Candlestick Pattern Probability Calculator
Free Candlestick pattern probability Calculator for technical analysis. Enter your numbers to see returns, costs, and optimized scenarios instantly.
Candlestick Pattern Probability Calculator
Calculator
Adjust values & calculateEnter your values below. Every result is computed in your browser — no data is sent to any server.
Formula: Adjusted Reading = Your Measured Win Rate x Volume Weight x Trend Weight x S/R Weight x Timeframe Weight x ATR Weight
Worked example — Adjusted reading: 80% | Four favourable weights applied to a measured 60% base
Formula
Adjusted Reading = Your Measured Win Rate x Volume Weight x Trend Weight x S/R Weight x Timeframe Weight x ATR Weight
The base rate is the historical win percentage you supply from your own backtest or broker statement. This page does not assert a success rate for any pattern, because published figures depend heavily on how each study defined the pattern and its exit rule. That base rate is then scaled by round, illustrative weights for volume confirmation, trend alignment, proximity to support or resistance, chart timeframe and volatility (ATR). Those weights show the direction and rough scale of each factor rather than measured effect sizes, so the output is a sensitivity scenario, capped at 95 percent and shown to whole percentages.
Worked Examples
Example 1: Hammer at Support with Volume Confirmation
Problem:Your own records show the hammers you traded resolved as reversals 60% of the time across 500 occurrences. One now forms at a key support level during a downtrend on the daily chart with above-average volume. ATR is 2.5%. Scale that base rate for the confluence.
Solution:Your measured base rate: 60% Volume weight (above average): x1.05 Trend alignment (downtrend + bullish pattern): x1.10 Support/Resistance (at support): x1.10 Timeframe (daily): x1.00 ATR weight (2.5%, above the 2% cutoff): x1.05 Adjusted: 60 x 1.05 x 1.10 x 1.10 x 1.00 x 1.05 = 80.0% 95% CI on the 60% base over 500 samples: +/-4.29 points
Result:Adjusted reading: 80% | Four favourable weights applied to a measured 60% base
Example 2: Doji in Mid-Range with Low Volume
Problem:Your records put doji reversals at 52%. One forms in the middle of a trading range on a 15-minute chart with below-average volume. ATR is 0.5%.
Solution:Your measured base rate: 52% Volume weight (below average): x0.95 Trend alignment (no clear trend): x0.95 S/R (mid-range): x0.95 Timeframe (15min): x0.85 ATR weight (0.5%): x0.90 Adjusted: 52 x 0.95 x 0.95 x 0.95 x 0.85 x 0.90 = 34.1% Break-even at the modelled 2:1 payoff is 33.3%, so this barely clears it before costs
Result:Adjusted reading: 34% | Every weight unfavourable - no confluence to work with
Frequently Asked Questions
How reliable are candlestick patterns for predicting price movements?
Candlestick patterns give probabilistic signals rather than certainties, and this page deliberately does not publish a success rate for any of them. Reported hit rates in the literature disagree sharply because each study defines the pattern differently, tests a different market and period, and above all uses a different exit rule: the same hammer can be scored a win under a fixed ten-bar hold and a loss under a trailing stop. Any single quoted percentage is therefore a property of the test design as much as of the pattern. The only base rate Candlestick Pattern Probability Calculator will use is the one you enter from your own backtest or broker statement, measured with the entry, stop and target you actually trade. Structure still matters directionally: a three-candle formation embeds more confirming price action than a single indecision candle, and confluence with volume, trend and a tested level tends to help. Treat published figures such as those collected by Bulkowski or Nison as starting points for your own testing rather than as inputs to a live expectancy calculation.
What factors increase the probability of a candlestick pattern working?
Several confluence factors significantly improve candlestick pattern reliability. Volume confirmation is among the most important: a reversal pattern forming on above-average volume has substantially higher success rates because it indicates genuine participation by market participants. Occurring at key support or resistance levels adds another layer of validation, as these are price points where supply and demand dynamics historically shift. Trend alignment matters: bearish reversal patterns are most reliable after extended uptrends, and bullish reversals after extended downtrends. Timeframe also plays a role, since each daily or weekly candle aggregates far more trading activity than an intraday one. What can be said about stacking these factors is directional rather than numerical: each additional independent confirmation filters out more of the setups that were going to fail, at the cost of leaving fewer trades to take. How much that is worth on your instrument is measurable only by scoring your own trades with and without each filter.
What is the difference between reversal and continuation candlestick patterns?
Reversal patterns signal a potential change in the prevailing trend direction, while continuation patterns suggest the current trend will persist after a brief pause. Key reversal patterns include hammer and inverted hammer (bullish reversal), shooting star and hanging man (bearish reversal), and multi-candle patterns like morning star (bullish) and evening star (bearish). Engulfing patterns can be either bullish or bearish depending on whether the engulfing candle is bullish or bearish. Continuation patterns include rising and falling three methods, tasuki gaps, and in-neck patterns. The distinction is critical for trading strategy: reversal patterns at support or resistance levels in a trending market provide high-probability entry signals, while continuation patterns within a trend confirm that pullbacks are temporary and the trend remains intact.
How should the Kelly criterion be applied to candlestick-based trading?
