The kicking pattern, usually called the kicker, is two long candles of opposite colour with almost no wicks and a gap between them: the market opens on the other side of where it closed and never comes back through.

On eight months of EURUSD and a German index, across four timeframes, the pattern as TA-Lib defines it did not occur once. Three of the four published readings of the gap were satisfied zero times. The fourth produced 28 hits, and all 28 of them open inside the first candle's own range, which is not a gap by any ordinary meaning of the word. Gaps happen in that sample, 66 of them, and opposite-colour marubozu pairs happen, 75 of them. The two sets never once intersect.

What is the kicking (kicker) candlestick pattern?

Two candles. The first is a long candle in the direction of the prevailing move with very short shadows at both ends, which is a marubozu. The second is a marubozu of the opposite colour, and it opens on the far side of a gap.

TA-Lib publishes it as CDLKICKING: "Two-candle pattern of two opposite-color marubozu (long bodies with very short shadows) separated by a price gap. A reversal signal whose direction is set by the second candle's color".

TradingView states the same shape in more detail and is among the strictest sources here. Its bullish page requires the first candle to be "a bearish marubozu candlestick with next to no upper or lower shadow", the second to be "a bullish marubozu", and says "the second day gaps up extensively and opens above the opening price of the day before". That clause is where TradingView pins the gap to two named prices.

How short is "next to no shadow", and how long is "long"?

Both have published numbers. TA-Lib states its thresholds as a range type, an averaging window and a factor, and the two a marubozu uses are BodyLong = RealBody, 10, 1.0 and ShadowVeryShort = HighLow, 10, 0.1.

In plain terms: a long body is one longer than the mean real body of the previous ten bars, and a very short shadow is one no longer than a tenth of the mean high-to-low range of the previous ten bars. No explainer page puts a number on either.

Note what that second threshold permits. A marubozu does not have to have a flat top and bottom. Its shadows only have to be small relative to recent range, so the close can sit slightly away from the extreme and still qualify.

Is the kicking pattern bullish or bearish?

The direction is the second candle's colour: white for bullish, black for bearish. CDLKICKING emits +100 when the second candle is white and -100 when it is black.

The library ships a second function for the same shape that answers the question differently. CDLKICKINGBYLENGTH has the identical two-candle definition but sets its sign from "candlecolor of the candle with the larger realbody (i if realbody(i) > realbody(i-1), else i-1; tie goes to i-1)". When the first candle is the longer one, the two functions disagree about which way the pattern points.

No explainer page mentions that the second function exists. We measured the disagreement on the only reading that produces any hits at all: of 28 occurrences, 13 have the first candle as the longer body, so the two functions return opposite signs on 13 of 28. On EURUSD's 15-minute chart alone it is 8 of 14. If your platform offers a "Kicking" scan, it is worth finding out which of the two it runs.

There is also an asymmetry inside TradingView's own pair of pages. The bullish page requires both candles to be marubozu. The bearish page names the marubozu for the first candle only, "The first day candlestick is a bullish marubozu", and says nothing about the second candle's shadows.

Which of the two does ChartLabs run?

Both of them, as separate switches. Kicking and Kicking by Length are two of the 61 pattern switches in the Candlestick Patterns indicator, which you add from the chart's Indicators menu under the Patterns heading, and it draws on the price pane with your candles. Both sit in the two-candle section and both are on as soon as you add the indicator, so neither needs turning on by hand.

Kicking marks the bar with K and takes its direction from the second candle. Kicking by Length marks it with KL and takes its direction from the longer body. The indicator puts a bullish mark below the bar's low and a bearish one above the high, so where the first candle carries the longer body you get the K and the KL at opposite ends of the same bar, which is the disagreement above drawn on the chart.

Where exactly is the gap measured from?

It is answered four ways:

Reading of the gap	Who publishes it
Candle two opens beyond candle one's open	tradingview.com: "opens above the opening price of the day before"
Candle two opens beyond candle one's close	strike.money: "opens above the previous day's closing price"
Candle two opens beyond candle one's high	strike.money, further down the same page: "a long green candlestick that opens above the previous day's high"
The two candles do not overlap at all	tradingsim.com: "the candles never overlap. This is essential when validating the pattern"; chartguys.com: "a strong bullish candle follows a bearish one with no overlap between the two"

strike.money publishes two of those four on one page. For a bearish first candle, its close is at the bottom of the bar and its high is at the top, so "above the previous close" and "above the previous high" are the loosest and the strictest possible readings of the same sentence, on the same page.

