Breaker Blocks in Futures: When a Failed Order Block Flips Sides

Order blocks fail. Price returns to a zone you marked as support, closes through it, and the usual reaction is to hold on and hope. A breaker block is ICT's way of using that failure instead of ignoring it. The zone that was supposed to hold gets reclassified, and the opposite side is the one to watch.

What a breaker block is

A bullish order block is a zone you expected to act as support. If price closes below it, ICT calls the zone a bearish breaker block, meaning it's now treated as resistance on a rally back to it. A bearish order block that price closes above becomes a bullish breaker block and is treated as support on a pullback. The reasoning is positional: the buyers or sellers who built the original zone were overwhelmed, so when price returns, those trapped traders may use the level to get out. That explanation is a hypothesis about who holds positions. What you can observe and rule on is the zone, the close-through and the retest. Teachers also define the breaker differently. Some require a sweep of a prior swing and a shift in structure before the failed block qualifies, so pick one definition and apply it the same way every time.

Marking one on ES

1. Mark the order block: the last opposing candle or two before an impulsive move. Record the zone's high and low.
2. Wait for a failure. Price returns and closes through the far edge of the zone by your buffer. The source material used 3 points on ES. Treat that as a starting number to test, and scale it to volatility, since the same dollar move is a different number of points on NQ ($20 per point) than on ES ($50 per point).
3. Reclassify the zone, then wait for a retest. The trade is the pullback into the breaker, not the break itself.

A hypothetical bullish example

Say a bearish order block sits at 5,010 to 5,015 on the 15-minute ES chart. A 15-minute bar closes at 5,019, above the top of the zone. The zone is now a bullish breaker. Later, price pulls back and trades into the top of the zone at 5,014.

With about $29 of friction on ES (a $4 round-turn commission plus a tick each way), the reward nets to $771 and the realistic loss is closer to $379. That's roughly 2 to 1 after costs. The stop sits below the zone because a close through it means the reclassification was wrong. A stop order can slip through fast markets, so a worse fill than $350 is possible. The bearish version mirrors this: rally back into a former bullish block that failed, stop above the zone high.

How it fails

• The zone is wide. A 15 point zone makes the stop wide too, and the reward shrinks relative to risk.
• Price slices through the breaker with no reaction. Then the original close-through was just the start of a move, and the stop does the work.
• The retest never comes. Price leaves and you either chase it or skip. Chasing changes the risk math above.
• The break was a news spike, not a structural move. A wick through the zone that closes back inside isn't a failure of the block, so the close rule matters.

Breaker block or support that became resistance

Classic technical analysis describes the same behavior as support turning into resistance. The practical difference is that a breaker block is a zone rather than a line, which gives a defined entry area and a stop placed beyond the zone instead of a tight stop beside a single price. That's a benefit for stop placement, not evidence that breakers outperform plain level flips. I haven't been given any data showing that, and I wouldn't claim it.

KPL levels and breakers

KPL levels are fixed reference prices anchored to the regular session opening print, with four levels on the standard ES and NQ map. They don't flip polarity or move intraday, so there's no such thing as a KPL level becoming resistance. What you can do is note whether a breaker zone sits near a KPL level and record it as a tag. Whether that overlap helps is a question for your own log, not an assumption. The KPL strategy page covers the construction.

Testing it

Log every breaker you mark, including the ones you skip: zone size, buffer used, whether a retest occurred, result in points and net dollars. A small sample proves little. 30 trades is not a statistically proven edge, 50 SIM trades is not live proof, and a simulated fill doesn't validate queue position or slippage on the retest.