Pattern direction

CandlestickBullish key reversal barBearish key reversal bar

Key Reversal Bar Pattern: How to Spot It, Trade It and What the Data Shows

One bar that pushes the trend to a new extreme, then closes back beyond the previous closes. This guide covers the bull and bear versions, the trade we test and how each one did.

· Founder of PatternsWizard
Updated Statistics updated 13 min read
Bullish key reversal bar patternBearish key reversal bar pattern

How we tested this pattern

Sample
63,148 occurrences on 6 markets (crypto, forex, US stocks, ETFs, indices and commodities), 1970 to 2026.
Detection
Found by code, never picked by hand: every match of the rule on full price history, winners and losers alike.bullish key reversal bar pattern: After a downtrend: an outside bar (higher high and lower low than the previous bar) making the lowest low of the previous 10 bars, at least as wide as their average, that closes above the two previous closes.bearish key reversal bar pattern: After an uptrend: an outside bar (higher high and lower low than the previous bar) making the highest high of the previous 10 bars, at least as wide as their average, that closes below the two previous closes.
Benchmark
Each occurrence is compared with the same assets on random bars: the move 10 bars later, and a 2:1 reward-to-risk trade.
Checks
Block bootstrap and a correction for the thousands of tests we run, plus a random-walk self-test that the bench passes: on random prices, no pattern should win.

Every pattern's results on one sheet

The only pattern cheat sheet with tested stats: 152 patterns, what the textbook says and what really happened on 6 markets. Free PDF, sent to your inbox.

One email with the PDF, then a note when a result changes. Unsubscribe anytime. Privacy.

At a glance: what textbooks say, what we measured
Pattern
Bullish key reversal bar
Occurrences tested
23,281 occurrences, all markets, daily
Textbook: price should rise after it
Too close to call · −0.10% vs the market, 10 bars
Where it did best, US stocks left out
Crypto, 4-hour: +0.11 R 2:1 trade vs random dates, not proven
Verdict
10-bar move: No clear edge2:1 trade: No clear edge
Pattern
Bearish key reversal bar
Occurrences tested
39,867 occurrences, all markets, daily
Textbook: price should fall after it
Too close to call · +0.01% vs the market, 10 bars
Where it did best
Forex, daily: +0.13 R 2:1 trade vs random dates, not proven
Verdict
10-bar move: No clear edge2:1 trade: No clear edge

See the statistics

On this page
  1. What is a key reversal bar pattern?
  2. Features of a bull reversal bar
  3. Features of a bear reversal bar
  4. Identifying key reversal patterns on your chart
  5. Common types of reversal bar patterns
  6. How to trade key reversal patterns
  7. Risk management and stop-loss placement
  8. Common mistakes when trading reversal bars
  9. Using technical indicators to validate reversals
  10. Reliability of the key reversal bar: what our tests show
  11. Real examples from our test bench
  12. Comparing reversal bars to candlestick patterns
  13. Frequently asked questions about reversal bars
  14. Summary of trading reversal bars in 2026

A key reversal bar pattern is a single price bar that pushes the trend to a new extreme, then turns and closes beyond the previous closes. In an uptrend, the bar makes a higher high, then closes below the prior closes (a bear reversal bar); in a downtrend, it makes a lower low, then closes above them (a bull reversal bar). Its older name is the key reversal day.

This guide covers the definition, how to trade the pattern, the mistakes to avoid, and what happened next. Our test bench detected 23,281 bullish and 39,867 bearish key reversals on daily price history across stocks, crypto, forex, ETFs, indices and commodities, and compared each outcome with random days on the same assets.

What is a key reversal bar pattern?

Defining the mechanics of a reversal bar

Definitions vary between authors. The loosest version only asks for a new high followed by a close below the previous close. The stricter version, the one our bench uses, has three conditions:

  1. A prior trend. The close before the pattern is at least one ATR above (for a top) or below (for a bottom) the close 10 bars earlier, and in the upper (for a top) or lower (for a bottom) half of the last 40 closes.
  2. An outside bar. The bar trades both above the previous high and below the previous low: its range covers the whole range of the prior bar. It also sets a new high (for a top) or low (for a bottom) of the previous 10 bars and is at least as wide as their average.
  3. A close beyond the two previous closes. Below both for a top, above both for a bottom.

The chart below is not drawn by hand. It is the median shape of the clean key reversals our bench detected, with the faded candles showing the trend before the pattern, and the lines of the trade we test: entry beyond the extreme of the reversal bar, stop at the other end, target at twice the risk.

TPSLEntryBullish
Bullish setup on the median shape of 2,172 daily occurrences drawn at random from all markets, with entry, stop loss (SL) and 2:1 target (TP)
TPSLEntryBearish
Bearish setup on the median shape of 2,047 daily occurrences drawn at random from all markets, with entry, stop loss (SL) and 2:1 target (TP)

What a reversal bar shows

The pattern shows a change of direction inside one session. The market first extends the move beyond the previous bar, then the push fails and price closes on the other side. That failed push is the signal: the trend tried and could not hold. It also gives a clear risk level, the extreme of the session.

Features of a bull reversal bar

The bull version forms at the end of a decline. It trades below the previous low, then rallies to close above the two previous closes, ideally near its high. Its low is the new low of the move. On the median daily occurrence, the range was 4.55% of price, or 1.43 times the average true range (ATR): a large candle, but not an extreme one.

Features of a bear reversal bar

The bear version is the mirror image at the top of a rally. It trades above the previous high, then falls to close below the two previous closes, ideally near its low. Its high becomes the high of the move, and the stop level for anyone selling the break.

