Browse the library

Candlesticks · Single-candle patterns

Rickshaw man candlestick: the centered doji, rules and results

How to spot a rickshaw man, the midpoint test TA-Lib applies, why it fires on 10% of trading days, and what 3,133 real cases did next.

On this page
  1. Anatomy of a rickshaw man
  2. The exact rules Finabase uses
  3. What the balance means
  4. Confirmation
  5. Two rickshaw men from 2026
  6. What the data says
  7. Rickshaw man versus similar candles
  8. An educational outline
  9. How Finabase shows the rickshaw man
  10. Frequently asked questions
  11. Summary

The rickshaw man is the most symmetrical candle in the doji family. The open and close are virtually the same, there is a long shadow above and a long shadow below, and the tiny body sits close to the middle of the day's range. If you squint, it looks like the two long poles and the seat of a pulled rickshaw, which is where the name comes from.

Where the Long-legged doji: rules, the TA-Lib quirk and real results only says "long shadows", the rickshaw man adds balance: the market travelled about as far up as it did down and finished in the center. It is the purest picture of a session that found no new value.

This page covers the anatomy, the exact three-part test in TA-Lib's CDLRICKSHAWMAN (including how loose "near the midpoint" can be), two real cases from 2026 with very different outcomes, a test on 3,133 detections, and how Finabase's scanner and gauge treat a candle that is neutral by design.

Anatomy of a rickshaw man

A rickshaw man after an advance (illustration, prices not real) Body at the midpoint
An idealized rickshaw man at the end of an advance. The body sits at the center of a wide range.

Three features define it:

  1. A doji body. Open and close within a few cents of each other.
  2. Two long shadows. The session traded well above and well below the open.
  3. A centered body. The open and close are near the midpoint between the high and the low, so neither shadow dominates.

The symmetry is what distinguishes it from its neighbours. A doji with a long lower shadow and its body at the top is a Dragonfly doji: meaning, TA-Lib rules and a 380-case test; the mirror image is a Gravestone doji: meaning, rules and why Finabase files it bullish. Both of those lean one way. The rickshaw man refuses to lean.

The exact rules Finabase uses

Finabase's markers come from TA-Lib's CDLRICKSHAWMAN, run on daily bars. Every condition must hold:

Condition in TA-Lib Default setting
Doji body Real body ≤ 10% of the average high–low range of the previous 10 candles
Long upper shadow Upper shadow longer than the candle's own real body
Long lower shadow Lower shadow longer than the candle's own real body
Body near midpoint Body overlaps the band of the range midpoint ± 20% of the average high–low range of the previous 5 candles
Trend before it Not tested
Output +100 on every match

Two details deserve attention.

"Long" is measured against the body. As with the long-legged doji, TA-Lib's ShadowLong setting compares each shadow with the candle's own real body. A doji's body is tiny, so both shadows usually clear it easily. The shadows do not have to be long compared with other candles, and the session does not need a wide range.

"Near the midpoint" is measured against recent ranges, not this candle's range. The tolerance is 20% of the average range of the last five sessions, on either side of the midpoint. When today's range is small compared with that average, the tolerance can cover most of the candle, and a body sitting three-quarters of the way up still passes. The Intel example below is exactly that case.

A long-legged doji with its body off-center (illustration, prices not real) Body too high: not a rickshaw man
A long-legged doji that is not a rickshaw man: the body is too far from the midpoint of this wide range.

Of the 4,119 TA-Lib dojis in our sample, 986 were not rickshaw men. In 326 of them one shadow was no longer than the body; in the other 660 the body sat too far from the midpoint.

What the balance means

A centered doji describes a session in which the market explored both directions with roughly equal force and rejected both. Buyers paid up early or late, sellers knocked price down at some point, and by the close the two sides had cancelled out at the middle of the day's range.

Context changes how traders read that balance:

  • After a strong rally it suggests the buyers' advantage has evaporated. Nison and others treat it as a warning that the trend may be tiring.
  • After a sharp decline it suggests the sellers' advantage has gone. Same logic, opposite direction.
  • Inside a range it is simply what a balanced market looks like day to day.

A rickshaw man with a range much wider than its neighbours is more informative than a narrow one. A wide one says a lot of shares changed hands while the price got nowhere, which is a sign of genuine disagreement about value.

Confirmation

Because the rickshaw man has no lean, everything rests on what comes next:

  • After an advance, a close below the candle's low is the bearish follow-through.
  • After a decline, a close above its high is the bullish one.
  • A close back inside the range means the market is still undecided, and many traders simply wait.

Two rickshaw men from 2026

Intel (INTC) daily — a rickshaw man after a 27% rally, January 2026 $40 $50 Rickshaw man 2026-01-22 −11.2% after 10 sessions Jan 2026 Feb
Intel (INTC), December 2025 to February 2026. The rickshaw man on 2026-01-22 came after a 27% rally and was followed by a 17% one-day drop.

Intel, January 22, 2026. INTC had gained 27.4% in ten sessions. On January 22 it opened at $54.21 and closed at $54.32, with a high of $54.60 and a low of $53.08. Volume ran about 1.85 times its 20-day average. Strictly, this is not a symmetrical candle: the lower shadow ($1.13) was four times the upper one ($0.28), and the body sat about three-quarters of the way up the range. It passed TA-Lib's midpoint test because the session's $1.52 range was only about half the recent average, which widened the tolerance. The next session closed at $45.07, 17.0% lower, far below the candle's low. Ten sessions after the pattern INTC was 11.2% lower. In textbook terms this was a warning at the top that came true. In Finabase's scanner it would have been listed as a bullish pattern the evening before the drop.