LiteFinance drops the condition. It says the second candle is "often opening with a gap", then that the candles "as a rule, open with a small price gap", then that "there are also Kickers where the second candlestick opens at the opening level of the first candle", which is no gap at all. By the end it has turned the gap into a strength rating: "If there is a price gap between the opening prices of the two candlesticks, it is a powerful Kicker pattern."

Does the kicking pattern actually happen?

Here is what each reading selects on our own cleaned data: EURUSD (246,778 one-minute bars) and a euro-quoted German index (225,951 one-minute bars), 1 January to 31 August 2026, aggregated to the 15-minute, 1-hour, 4-hour and daily timeframes, eight symbol-and-timeframe cells and 42,252 bars in all, prices held in whole ticks.

Reading of the gap	With the marubozu condition	Without it
Opens beyond candle one's open (TradingView)	0	318
Opens beyond candle one's close (strike.money)	28	9,101
Opens beyond candle one's high (strike.money, same page)	0	87
The candles do not overlap (TradingSim, ChartGuys)	0	66
Opens at candle one's open (LiteFinance's no-gap case)	0	130

ChartGuys calls the pattern "rare but extremely telling" and gives no count. That is the count, for two instruments and eight months.

The 28 are not gaps

The one reading that produces hits is strike.money's loosest, and it does not describe a gap. When the first candle is a black marubozu its close sits at or very near the bottom of the bar, so opening above the previous close only asks the second candle to begin somewhere above the bottom of the first.

All 28 of those occurrences have the second candle opening inside the first candle's own high-to-low range, and not one opens beyond the first candle's far extreme. On the first candle the trailing shadow runs from 0 to 13.6 per cent of its own range, with a median of 4.3 per cent and exactly zero on 10 of the 28, which is what the ShadowVeryShort threshold allows. So the 28 are the count of one source's reading, which does not test what the pattern claims to test.

Placing the zero properly

A zero can come from the wrong place, so each condition was measured on its own over the same eight cells:

2,157 marubozu bars, between 2.90 and 5.79 per cent of all bars depending on the cell. Marubozu are ordinary.
75 opposite-colour marubozu pairs, with no gap condition applied at all. The pair happens.
66 whole-candle gaps, with no marubozu condition applied at all. Gaps happen.
0 pairs with a gap.

It is not the marubozu test being too strict, and it is not that nothing ever gaps. In our sample the two things never coincided.

All of that counts shapes under published rules, and none of it describes what price did after one.

Why does it hardly ever appear in forex?

Because the pattern needs a gap, and the instrument that gaps is the one whose session has a break in it.

Those 66 whole-candle gaps are not evenly spread. On the 15-minute chart, 31 of them are on the German index, which quotes in every hour of the day but pauses for roughly an hour each evening, against 4 on EURUSD, which quotes without a daily break on weekdays. On the open-to-open reading it is 205 against 37 on the same two cells. One short pause a day is enough, and a market with no pause at all has far fewer places to leave a hole.

A pattern built on gaps behaves differently on instruments that gap, and testing it on a forex pair is close to testing whether the pattern can exist at all. If it interests you, test it where price pauses and restarts: indices and commodities with a daily break in the session, and the weekly open.

One caveat about what we counted. These are gaps between consecutive aggregated bars on the chart, which is a statement about where price jumped between one bar's close and the next bar's open, and nothing more. It is not a claim about our underlying data: coverage inside a symbol's range is not uniform, and absent periods are normal rather than defects. Where the data comes from and how far back it runs is in clean forex historical data.

What is the difference between "kicking" and "kicker"?

Nothing. They are two names for the same two candles, and which one you see depends on where you look. TA-Lib and TradingView both call it Kicking; most of the explainer pages title it Kicker. People searching often type kicking. If you are hunting for this pattern in a platform's scan list or an indicator library, try both words before you decide it is missing.