Identifying key reversal patterns on your chart

123
  1. Previous barits high and low are both exceeded
  2. Closes above the two previous closes
  3. Outside barhigher high and lower low than the previous bar
Median shape of 2,172 daily occurrences drawn at random from all markets, those with the candle colors drawn here (72% of the 3,000 drawn), found by our test bench.
123
  1. Previous barits high and low are both exceeded
  2. Outside barhigher high and lower low than the previous bar
  3. Closes below the two previous closes
Median shape of 2,047 daily occurrences drawn at random from all markets, those with the candle colors drawn here (68% of the 3,000 drawn), found by our test bench.

The importance of context in market trends

The classic reading gives the pattern meaning only at the end of a trend: at a new high or low of the move, at a support or resistance level, or at the end of a completed chart pattern. Our bench requires the prior trend and a new 10-bar extreme but does not test the other locations.

The larger trend changed little in our data. We split daily bull reversal bars by the 200-day moving average: below it, the trade reached its target 39.1% of the time (+0.21 R); above it, 37.5% (+0.14 R). The bear version showed an even smaller gap, in the other direction: 27.7% below and 27.9% above. These breakdowns are descriptive and not corrected for the number of tests.

Volume confirmation for reversal signals

The textbook rule wants high volume on the reversal session. On daily charts, bull reversal bars with volume above 1.5 times the 20-day average reached the 2:1 target 36.8% of the time against 39.3% on normal volume. The bear version on high volume did slightly worse: 26.0% against 28.1%. High volume is not a reliable filter on its own.

Common types of reversal bar patterns

The two-bar reversal pattern

A two-bar reversal is a strong trend candle followed by a strong candle in the opposite direction that closes near the start of the first. The key reversal is a stricter two-bar pattern, because the second bar must also take out both extremes of the first. The engulfing pattern is the candlestick version based on bodies only: over 48,554 daily bullish engulfing trades, the 2:1 setup reached its target 36.9% of the time against 35.7% on random dates, close to the results below.

The three-bar reversal pattern

A three-bar reversal adds a confirmation: the next bar closes beyond the reversal bar, above its close for a bull reversal or below it for a bear one; a stricter version asks for a close beyond its extreme. The three outside up and down patterns use the first rule, on an engulfing candle. Confirmation gives a later entry and a wider stop.

Outside bars and their significance

Every key reversal is an outside bar, but an outside bar alone is not a reversal signal. The outside bar shows an expansion of range; the close beyond the prior closes and the location at the end of a trend turn it into a reversal. The island reversal is another fast turn, isolated by gaps on both sides.

Inside bars and pattern transformation

An inside bar is the opposite: a session whose high and low stay within the previous range. After a reversal, an inside bar shows a pause, and its break in the new direction can be traded with a tighter stop. A false break of an inside bar is the hikkake pattern.

How to trade key reversal patterns

The pattern is a setup, not a trigger. The trade starts only when price breaks the extreme of the reversal bar in the new direction. This is the trade our bench simulates on every occurrence.

TPSLEntryBullish
Entry
Buy stop at the pattern high
Stop loss (SL)
At the pattern low
Target (TP)
Twice the risk (2:1)
The trade our bench tests, drawn on the typical shape. Median shape of 2,172 daily occurrences drawn at random from all markets, those with the candle colors drawn here (72% of the 3,000 drawn).
TPSLEntryBearish
Entry
Sell stop at the pattern low
Stop loss (SL)
At the pattern high
Target (TP)
Twice the risk (2:1)
The trade our bench tests, drawn on the typical shape. Median shape of 2,047 daily occurrences drawn at random from all markets, those with the candle colors drawn here (68% of the 3,000 drawn).

A clean case from our bench: US stocksChosen on its shape alone, never on the outcome. Stock prices are adjusted for splits and dividends, so they can differ from the prices quoted at the time.

◂ Pattern43.2228.84+10 barsTPSLEntry2022-09
TextbookLRCX October 13, 2022Daily chart14.2% in the expected direction after 10 bars2:1 trade: target hit, +2.00 R

A clean case from our bench: US stocksChosen on its shape alone, never on the outcome. Stock prices are adjusted for splits and dividends, so they can differ from the prices quoted at the time.

◂ Pattern28.0714.43+10 barsTPSLEntry2008-08
TextbookWFC September 15, 2008Daily chart7.3% against the expected direction after 10 bars2:1 trade: stopped out, −1.00 R

Entry strategies for long positions

After a bull reversal bar, place a buy stop just above its high. On daily charts, that entry was triggered in 82.2% of cases; in 17.3% of cases price broke below the low first and the setup was canceled before any trade. The remaining setups expired: neither level was reached within 20 bars, or before the end of the data.

Entry strategies for short positions

After a bear reversal bar, place a sell stop just below its low. On daily charts, the entry was triggered in 79.0% of cases and canceled in 20.7%, when price took out the high first; the remaining setups expired, as on the long side.

Risk management and stop-loss placement

Setting protective stops below the reversal bar

Managing trade exits and profit targets

Our bench sets the target at twice the risk. On daily charts, bullish trades that reached that target took a median of 12 sessions, and stopped trades a median of 6. Also, of the bullish trades that hit their stop, 24.0% had first been up 1 R. A prior swing level is another reasonable target.

Common mistakes when trading reversal bars

Trading against the primary trend

A daily reversal against a clear weekly trend can be just a pullback. Check the larger trend first.

Ignoring support and resistance levels

The classic reading trusts a reversal at a prior high, a prior low or a round number more than one in empty space, because a level stands behind it. We have not tested support and resistance levels.