The Kelly criterion calculates the optimal position size to maximize long-term portfolio growth based on the edge (win probability) and the payoff ratio (reward-to-risk). The formula is: Kelly % = (p x b - q) / b, where p is win probability, q is loss probability (1-p), and b is the reward-to-risk ratio. For a candlestick pattern with 60 percent success rate and 2:1 reward-to-risk, Kelly suggests risking (0.60 x 2 - 0.40) / 2 = 40 percent of capital. However, most professional traders use fractional Kelly (typically one-quarter to one-half Kelly) because the full Kelly criterion assumes perfectly known probabilities and can lead to excessive drawdowns. With candlestick patterns where probabilities are estimated rather than precisely known, quarter-Kelly (10 percent in this example) provides a more conservative approach that protects against model uncertainty while still optimizing growth.
Why do candlestick patterns perform differently across timeframes?
Candlestick pattern reliability varies across timeframes primarily due to the amount of market noise relative to genuine price signals. On very short timeframes like one-minute and five-minute charts, random price fluctuations (market microstructure noise, bid-ask bounce, and algorithmic trading activity) create many false patterns that appear valid but lack meaningful participation from price-driving participants. As the timeframe increases, each candle represents more trading activity and more significant decision-making by market participants. Daily candles encompass all trading sessions and reflect institutional participation, making patterns more meaningful. Weekly and monthly candles carry more information still but generate far fewer opportunities. This page will not attach a number to that advantage, because it depends on the instrument, the session and the costs you pay. What is structural rather than statistical is the cost side: spread and commission are charged per trade regardless of timeframe, so a five-minute pattern targeting a few pips must clear a much larger proportional hurdle than a daily one targeting a multi-day swing.
How does the sample size input affect the 95 percent confidence band?
Sample size never moves the probability itself; it only sets how precisely your measured base rate is pinned down. The tool takes the standard error as the square root of p times one minus p divided by n, then multiplies by 1.96, where p is the historical win rate you entered and n is the number of occurrences it was measured over. At a 60 percent win rate over 500 observations that works out to roughly plus or minus 4.3 percentage points. Because the error falls with the square root of n, quadrupling the sample halves the band. Note that the band applies to that measured input alone and says nothing about the additional uncertainty the confluence weights introduce.
What does the expected value figure represent, and when does it turn positive?
Expected value is quoted in percent of price here, because the modelled win is two times the ATR percentage and the modelled loss is one times it. It equals the reversal probability times the win minus the failure rate times the loss. With that fixed two-to-one payoff the break-even probability is one in three, so any adjusted probability above 33.3 percent shows positive expectancy, although a thin edge leaves nothing for spread and slippage.
Why are the confluence modifiers multiplied together rather than added?
Each weight scales the running estimate instead of shifting it by a fixed number of points, so three favourable 1.1 factors compound to 1.331 rather than 1.3, and one 0.85 timeframe penalty trims every other gain proportionally. Multiplying keeps each adjustment in proportion to whatever base rate you started from, which matters because a factor worth six points on a 60 percent base should not be worth six points on a 30 percent one. The weights themselves are round illustrative values rather than measured effect sizes, so read the output as a sensitivity scenario, not a forecast. Because stacked favourable factors can otherwise push the figure past 100 percent, the output is capped at 95 percent and shown to whole percentages only.
Why does a low ATR reading reduce the estimated probability?
The volatility modifier sits at 0.9 when ATR is 1 percent of price or less, is neutral between 1 and 2 percent, and climbs to 1.05 and then 1.1 as ATR passes 2 and 3 percent. The reasoning is that in a very quiet market the bodies and shadows defining the pattern are small next to the spread and ordinary noise, leaving little room for follow-through. Remember the modelled win and loss also shrink with ATR.
What is the difference between a hammer and a hanging man?
They are the same candle shape, with a small body near the top of the range, a lower shadow roughly twice the body or longer, and little or no upper shadow. Only context separates them. Appearing after a decline the candle is a hammer and is read as a bullish reversal; the identical candle after a rally is a hanging man and is read as bearish. That is why the trend field changes the estimate even when the pattern choice does not.
References
Reviewed for accuracy by Daniel Agrici, Founder & Lead Developer · Editorial policy
Related Calculators
🧮Harmonic Pattern Calculator
Calculate harmonic pattern ratios for Gartley, Butterfly, Bat, and Crab patterns.
🧮Propulsion Block Calculator
Identify propulsion blocks (order blocks within FVGs) for high-probability ICT trade entries.
🧮Power of Three Calculator
Analyze the Power of Three (Accumulation, Manipulation, Distribution) pattern for ICT trading.
🧮Unicorn Model Calculator
Calculate ICT unicorn model setups combining breaker blocks with FVGs for high-probability entries.
🧮ICT Master Setup Calculator
The ultimate ICT setup validator — input market structure, FVG, order blocks, and killzone timing to score trade probability using Inner Circle Trader methodology.
🧮ICT Macro Time Calculator
Calculate ICT macro time windows at :50-:10 of each hour for high-probability entries.
🧮Pip Calculator
Calculate pip with inputs, formulas, and instant results.
🧮Forex Margin Calculator
Calculate forex margin with inputs, formulas, and instant results.