Is it the same as a gap, a kicker gap, or an exhaustion gap?

A gap is one event: price jumps between a close and the next open. The kicking pattern is a gap plus four other conditions, which is why it is so much rarer than a gap, 66 whole-candle gaps in our sample and none of them with the marubozu pair on either side.

The exhaustion gap comparison comes from TradingSim, the one page that makes it. Its framing is that both patterns involve "a gap against the prior trend", and it separates them by what drives the gap rather than by shape. It also gives the fullest statement of its own rule set, four conditions: "the final candle of the old trend must be in that trend's direction, a gap must appear against the trend, the candle after the gap must confirm the new direction, and the two candles bracketing the gap must not overlap."

That list has no marubozu condition, which is the condition doing all the work in TA-Lib's version. The same page also undercuts the rule it calls essential: having said the candles never overlap, its own bearish worked example is introduced by "The key differential of this example from the previous bullish example is the small size of the gap candlestick."

We are not repeating any success rate from these pages. TA-Lib's own note on CDLKICKING cites research we are not in a position to reproduce here. The rarity figures above are measured and are ours; the performance figures are not something we can pass on.

How do you test the kicking pattern yourself?

The counts above say the strict form is absent from our sample. What they cannot tell you is whether a looser version you define yourself is worth anything. That is the part replay is for.

The two Candlestick Patterns switches described earlier use TA-Lib's definition, the one that found nothing in our sample: two opposite-colour marubozu with the second candle's range clear of the first's. On the eight months above they would have marked nothing. There is no scanner and no price alerts. For a looser gap rule, the Pine Script editor will mark it as long as your script defines the pattern, and "Example: Mark bars by a rule" on the Pine Scripts tab opens a copy to start from. Pine reads each bar's open, high, low and close, so a script can test both marubozu conditions and whichever gap rule you picked, then mark the second bar with plotshape, plotchar or a label, or draw the gap itself with a box. There is no Pine backtesting in ChartLabs, so a script draws the line and you judge the rule yourself on the chart.

Then the manual pass:

Write the gap test down in prices before you start. "A gap" is not a rule. "Candle two's low is above candle one's high" is a rule, and so is "candle two opens above candle one's open". The table above is what the difference costs: before the marubozu test those two are 66 and 318, and after it both are 0.
Pick an instrument whose session has a break in it. An index or commodity symbol that pauses each evening gives you something to look at, and those sit alongside forex and crypto in what we carry. The data behind all of them comes from Dukascopy, tick level where it is available, cleaned to remove anomalies and built into 1-minute candles as the lowest timeframe. Depth runs up to around 22 years on the oldest forex pairs, with other instruments starting later.
Replay the session open bar by bar. This pattern lives entirely in the first bar after a break, so stepping into the open one bar at a time is the only way to see what you would actually have had in front of you.
Decide whether you require the marubozu, and keep both versions. On our data that single condition is the difference between 66 and 0, so it is the choice that decides whether you will ever see the pattern.
Mark each candidate, tag it with which reading fired, and filter the analytics by tag later, so "non-overlap" and "open-to-open" stay separable. A note on the trade saying which reading fired, how big the gap was in your own units and whether it filled follows through to the trades CSV export.

Per candidate, record the symbol, timeframe, date and time, which gap reading it satisfied, whether both candles met the marubozu test, the size of the gap in ticks, and whether the gap filled and when. Those last two are the columns that tell you whether the pattern is doing anything your gap rule was not already doing. Record whether the gap sat on a session break, a weekly open or neither as well: on our data that one column explains most of the difference between instruments.

What you will not get, from your own testing or from anything published about the pattern, is a reliable success rate. What you can get is an honest count of how often the thing you defined shows up where you trade. The general method is in our beginner's guide to backtesting a strategy, and the session side of it is in the forex backtesting guide.

Start testing it

Step through real historical charts bar by bar, watch what happens at a session open, and take the practice trades by hand. One plan, $20 a month or $200 a year, with a 7-day free trial.

Start your free trial

Kicking Candlestick Pattern: Where Is the Gap Measured?