Failing to wait for confirmation

Using technical indicators to validate reversals

Moving averages and trend exhaustion

The 200-day moving average is the filter we tested, and it changed little. A shorter average can show whether price is stretched before the reversal; we have not tested it.

Oscillator divergence as a secondary signal

Another filter is momentum divergence: price makes a new extreme while an oscillator does not. We have not tested this combination: if you use it, test it on your own market first.

Reliability of the key reversal bar: what our tests show

The key reversal bar pattern is not reliable on its own: on no market or timeframe we tested did its average excess return over the rest of the market pass our correction for multiple testing in the expected direction. Where the gap is large enough to measure, it mostly points the wrong way: on hourly crypto and daily index charts the bear version moved against the expected direction relative to its market (−0.06% and −0.33%), by a gap that survives our correction, and on daily index charts the bull version lagged slightly, a gap that passes a simple test but not our correction. The one exception is weekly forex, where the bear version is rated Promising on 92 occurrences (58.9% lower 10 weeks later against 49.0% for random weeks, excess return +0.81%): it passes a simple test, not our correction.

54.0%vs 53.2% at random

Moved the expected way 10 bars later, compared with random bars.

Measured on 23,281 daily bullish key reversal bar occurrences across 2,235 assets. Random bars on the same assets are the honest baseline.

No clear edge
45.4%vs 45.5% at random

Moved the expected way 10 bars later, compared with random bars.

Measured on 39,867 daily bearish key reversal bar occurrences across 2,216 assets. Random bars on the same assets are the honest baseline.

No clear edge

Bullish key reversal bar against random bars on the same assets, by market, on daily candles. Bars of almost the same length mean no edge.

Moved the expected way 10 bars laterBullish key reversal barRandom bars
  • All markets
    All markets, Bullish key reversal bar: 54.0%
    All markets, Random bars: 53.2%
  • Crypto
    Crypto, Bullish key reversal bar: 43.1%
    Crypto, Random bars: 43.5%
  • US stocks
    US stocks, Bullish key reversal bar: 54.9%
    US stocks, Random bars: 53.9%
  • Forex
    Forex, Bullish key reversal bar: 50.4%
    Forex, Random bars: 49.9%
  • ETFs
    ETFs, Bullish key reversal bar: 52.1%
    ETFs, Random bars: 56.2%
  • Indices
    Indices, Bullish key reversal bar: 51.5%
    Indices, Random bars: 57.2%
  • Commodities
    Commodities, Bullish key reversal bar: 50.9%
    Commodities, Random bars: 51.6%
2:1 trade hit its targetBullish key reversal bar tradeSame trade, random dates
  • All markets
    All markets, Bullish key reversal bar trade: 38.4%
    All markets, Same trade, random dates: 36.0%
  • Crypto
    Crypto, Bullish key reversal bar trade: 31.0%
    Crypto, Same trade, random dates: 34.3%
  • US stocks
    US stocks, Bullish key reversal bar trade: 39.2%
    US stocks, Same trade, random dates: 36.2%
  • Forex
    Forex, Bullish key reversal bar trade: 27.9%
    Forex, Same trade, random dates: 31.2%
  • ETFs
    ETFs, Bullish key reversal bar trade: 35.0%
    ETFs, Same trade, random dates: 37.9%
  • Indices
    Indices, Bullish key reversal bar trade: 37.3%
    Indices, Same trade, random dates: 33.8%
  • Commodities
    Commodities, Bullish key reversal bar trade: 39.3%
    Commodities, Same trade, random dates: 38.1%

Bearish key reversal bar against random bars on the same assets, by market, on daily candles. Bars of almost the same length mean no edge.

Moved the expected way 10 bars laterBearish key reversal barRandom bars
  • All markets
    All markets, Bearish key reversal bar: 45.4%
    All markets, Random bars: 45.5%
  • Crypto
    Crypto, Bearish key reversal bar: 50.1%
    Crypto, Random bars: 54.5%
  • US stocks
    US stocks, Bearish key reversal bar: 45.5%
    US stocks, Random bars: 45.3%
  • Forex
    Forex, Bearish key reversal bar: 48.8%
    Forex, Random bars: 48.5%
  • ETFs
    ETFs, Bearish key reversal bar: 38.8%
    ETFs, Random bars: 42.3%
  • Indices
    Indices, Bearish key reversal bar: 41.3%
    Indices, Random bars: 42.2%
  • Commodities
    Commodities, Bearish key reversal bar: 48.9%
    Commodities, Random bars: 47.2%
2:1 trade hit its targetBearish key reversal bar tradeSame trade, random dates
  • All markets
    All markets, Bearish key reversal bar trade: 27.8%
    All markets, Same trade, random dates: 28.5%
  • Crypto
    Crypto, Bearish key reversal bar trade: 28.3%
    Crypto, Same trade, random dates: 29.2%
  • US stocks
    US stocks, Bearish key reversal bar trade: 27.8%
    US stocks, Same trade, random dates: 28.4%
  • Forex
    Forex, Bearish key reversal bar trade: 36.5%
    Forex, Same trade, random dates: 32.2%
  • ETFs
    ETFs, Bearish key reversal bar trade: 23.9%
    ETFs, Same trade, random dates: 28.4%
  • Indices
    Indices, Bearish key reversal bar trade: 29.5%
    Indices, Same trade, random dates: 29.9%
  • Commodities
    Commodities, Bearish key reversal bar trade: 29.7%
    Commodities, Same trade, random dates: 29.5%
All statistics

Full results for the key reversal bar pattern, market by market

Every occurrence our rule finds on full price history, compared with what the same assets did on random bars. A pattern only earns an edge if it beats random entries after correcting for the thousands of tests we run.

Our verdict

Not a reliable signal on its own here. Use it as context, not as a trigger.

10-bar move No clear edge2:1 trade No clear edge

Bullish key reversal bar: 23,281 occurrences tested

Bullish pattern · All markets · Daily · 2,235 assets · 1970-04 to 2026-10

Moved the expected way after 10 bars
54.0%
vs 53.2% on random bars
Excess return vs its market, 10 bars
−0.10%
95% interval −0.30% to +0.07%
its market−3%+3%
2:1 trade hit the target
38.4%
vs 36.0% for the same trade on random dates
Average result per trade
+0.18 R
vs +0.10 R on random dates

What this means. No measurable difference from the rest of its market over the next 10 bars.

Timeline

Entry triggered, after the pattern
1 day · middle half 1 to 1 · on 19,147 triggered entries
Stop order beyond the pattern's extreme. 77 of these entries are not among the 19,070 trades: still open at the end of our data, or left out by the trade rules.
Time to the 2:1 target, when hit
12 days · middle half 6 to 22 · on 7,330 trades that hit it
Time to the stop, when hit
6 days · middle half 2 to 13 · on 11,304 trades that hit it
Time in the trade, all exits
8 days · middle half 3 to 18 · on 19,070 trades

Size and trade path

Pattern range, high to low
4.6% · middle half 3.0% to 7.3% · on 23,281 occurrences
Range vs average bar (ATR)
1.43× · middle half 1.19× to 1.78× · on 23,281 occurrences
Best open profit during the trade
1.22 R · middle half 0.37 R to 2.11 R · on 19,070 trades
Stopped trades that were first up 1 R
24.0% · on 11,304 stopped trades
Trades a stop moved to breakeven at +1 R would have closed flat instead of at a loss
Entry order triggered
82.2% · on 23,281 occurrences

Medians, each on the cases named on its line; "middle half" spans the 25th to 75th percentile; no value under 30 cases.

How we detect it. After a downtrend: an outside bar (higher high and lower low than the previous bar) making the lowest low of the previous 10 bars, at least as wide as their average, that closes above the two previous closes. Candlestick counts are not checked against planted examples, so they may not be exhaustive.

All measurements (20 rows)

10-bar move: did price go the expected way?

Figures in gray: fewer than 30 occurrences, not interpreted. Excess return is signed so that positive favors the pattern. The test uses the bootstrap’s standard error and the interval its percentiles: near the 0.05 line, or on a few extreme moves, an interval can leave out zero while the test does not pass, and the test decides. Stock data misses most delisted companies, a survivorship bias we have not sized, which can flatter stock rows.
MarketOccurrencesPatternRandom barsExcess10-bar move verdict
Crypto · 1-hour42,11347.4%48.2%−0.04%No clear edge
Forex · 1-hour8,64749.1%50.5%+0.001%No clear edge
All markets · 1-hour50,76047.7%48.5%−0.04%No clear edge
Crypto · 4-hour11,35549.1%47.1%−0.01%No clear edge
Forex · 4-hour2,57049.4%50.4%−0.001%No clear edge
All markets · 4-hour13,92549.2%47.7%−0.01%No clear edge
Crypto · Daily1,45843.1%43.5%−0.03%No clear edge
US stocks · Daily20,34854.9%53.9%−0.10%No clear edge
ETFs · Daily53952.1%56.2%−0.30%No clear edge
Indices · Daily26851.5%57.2%−0.54%Weak, below chance
Forex · Daily39850.4%49.9%−0.10%No clear edge
Commodities · Daily27050.9%51.6%+0.36%No clear edge
All markets · Daily23,28154.0%53.2%−0.10%No clear edge
Crypto · Weekly10536.0%37.2%+16.24%Not enough data
US stocks · Weekly4,04658.6%56.7%−0.18%No clear edge
ETFs · Weekly13951.1%60.2%−1.42%No clear edge
Indices · Weekly5567.3%61.8%−0.19%No clear edge
Forex · Weekly6857.4%50.1%+0.47%No clear edge
Commodities · Weekly5542.6%53.3%−0.58%No clear edge
All markets · Weekly4,46857.8%56.3%+0.10%No clear edge

2:1 trade: target hit and average result

R = initial risk of the trade. Entry by an order at the trigger level, filled at the open when a bar opens beyond it. Exit at the stop or the target, or at the open when a bar opens beyond either one; otherwise at the close after 100 bars. When one bar touches both the stop and the target, the loss is counted. No fees or slippage. n/a = not measured. Per trade: average result, against the same trade on random dates. Figures in gray: fewer than 30 closed trades, not interpreted.
MarketTradesTarget hitSame trade, random datesPer trade2:1 trade verdict
Crypto · 1-hour35,16832.3%32.6%0.00 R vs +0.01 RNo clear edge
Forex · 1-hour7,08930.5%31.9%−0.04 R vs −0.01 RNo clear edge
All markets · 1-hour42,25732.0%32.5%−0.01 R vs 0.00 RNo clear edge
Crypto · 4-hour9,12836.3%33.0%+0.11 R vs +0.01 RNo clear edge
Forex · 4-hour2,17729.9%31.3%−0.09 R vs −0.04 RNo clear edge
All markets · 4-hour11,30535.1%32.6%+0.08 R vs 0.00 RNo clear edge
Crypto · Daily1,06031.0%34.3%−0.06 R vs +0.04 RNo clear edge
US stocks · Daily16,82239.2%36.2%+0.20 R vs +0.10 RNo clear edge
ETFs · Daily42635.0%37.9%+0.09 R vs +0.16 RNo clear edge
Indices · Daily22037.3%33.8%+0.11 R vs +0.02 RNo clear edge
Forex · Daily32327.9%31.2%−0.13 R vs −0.04 RNo clear edge
Commodities · Daily21939.3%38.1%+0.20 R vs +0.16 RNo clear edge
All markets · Daily19,07038.4%36.0%+0.18 R vs +0.10 RNo clear edge
Crypto · Weekly7145.1%33.2%+0.35 R vs 0.00 RNot enough data
US stocks · Weekly3,44140.9%40.1%+0.24 R vs +0.23 RNo clear edge
ETFs · Weekly11741.9%46.9%+0.27 R vs +0.44 RNo clear edge
Indices · Weekly5143.1%35.8%+0.30 R vs +0.10 RNo clear edge
Forex · Weekly6232.3%30.6%+0.08 R vs −0.07 RNo clear edge
Commodities · Weekly4341.9%45.0%+0.26 R vs +0.35 RNo clear edge
All markets · Weekly3,78540.9%40.1%+0.24 R vs +0.23 RNo clear edge

Our verdict

Not a reliable signal on its own here. Use it as context, not as a trigger.

10-bar move No clear edge2:1 trade No clear edge

Bearish key reversal bar: 39,867 occurrences tested

Bearish pattern · All markets · Daily · 2,216 assets · 1970-09 to 2026-10

Moved the expected way after 10 bars
45.4%
vs 45.5% on random bars
Excess return vs its market, 10 bars
+0.01%
95% interval −0.08% to +0.11%
Excess return is signed so that positive favors the pattern. On a bearish setup, it means the asset lagged its market.
its market−3%+3%
2:1 trade hit the target
27.8%
vs 28.5% for the same trade on random dates
Average result per trade
−0.16 R
vs −0.14 R on random dates

What this means. No measurable difference from the rest of its market over the next 10 bars.

Timeline

Entry triggered, after the pattern
1 day · middle half 1 to 1 · on 31,513 triggered entries
Stop order beyond the pattern's extreme. 183 of these entries are not among the 31,330 trades: still open at the end of our data, or left out by the trade rules.
Time to the 2:1 target, when hit
9 days · middle half 4 to 18 · on 8,715 trades that hit it
Time to the stop, when hit
5 days · middle half 2 to 11 · on 22,220 trades that hit it
Time in the trade, all exits
6 days · middle half 3 to 14 · on 31,330 trades

Size and trade path

Pattern range, high to low
3.3% · middle half 2.3% to 5.1% · on 39,867 occurrences
Range vs average bar (ATR)
1.41× · middle half 1.18× to 1.75× · on 39,867 occurrences
Best open profit during the trade
0.82 R · middle half 0.27 R to 2.03 R · on 31,330 trades
Stopped trades that were first up 1 R
22.9% · on 22,220 stopped trades
Trades a stop moved to breakeven at +1 R would have closed flat instead of at a loss
Entry order triggered
79.0% · on 39,867 occurrences

Medians, each on the cases named on its line; "middle half" spans the 25th to 75th percentile; no value under 30 cases.

How we detect it. After an uptrend: an outside bar (higher high and lower low than the previous bar) making the highest high of the previous 10 bars, at least as wide as their average, that closes below the two previous closes. Candlestick counts are not checked against planted examples, so they may not be exhaustive.

All measurements (20 rows)

10-bar move: did price go the expected way?

Figures in gray: fewer than 30 occurrences, not interpreted. Excess return is signed so that positive favors the pattern. The test uses the bootstrap’s standard error and the interval its percentiles: near the 0.05 line, or on a few extreme moves, an interval can leave out zero while the test does not pass, and the test decides. Stock data misses most delisted companies, a survivorship bias we have not sized, which can flatter stock rows.
MarketOccurrencesPatternRandom barsExcess10-bar move verdict
Crypto · 1-hour43,85648.6%50.6%−0.06%Opposite of the textbook
Forex · 1-hour9,52149.7%49.1%+0.002%No clear edge
All markets · 1-hour53,37748.8%50.3%−0.05%Opposite of the textbook
Crypto · 4-hour12,65048.7%51.8%+0.04%No clear edge
Forex · 4-hour2,99149.8%49.2%+0.02%No clear edge
All markets · 4-hour15,64148.9%51.3%+0.04%No clear edge
Crypto · Daily1,76350.1%54.5%+0.26%No clear edge
US stocks · Daily34,73345.5%45.3%+0.02%No clear edge
ETFs · Daily1,50038.8%42.3%−0.16%No clear edge
Indices · Daily95941.3%42.2%−0.33%Opposite of the textbook
Forex · Daily48148.8%48.5%−0.01%No clear edge
Commodities · Daily43148.9%47.2%+0.04%No clear edge
All markets · Daily39,86745.4%45.5%+0.01%No clear edge
Crypto · Weekly9757.0%59.3%+5.08%Not enough data
US stocks · Weekly11,38842.1%41.6%−0.04%No clear edge
ETFs · Weekly48433.0%36.2%−0.74%No clear edge
Indices · Weekly29534.9%37.1%−0.09%No clear edge
Forex · Weekly9258.9%49.0%+0.81%Promising
Commodities · Weekly9440.2%44.9%−1.06%No clear edge
All markets · Weekly12,45041.8%41.5%−0.04%No clear edge

