither sees participation. In an illiquid instrument gaps form because nobody was trading between the prices, and a block forms on a move a single order produced.
The original data
Of the 24,971 unique videos in research/search-study-corpus.jsonl, 3 compare these two directly in
the title, at a median of 198,814 views. Separately, fair value gaps appear in 150 titles at a median
of 28,170 and order blocks in 391 at 2,786. The counts come from site/rank_compare.py and
site/corpus_count.py, both deduplicating by video id.
391 videos on the block at 2,786 against 150 on the gap at 28,170. Two and a half times the coverage and a tenth of the audience per video — and the comparison between them outperforms both, at 198,814, which is what happens when people already know the two names and want to know which to use.
The answer to the question on that chart is that a gap inside a block is one zone, not two confirmations. They frequently overlap by construction — the block is the candle before a fast move, and a fast move is what leaves a gap — so treating the overlap as agreement is counting one observation twice.
When it fails
The failure is using both and calling the overlap confirmation, and it is almost invisible because the two really do coincide. A decisive move leaves a gap and, by definition, had a last opposing candle in front of it. Marking both and requiring them to agree feels like a filter and removes almost nothing, because the second one was produced by the same move as the first. The trade is taken on one piece of evidence wearing two names.
The second failure is scanning for blocks. The judgement cannot be automated.
A third is dropping the structural break. Then any candle is a block.
A fourth is no size filter on gaps. They cover the chart within a week.
A fifth is widening either zone after a reaction. Both stop being able to fail.
And a sixth is trading the order-flow story. It is untestable from either pattern.
Related
Fair value gap covers the mechanical pattern. Order block covers the judgement-based one. And smart money concepts is the framework both belong to.
I trust the gap more, and the reason is not that it works better — it is that I can write it in code. Two comparisons on three candles gives the same answer for everybody. Which candle counts as the block is a judgement, and a judgement made after the move is not a rule.
— Michael Whitman
This page is educational, not financial advice. Test every idea on your own charts before risking money.