Salesforce (CRM) daily — a rickshaw man after a 19% decline, June 2026 $160 $180 $200 Rickshaw man 2026-06-22 +12.9% after 10 sessions Jun Jul
Salesforce (CRM), May to July 2026. The rickshaw man on 2026-06-22 followed a 19% decline; ten sessions later the stock was 12.9% higher.

Salesforce, June 22, 2026. This one is close to the textbook. CRM had fallen 19.1% in ten sessions. The June 22 candle opened at $150.16 and closed at $150.12, a body of 4 cents, with an upper shadow of $3.71 and a lower shadow of $3.80. Its range was about 22% wider than the recent average. The next session closed at $153.42, just inside the candle's high of $153.87, so the strict confirmation came later. Ten sessions after the pattern CRM was 12.9% higher, and it traded as high as $172.38 in that window. A wide, balanced session after a steep fall marked the low of the move this time.

What the data says

We ran CDLRICKSHAWMAN over daily bars from October 30, 2024 to October 29, 2026 for 60 large US stocks and ETFs. It fired 3,133 times on all 60 symbols; 3,055 cases had ten sessions of data after them. The last three rows are our own split of the same data.

Group Cases Higher after 5 sessions Avg 5-session return
Any trading day (the baseline) 29,760 53.1% +0.35%
Every rickshaw man 3,055 53.4% +0.44%
Confirmed: next close above the high, from that close 958 54.2% +0.52%
Range at least the 10-session average 251 54.2% +0.27%
… of those, after a 10-session advance 147 58.5% +0.75%
… of those, after a 10-session decline 104 48.1% −0.41%

The average ten-session return after a rickshaw man was +0.66%, identical to the baseline. 46.9% of the measured cases followed a ten-session decline.

The plain reading:

  • As detected, the rickshaw man was a coin flip weighted exactly like the market. One trading day in ten qualifies, and those days behaved like the rest.
  • Wide-range rickshaw men did not behave as the textbook predicts. If anything, the ones after an advance were followed by gains more often than average, and the ones after a decline by losses. With 147 and 104 cases, those splits are well within the range of chance, so they are not evidence of a reversed signal either.

Rickshaw man versus similar candles

Pattern What it requires Lean
Rickshaw man Doji, both shadows longer than body, body near the midpoint None
Long-legged doji Doji, at least one shadow longer than body None
Dragonfly doji Doji, almost no upper shadow, a real lower shadow Up
Gravestone doji Doji, almost no lower shadow, a real upper shadow Down
High-wave candle Short body (not necessarily a doji), both shadows over 2× body None
Spinning top Short body, both shadows longer than the body None

The High-wave candle: what it means, TA-Lib rules and test results candle and the rickshaw man often overlap. In our data 2,670 candles were both, because a centered doji with long shadows easily meets the high-wave condition of shadows more than twice the body.

An educational outline

This is a study framework, not a recommendation to trade.

  1. Check the width. Compare the candle's range with the ten-session average. Skip narrow ones; they are ordinary quiet days.
  2. Check the trend. Look for a clear move of several percent in the prior two weeks.
  3. Wait for a close outside the range, against the prior trend.
  4. Define risk at the opposite extreme. For a bullish read, just below the low; for a bearish read, just above the high.

How Finabase shows the rickshaw man

  • Chart markers. Add the Rickshaw Man study to a chart and Finabase marks each CDLRICKSHAWMAN detection. Expect roughly one marked candle every two weeks on a typical large stock.
  • Scanner. TA-Lib returns +100 for every rickshaw man, and the scanner files detections by that sign, so they always appear in the Bullish list (S&P 500 rickshaw men). The Intel case above would have been listed there.
  • Signals-tab gauge. On the Signals tab, each of the 61 patterns votes from its most recent detection within the last eight bars. In our data the rickshaw man's vote was Buy on 58.5% of stock-days and never Sell. Because every rickshaw man is also a doji and a long-legged doji, one centered candle usually delivers three Buy votes together, enough on its own for a Buy verdict on the candlestick gauge (3 ÷ (3 + 0 + 3) = 0.5).

Frequently asked questions

What does a rickshaw man candle mean?

It means the session moved well above and well below its open and closed in the middle, with no winner. After a strong trend it can signal that the trend has lost its driving force. On its own it predicts nothing.

How is it different from a long-legged doji?

A rickshaw man is a long-legged doji whose body sits near the center of the range. In TA-Lib it must also have both shadows longer than the body, while the long-legged doji needs only one. That is why TA-Lib found 3,133 rickshaw men and 4,119 long-legged dojis in our sample.

Why does Finabase call it bullish?

Because TA-Lib's function returns +100 for every match and Finabase sorts patterns by that sign. The candle itself is neutral; the label is a side effect of how the output is encoded.

Does a wider range make it more reliable?

It makes the candle more meaningful as a description of the session. In our sample it did not make the outcome more predictable, but the wide-range group was small.

Summary

The rickshaw man is the balanced doji: long shadows on both sides and the close back at the middle. TA-Lib finds it when a doji's two shadows beat its own body and its body falls near the midpoint, a test met on about one day in ten, and generous enough to admit lopsided candles on narrow days. Across 3,055 measured cases the outcome matched the market's baseline. Finabase marks it, scans for it and lets it vote in the gauge as bullish, so the direction is your job: look at the trend, the range and the next close.

Sources and further reading

  1. Steve Nison, Japanese Candlestick Charting Techniques (New York Institute of Finance, 1991)
  2. TA-Lib source code — CDLRICKSHAWMAN and the default candle settings
  3. Finabase chart data — daily bars for 60 US stocks and ETFs, 2024-10-30 to 2026-10-29; patterns detected with TA-Lib CDLRICKSHAWMAN