2:1 trade: target hit and average result

R = initial risk of the trade. Entry by an order at the trigger level, filled at the open when a bar opens beyond it. Exit at the stop or the target, or at the open when a bar opens beyond either one; otherwise at the close after 100 bars. When one bar touches both the stop and the target, the loss is counted. No fees or slippage. n/a = not measured. Per trade: average result, against the same trade on random dates. Figures in gray: fewer than 30 closed trades, not interpreted.
MarketTradesTarget hitSame trade, random datesPer trade2:1 trade verdict
Crypto · 1-hour35,89828.8%31.4%−0.07 R vs −0.01 RWeak, below chance
Forex · 1-hour7,86230.2%32.0%−0.06 R vs −0.01 RWeak, below chance
All markets · 1-hour43,76029.0%31.5%−0.07 R vs −0.01 RWeak, below chance
Crypto · 4-hour9,86829.3%30.4%−0.04 R vs −0.02 RNo clear edge
Forex · 4-hour2,49131.8%32.0%−0.02 R vs −0.02 RNo clear edge
All markets · 4-hour12,35929.8%30.7%−0.04 R vs −0.02 RNo clear edge
Crypto · Daily1,29528.3%29.2%−0.07 R vs −0.04 RNo clear edge
US stocks · Daily27,45127.8%28.4%−0.16 R vs −0.15 RNo clear edge
ETFs · Daily1,10223.9%28.4%−0.30 R vs −0.16 RWeak, below chance
Indices · Daily73929.5%29.9%−0.12 R vs −0.10 RNo clear edge
Forex · Daily40036.5%32.2%+0.11 R vs −0.01 RNo clear edge
Commodities · Daily34329.7%29.5%−0.08 R vs −0.10 RNo clear edge
All markets · Daily31,33027.8%28.5%−0.16 R vs −0.14 RNo clear edge
Crypto · Weekly3215.6%16.9%−0.38 R vs −0.39 RNot enough data
US stocks · Weekly8,89622.8%23.5%−0.30 R vs −0.28 RNo clear edge
ETFs · Weekly35623.3%25.1%−0.28 R vs −0.23 RNo clear edge
Indices · Weekly21823.4%27.4%−0.29 R vs −0.17 RNo clear edge
Forex · Weekly7739.0%28.5%+0.18 R vs −0.15 RNo clear edge
Commodities · Weekly7112.7%21.9%−0.59 R vs −0.31 RNo clear edge
All markets · Weekly9,65022.8%23.6%−0.30 R vs −0.27 RNo clear edge

Data updated November 5, 2026 · code version c2a1683f · Read the methodology. We measure every pattern on crypto and forex (1-hour, 4-hour, daily and weekly charts) and US stocks, ETFs, indices and commodities (daily and weekly charts). A market or timeframe from that list missing from the tables above had no occurrence.

What you often read

“The bullish key reversal bar is a reliable reversal signal.”

What our data shows

On 23,281 daily occurrences, it moved in the expected direction 54.0% of the time, against 53.2% for random bars: no clear edge, since the average excess return is not statistically significant.

What you often read

“The bearish key reversal bar is a reliable reversal signal.”

What our data shows

On 39,867 daily occurrences, it moved in the expected direction 45.4% of the time, against 45.5% for random bars: no clear edge, since the average excess return is not statistically significant.

Table quoted in the text (verdicts and every market: see All measurements)
Version, market and timeframeCasesMoved as expected after 10 barsRandom datesClosed trades2:1 target hitSame trade, random datesExpectancy (R)
Bull, all markets, daily23,28154.0%53.2%19,07038.4%36.0%+0.18 vs +0.10
Bull, US stocks, daily20,34854.9%53.9%16,82239.2%36.2%+0.20 vs +0.10
Bull, crypto, hourly42,11347.4%48.2%35,16832.3%32.6%0.00 vs +0.01
Bull, forex, hourly8,64749.1%50.5%7,08930.5%31.9%−0.04 vs −0.01
Bull, all markets, weekly4,46857.8%56.3%3,78540.9%40.1%+0.24 vs +0.23
Bear, all markets, daily39,86745.4%45.5%31,33027.8%28.5%−0.16 vs −0.14
Bear, crypto, hourly43,85648.6%50.6%35,89828.8%31.4%−0.07 vs −0.01
Bear, forex, hourly9,52149.7%49.1%7,86230.2%32.0%−0.06 vs −0.01
Bear, all markets, weekly12,45041.8%41.5%9,65022.8%23.6%−0.30 vs −0.27

Read the table in pairs of columns. On daily charts, 54.0% of bull reversal bars were followed by a higher price 10 days later, against 53.2% for random days on the same assets. The excess return over the rest of the market was −0.10%, with a 95% confidence interval from −0.30% to +0.07%: no clear edge. The bear version was followed by a lower price 45.4% of the time, almost exactly the rate of random days.

The 2:1 trade tells the same story. The bull version reached its target in 38.4% of 19,070 trades against 36.0% on random dates (+0.18 R against +0.10 R before costs): the gap in average result per trade does not pass our statistical test. The bear version lost money before costs on daily charts (−0.16 R), as did the same trade on random dates (−0.14 R): selling assets that rose over our data is a headwind. On hourly crypto, the bear version's trade read weak, below chance: its average result per trade (−0.07 R against −0.01 R) passes a simple test but not our correction. On hourly forex, the bull version stayed close to the same trade on random dates (−0.04 R against −0.01 R), while the bear version's trade read weak, below chance (−0.06 R against −0.01 R), again a gap that passes a simple test but not our correction.

Which version works better?

Neither showed a proven edge. The bull version's best rows, such as the 2:1 trade on daily US stocks (see the FAQ below), do not pass our statistical test, and on hourly crypto the bear version's trade read weak, below chance. The bull version had a positive expectancy on daily and weekly charts, but so did the same trade on random dates (+0.10 R on daily charts, +0.23 R on weekly charts), because these assets rose over our data. The island reversal gives a similar picture: on 262 daily tops, 43.5% were lower 10 days later against 45.1% for random days.

Real examples from our test bench

The gallery in this section shows real key reversal bars picked by our bench (best and worst results, textbook shapes, most recent, random draws), with their actual outcome. The cases written up in this section come from daily crypto charts.

Real occurrences from our test bench, each on its market's main timeframe. Crypto and US stock examples come from liquid assets only (the largest coins, the most traded stocks). Best and Worst are the best and worst 2:1 trade results, Textbook the cleanest shapes (chosen on geometry alone, never on the outcome), Most recent the latest detections among the assets the gallery draws from, Typical a random draw. Shaded area: the pattern. Dashed lines: entry, stop loss (SL) and 2:1 target (TP) of the tested trade. Dotted line: 10 bars after the signal. A result above +2 R or below −1 R means a bar opened past the target or the stop, and the trade was closed at that open. Stock prices are adjusted for splits and dividends, so they can differ from the prices quoted at the time. We left out 15 cards: 1 with a bar far out of line with the bars around it (a likely bad quote), 13 with a shape that does not read as the definition on this page, 1 with a single bar so large that the pattern or the trade levels could not be read. They still count in the statistics, and are not replaced: our bench picks a fixed number of cards per category.

Market
Examples
Crypto overallAcross 42,113 bullish key reversal bar occurrences on 1-hour charts, price moved the expected way 10 bars later 47.4% of the time, against 48.2% for random bars. The average excess return does not pass our significance test.No clear edgeThe examples below are single cases.
◂ Pattern24.0921.61+10 barsTPSLEntry2023-12
Best tradeAVAX December 5, 20231-hour chart19.8% in the expected direction after 10 bars2:1 trade: target hit, +2.00 RAfter the pattern, price left the chart: high 26.96 on December 6, 2023.
◂ Pattern17.9517.36+10 barsTPSLEntry2023-08
TextbookETC August 14, 20231-hour chart0.2% against the expected direction after 10 bars2:1 trade: stopped out, −1.00 R
◂ Pattern5.104.95+10 barsTPSLEntry2023-082023-08
TextbookDOT August 14, 20231-hour chart0.1% against the expected direction after 10 bars2:1 trade: stopped out, −1.00 R
◂ Pattern5.955.50+10 barsTPSLEntry2023-03
TypicalDOT March 8, 20231-hour chart1.5% against the expected direction after 10 bars2:1 trade: stopped out, −1.00 R
◂ Pattern2.662.46+10 barsTPSLEntry2025-09
TypicalFIL September 17, 20251-hour chart2.9% in the expected direction after 10 bars2:1 trade: target hit, +2.00 R
◂ Pattern3.593.24+10 barsTPSLEntry2025-07
TypicalDOT July 2, 20251-hour chart1.1% in the expected direction after 10 bars2:1 trade: target hit, +2.00 R
Examples
US stocks overallAcross 20,348 daily bullish key reversal bar occurrences, price moved the expected way 10 bars later 54.9% of the time, against 53.9% for random bars. The average excess return does not pass our significance test.No clear edgeThe examples below are single cases.
◂ Pattern299.5242.0+10 barsTPSLEntry2024-03
Best tradeAMGN April 19, 2024Daily chart15.8% in the expected direction after 10 bars2:1 trade: target hit, +4.75 R
◂ Pattern43.2228.84+10 barsTPSLEntry2022-09
TextbookLRCX October 13, 2022Daily chart14.2% in the expected direction after 10 bars2:1 trade: target hit, +2.00 R
◂ Pattern173.1133.6TPSLEntry2026-092026-10
Most recentMSTR October 20, 2026Daily chart2:1 trade: in progress
◂ Pattern336.9255.5TPSLEntry2026-092026-10
Most recentAPP October 15, 2026Daily chart2:1 trade: stopped out, −1.00 R
◂ Pattern14.157.19+10 barsTPSLEntry2008-08
TypicalMRVL September 11, 2008Daily chart7.9% against the expected direction after 10 bars2:1 trade: invalidated before entry
◂ Pattern11.899.43+10 barsTPSLEntry2002-12
TypicalVZ January 28, 2003Daily chart2.4% in the expected direction after 10 bars2:1 trade: target hit, +2.00 R
◂ Pattern35.7029.33+10 barsTPSLEntry2002-12
TypicalLLY January 29, 2003Daily chart3.3% against the expected direction after 10 bars2:1 trade: invalidated before entry
Examples
Crypto overallAcross 43,856 bearish key reversal bar occurrences on 1-hour charts, price moved the expected way 10 bars later 48.6% of the time, against 50.6% for random bars. The verdict rests on the average excess return over the rest of the market, −0.06% (positive favors the pattern), not on this rate.Opposite of the textbookThe examples below are single cases.
◂ Pattern3.232.61+10 barsTPSLEntry2025-10
Best tradeNEAR October 10, 20251-hour chart23.0% in the expected direction after 10 bars2:1 trade: target hit, +2.00 RAfter the pattern, price left the chart: low 1.55 on October 10, 2025.
◂ Pattern2.371.39+10 barsTPSLEntry2020-11
Worst tradeSOL November 6, 20201-hour chart27.2% against the expected direction after 10 bars2:1 trade: stopped out, −1.00 R
◂ Pattern15.8714.29+10 barsTPSLEntry2023-122023-12
TextbookLINK December 22, 20231-hour chart1.5% against the expected direction after 10 bars2:1 trade: stopped out, −1.00 R
◂ Pattern5.704.94+10 barsTPSLEntry2023-02
TextbookFIL February 7, 20231-hour chart2.8% against the expected direction after 10 bars2:1 trade: invalidated before entry
◂ Pattern1.271.06+10 barsTPSLEntry2024-11
TypicalXRP November 16, 20241-hour chart8.2% against the expected direction after 10 bars2:1 trade: invalidated before entry
◂ Pattern0.10880.1018+10 barsTPSLEntry2022-07
TypicalXLM July 14, 20221-hour chart1.3% against the expected direction after 10 bars2:1 trade: stopped out, −1.00 R
◂ Pattern0.06270.0623+10 barsTPSLEntry2022-10
TypicalTRX October 4, 20221-hour chart0.5% against the expected direction after 10 bars2:1 trade: stopped out, −1.00 R
◂ Pattern656.0634.9+10 barsTPSLEntry2026-05
TypicalBNB May 9, 20261-hour chart0.9% in the expected direction after 10 bars2:1 trade: target hit, +2.00 R

A bull reversal bar that reached its target

On November 22, 2022, AVAX/USDT traded below the previous low (11.57) to 11.44, then closed at 12.36, above the two previous closes. The buy stop at 12.38 was triggered the next day, with a stop at 11.44 and a 2:1 target at 14.26. Price reached the target on December 5, and AVAX was 9.9% higher 10 days after the pattern.

A bull reversal bar stopped the next day

On May 22, 2023, AVAX/USDT printed a bull reversal bar from 14.08 to 14.79, closing at 14.68. The buy stop at 14.79 was triggered the next day, and the day after, price fell to 14.00, through the stop at 14.08: a full 1 R loss. AVAX was 4.0% lower 10 days later.

A bear reversal bar that hit its stop

On February 16, 2024, NEAR/USDT traded up to 3.461, above the previous high, then closed at 3.279, below the two previous closes. The sell stop at 3.172 was triggered the next day, but price rallied and hit the stop at 3.461 on February 18. NEAR was 24.9% higher 10 days after the pattern: the reversal never came.

A bear setup canceled before entry

On March 5, 2024, a bear setup on AAVE/USDT never triggered: price took out its high (119.98) on March 6 before its low, and the setup was canceled. AAVE was 22.1% higher 10 days after the pattern.

Comparing reversal bars to candlestick patterns

The key reversal is a bar chart concept: it uses the high, low and close, and ignores the open. Candlestick patterns such as the engulfing, the hammer or the morning star focus on the body. A key reversal can also be an engulfing candle, and a bull version with a long lower wick looks like a hammer. The labels differ, the idea is the same: a failed push followed by a close on the other side.

Frequently asked questions about reversal bars

Which time frame is best for reversal patterns?

None gave a proven edge. Among the bull version's 2:1 trades, the largest gap is weekly crypto (45.1% of targets hit on 71 trades against 33.2% on random dates). The larger samples include 4-hour crypto (36.3% of targets hit against 33.0% on random dates) and daily US stocks (39.2% of targets hit on 16,822 trades against 36.2% on random dates). The last two are rated no clear edge, and our bench gives weekly crypto, a small sample, no verdict at all. Among the all-markets rows, weekly bull reversals had the highest raw results (40.9% of targets hit on 3,785 trades), but random weeks on the same assets came close (40.1%), and the average result per trade was about the same (+0.24 R against +0.23 R).

Can I use reversal bars in automated trading?

Yes. The definition is mechanical, which is why we can test it. Code the three conditions, add the entry, stop and target, and compare the results with random entries on the same assets before trading it.

How do I distinguish between a reversal and a pullback?

You cannot from one session alone. A reversal shows up later as a lower high (or higher low) and a break of the prior swing. Until then, treat it as a possible pullback.

How often does the pattern appear?

Often: about one bull version every 364 daily sessions and one bear version every 213, on average per asset.

Summary of trading reversal bars in 2026

  • A key reversal bar is an outside bar at the end of a trend that closes beyond the two previous closes.
  • The trade starts on the break of its extreme, with the stop at the other end.
  • Over 19,070 daily bull reversal trades, the 2:1 setup reached its target 38.4% of the time against 36.0% on random dates, a slightly higher hit rate, but the average result per trade (+0.18 R against +0.10 R) is within the margin of error.
  • The bear version was followed by a lower price 45.4% of the time, almost exactly like random days.
  • Volume and the 200-day moving average changed the results very little.

Treat the pattern as a location for a trade, not as a reason for one. Define the stop before entry, keep the risk per trade small, and remember that past results do not guarantee future results.

About the author and the data. Written by Amaury, founder of PatternsWizard, who is passionate about finance, quantitative trading and data analysis. He built the PatternsWizard test bench, which detects each setup with fixed, published rules on full market history: crypto, forex, US stocks, ETFs, indices and commodities. Trades and win rates are compared with random bars on the same assets, excess returns with the rest of the market, with error margins and a correction for multiple testing, and the whole pipeline is checked on random-walk prices, where nothing can be predicted. Statistics on this page were generated on November 5, 2026. The full methodology explains every step.

, Founder of PatternsWizard

Passionate about finance, quantitative trading and data analysis. Amaury built the PatternsWizard test bench, which detects each pattern on full price history and measures what happened next against random entries.

Statistics on this page come from our open methodology: every occurrence our rule finds on full history, compared with chance. How we test patterns.

179 patterns testedRandom-walk self-test passed