The Inverse Head and Shoulders is one of the most widely taught bullish reversal patterns in technical analysis. Bulkowski’s encyclopedia, Warrior Trading, and TradingWithRayner have all covered it extensively. What almost none of those sources offer is real, segmented performance data. How often does the pattern actually win? Which setups are worth taking and which ones should you avoid? Where does the pattern work, where does it fail? How should you enter the trade?
The analysis that follows is drawn from a backtest of 67,041 completed Inverse Head and Shoulders trades, generated by scanning roughly 6,000 stocks every trading day across the NYSE and NASDAQ for sixteen bullish chart patterns and tracking what happens to each detection afterward.
The Unfiltered Pattern Is a Coin Flip
The first thing the data shows is that the Inverse Head and Shoulders without any quality filter is not a strategy. It is a coin flip with a slight tilt.
Across the lowest tier of detections, those scoring around 20 on a composite quality measure, we tracked 5,180 trades. The win rate landed at 47.2%. The market beat rate was 43.5%. The average return per trade was essentially zero. Move up to a quality score of 25 and the picture barely changes.
This is what most published studies of the pattern look like. They aggregate every detection into a single bucket and report the average. The average is unimpressive because it mixes the setups that work with the much larger pool of setups that do not.
That is the floor. It is also the credibility check on everything that follows. The edge comes from identifying statistical markers that separate high-quality patterns from the rest.
Scoring Separates the Winners From the Noise
When each detection is scored on structure, volume, and breakout readiness and reduced to a single composite, the population stops looking uniform. The full scoring curve from score 28 through 39 looks like this:
The slope steepens hard above 34. Below 30, the pattern is barely earning its commissions. From score 35 onward, win rates jump into the sixties, then the seventies. By the time you reach score 37 and above, the population looks like a different pattern entirely from the one at the bottom.
That single shift, from “does the pattern work” to “which sub-segment of the pattern works,” is the practical lesson of the entire dataset. Every other finding here is an extension of it.
Day One Tells You Which Trade You Are Actually In
Once you have a high-quality setup, the next question is how to trade it. The data points at two signals, both available within the first trading session after the pattern forms.
The first is the open gap. Stocks that open down on the day after the signal end up as winners much more often than stocks that gap up. The relationship is monotonic across the full distribution: trades opening more than 5% below entry win 82.7% of the time and return 4.2% on average, while trades gapping up more than 5% win 32.4% and return -4.5%. The intuition that an enthusiastic gap-up confirms the pattern is the opposite of what the data shows.
Day-1 Open Gap
Observations
Win Rate
Trade Return
>5%
173
32.4%
-4.5%
3% to 5%
596
38.4%
-2.6%
1% to 3%
3,514
44.8%
-0.9%
-1% to 1%
8,347
55.2%
0.4%
-3% to -1%
4,359
64.2%
1.6%
-5% to -3%
700
74.6%
3.3%
< -5%
156
82.7%
4.2%
The second is what happens during day one itself. Bucketing trades by how far the intraday low fell below entry on day one produces one of the cleanest monotonic curves in the dataset:
Day-1 Intraday Drawdown
Observations
Win Rate
Trade Return
Never below entry
103
78.6%
4.18%
Within 0.5% of entry
17,055
56.8%
2.18%
0.5% to 2.0% below
33,945
48.8%
1.22%
2.0% to 3.0% below
7,951
44.4%
0.07%
3.0% to 5.0% below
5,409
43.9%
-0.68%
More than 5.0% below
2,578
38.8%
-4.25%
The practical rule: watch the first session. The decision to keep or cut the trade can be made at the close of day one, not days later.
The Regime Matters More Than the Textbooks Admit
A reversal pattern is a bet on the market changing its mind. That bet should pay off differently in different environments. It does, but not in the direction most traders expect.
Each detection is tagged with a regime zone derived from the broader market state. The interesting result shows up when comparing the same quality of setup across the three bull regimes:
Score
Regime
Observations
Win Rate
Trade Return
37
High Confidence Bull
122
36.9%
-0.42%
37
Medium Confidence Bull
242
80.2%
1.89%
37
Low Confidence Bull
159
78.6%
3.52%
The same quality of pattern, in three different bull environments, produces wildly different outcomes. Score-37 setups in moderate and weak bull regimes win about 80% of the time. The same score in strong bull markets wins 36.9%. The gap is too large to be sample noise and it shows up at every score band where there is significant data.
The working theory: the Inverse Head and Shoulders is fundamentally a reversal. In moderate bull tape there is room to reverse a trend. Stocks finish corrections, base, and turn back up. In a strong bull regime almost everything is already going up, so the “reversal” is more often a temporary shakeout in something already trending, or a setup that triggers right before the broader market adjusts. The pattern has nowhere to actually reverse to.
The Rule the Textbooks Got Backwards
The single most interesting finding in the dataset is not about scoring or regimes. It is about pattern geometry.
Conventional trading advice states that shallow right shoulders are safer and deep right shoulders should be treated as a warning sign. The reasoning, stated most sharply by TradingWithRayner, is that a deep right shoulder creates extra selling pressure on the way back up to the neckline.
The data rejects this cleanly and loudly. Across 19,477 detections where the right shoulder depth could be measured:
Right Shoulder Depth
Observations
Win Rate
Trade Return
Shallow (1.6% to 5.8% below neckline)
2,071
32.8%
-0.07%
Medium (7.2% to 8.9%)
2,065
53.8%
0.60%
Deep (11.1% to 25.8%)
2,050
66.5%
0.84%
That is a 34-point spread in win rate from the same pattern at the same score level, and the difference is driven by the right shoulder depth alone, not by pattern quality. Shallow right shoulders are where the losers live.
This finding is not marginal. It is the strongest single independent geometric signal in the entire dataset, and it holds across every bull regime and into moderate bear markets. It is also directly inverted from the textbook advice every retail trader has read.
The result fits a broader phenomenon worth naming. Textbook-perfect chart patterns systematically underperform their messier counterparts, and the mechanism is generally the same one. Clean setups are what every retail trader is taught to look for, so they draw the most attention, get bought into earliest, and resolve fastest. By the time the textbook version of a pattern actually completes, the easy edge has already been extracted by the crowd. The deeper, less photogenic right shoulder is, in this sense, doing exactly what messier patterns do across the board: filtering out the trades that are too obvious to be profitable.
Five geometric theories were tested in total. Two held up: pattern efficiency and RSI divergence at the head. Two were rejected: declining volume into the right shoulder and shoulder symmetry. One, right shoulder depth, was inverted.
What This Means for Traders
A few rules of thumb derived from the data are worth considering for anyone trading this pattern.
Filter on pattern quality first. Scores below 32 are barely worth the position. Scores of 35 and above generate serious returns.
Read the next-day open. Gap-down opens win much more often than gap-ups. Wait for the boring entry, not the exciting one.
Use day one as confirmation. If the stock holds within half a percent of entry through the close of day one, you are sitting in a population that wins meaningfully more than the baseline.
Mind the regime. Moderate and weak bull markets are where the high-quality setups deliver. High confidence bull markets are where they go to die.
And pay attention to the right shoulder. Deeper is better. The textbook is wrong on this one.
A note on methodology: these results come from a systematic backtest with the standard caveats. Backtested performance does not guarantee future results, and the regime distribution of the underlying period skews toward the bull tape that has dominated US equities since 2020. The shoulder-depth and day-1 findings are robust enough across sub-samples to be worth attention, but every trader should validate against their own execution and risk model before acting.
