Triple Top — Is the Third Test of Resistance a Sure Short?

Three peaks at the same height. In the textbook narrative, the market tried three times to break resistance and got knocked back three times — proof of supply's strength and a short signal "stronger than a double top, because there were more tests." It sounds logical. Intuition says: the more rejections, the surer the decline.

The trouble is, the measurement says something else. In Thomas Bulkowski's database (1,964 cases on US stocks), the triple top lands in the bottom half of the pack: 24th out of 36 bearish patterns, with a measly average decline of 14% and a 25% failure rate. For comparison, double tops — the "weaker" pattern with one fewer test — deliver an average 16-19%. The third test didn't add strength. It subtracted it. This article is the autopsy of the confirmation myth — and a guide to trading a pattern that's weaker than it looks.

How to Recognize a Triple Top

Bulkowski's identification criteria:

1. An uptrend before the pattern. The pattern reverses a rise — without a prior move up, there's nothing to reverse. A practical note from the measurements: if the approach into the pattern was short, the decline after it tends to be short too.
2. Three clear peaks at a similar price. The peaks should look comparable; the middle one is sometimes marginally lower. If the middle one is clearly HIGHER, that's not a triple top — it's a head and shoulders, a different pattern with different (and better) statistics.
3. Volume fades across the pattern (62% of cases), with local spikes under each peak.
4. Confirmation: a close below the pattern's lowest trough. A hard requirement. Three peaks without a break of the troughs is just a range, not a pattern — and it can break either way.

A bit of context: triple tops show up more often in bear markets than in bull markets, which fits their character as a "tired market" pattern rather than a dramatic climax. Telling it apart from its neighbors: the triple bottom is its bullish mirror image (covered separately — and its statistics look completely different); a descending broadening formation can imitate a triple top, so check whether the troughs between the peaks sit at one level or keep getting lower.

What the Numbers Say (Bulkowski, Encyclopedia of Chart Patterns)

The full scorecard for confirmed triple tops in a bull market (n = 1,964 "perfect trades" — results measured from breakout to the extreme, i.e., the upper bound of what the pattern could actually deliver):

Let's translate that into plain terms. One in four confirmed patterns doesn't even deliver a 5% decline. The average reach — 14% — sits in the bottom part of the catalog, and that's measured on ideal trades, with no slippage or emotion. The full measure-rule target is reached by fewer than half of cases (49%), so the classic "pattern height projected downward" is usually too optimistic. And a pullback in 66% of cases means two-thirds of breakouts first return to the break line, shaking out anyone who entered right on the break with a tight stop.

The most interesting comparison is with its relatives. Double top: an average decline of 16-19% depending on the variant. Head and shoulders: about 22%, with a much lower failure rate. The triple top, the pattern in this family with the MOST resistance tests, performs the worst of the three. The intuition that "more tests equals a stronger signal" finds no support in the data — it's exactly backward.

The standard caveat: these measurements come from US stocks, the daily timeframe, and are mostly bull-market data. On crypto, forex, or lower timeframes the distributions could look different — nobody has measured them to this standard, so don't pretend (and don't let anyone else pretend) that we know.

How to Trade a Triple Top

Rule zero: don't short the third touch. Entering short when price touches resistance for the third time is a bet that the range breaks down — and the pattern hasn't said that yet. Wait for a close below the lowest trough. Yes, you'll get a worse entry — sometimes by several percent, since troughs can sit well below the peaks. But before confirmation you're not trading the pattern, you're guessing the direction of a range, and ranges with a heavily tested resistance have an ugly habit of breaking upward. If price reverses after the third peak and breaks resistance instead, there simply was no pattern — and there was no loss for you either.

Target: modest, weighted by 49%. Bulkowski's realistic version of the measure rule: pattern height × 49%, subtracted from the breakout level. Example: peaks at 110, lowest trough at 100 → height 10 → realistic target about 95.1, not 90. Want to aim for the full height? Remember only a minority of cases get there.

Stop and pullback. Stop above the pattern's last peak (or, more aggressively, above the last local high before the breakout). Since 66% of breakouts return to the break line, you have two sensible options: enter right away with a wider stop and ride out the return, or wait for the pullback and enter as it fizzles out — statistically, a pullback also tends to worsen the pattern's subsequent performance, so its occurrence is itself a piece of information about the move's weakness.

Filters that improve the odds (all from Bulkowski's measurements): a third peak BELOW the middle one — a failed attempt at a new high signals a stronger decline ahead; a fast, steep rally leading into the pattern — there's something to fall from; the pattern sitting in the bottom third of the yearly price range; a declining volume trend across the pattern's formation (the difference is small but real). And the mirror-image warnings: a shallow approach, a pattern sitting right above a yearly support low, or a third peak level with the previous ones all lower the odds.

Plan B: the busted triple top. Since one in four downside breakouts is a dud, have a plan for failure. If, after confirmation, price drops only a little, reverses, and CLOSES above the pattern's peaks, the short is dead — and the setup's meaning flips: bear stops and breakout-trader orders sat above the triple top, so a break can fire price higher through short-squeeze mechanics. Bulkowski measured the equivalent effect directly for double tops (a busted double top later rises an average of 38%); the same liquidity logic applies to triple tops. Bare minimum discipline: don't add to a losing short "because the pattern was pretty" — busted bearish patterns can be better long signals than the original short call.

The Confirmation Myth — Why a Third Test Does NOT Strengthen the Signal

Where does the gap between intuition and measurement come from? The mechanics suggest two things. First, every additional test of resistance also consumes supply: sellers at that level gradually run out, so a level tested three times tends to be closer to breaking, not further from it — which is exactly why heavily tested ranges scarily often break UPWARD (the same mechanism that drives the ascending triangle). Second, by the time a triple top confirms, price has to travel all the way down to the lowest trough — a large chunk of the downside move has already happened by the time you enter. The signal arrives late, with little range left: hence the meager 14%.

There's a third, broader lesson here: matching names don't mean matching results. The triple bottom, this pattern's mirror twin, has a failure rate of 13%, an average rise of about 46%, and ranks 12/39. Same geometry, opposite direction, a chasm in performance (partly from the bullish bias of the stock market, but the scale of the gap survives that adjustment). The takeaway, no hype attached: you can't assume that because a bullish pattern works, its mirror image works the same way. Every direction needs its own measurement — and this particular short, however much the textbooks love it, ranks among the weaker plays in the catalog.

FAQ

Is a triple top stronger than a double top? No. Bulkowski's measurements: triple top — average decline 14%, 25% failures, rank 24/36; double tops — 16-19% decline with fewer failures. A third resistance test statistically weakens the signal, partly because it consumes supply and delays confirmation.

When is the pattern confirmed? On a candle close below the lowest trough between the peaks. Before that, it's just a range with three resistance tests that could just as easily break upward. Shorting the third touch alone is guessing the direction, not trading the pattern.

How do you calculate the target? Pattern height (highest peak minus lowest trough) × 49%, subtracted from the breakout level — the full height is reached by fewer than half of cases. Also account for a pullback to the breakout line in 66% of cases, and keep the stop above the last peak.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero