2004issue C081-4
Strength-rated hammer candlestick construction
A hammer can be recognized by ranking OHLC tests from most decisive to least decisive and counting how many a bar meets. Nest those checks, then use one stringency setting to loosen or tighten the same single-candle reversal hypothesis.
- A candlestick pattern can be strength-rated by ranking recognition conditions from most important to least important and counting how many of those conditions a bar satisfies.
- Criteria order is part of construction: a bearish-trend test comes first so the same shape in a bullish trend is treated as a hanging man and later hammer tests are not run.
- The first shape test places the real body at the upper end of the session range by requiring the real-body low to sit above the range midpoint; later checks cover shadow length, recency of the low, and a white real body.
- A stringency parameter grades the nested tests. A larger value lets fewer tests pass, a smaller value requires more tests, and a default of 6 keeps only the bearish-trend and upper-range real-body conditions.
Strength rating as a ranked count
A candlestick pattern can be strength-rated by ranking its recognition conditions from most important to least important and counting how many of those conditions a bar satisfies.
The overlay that count produces is a strength rating. It runs from a fully qualified hammer down to weaker numeric grades.
Trend test before shape tests
A hammer is a single-candle bullish-reversal construction that requires a bearish prior trend, a real body near the top of the range, and a long lower shadow.
Criteria order is part of the construction. A bearish-trend test comes first so the same shape in a bullish trend is treated as a hanging man and later hammer tests are not run.
Real body above the range midpoint
The real body is the absolute difference between the last price and the open. The session range is the high-minus-low span of the bar and is the scale for body and shadow percentages.
The hammer’s first construction test places the real body at the upper end of the session range by requiring the real-body low to sit above the range midpoint. That range midpoint is constructed as the low plus half the high-minus-low span.
Lower and upper shadow construction
The second construction test requires the lower shadow to be at least twice as long as the real body, and both lengths can also be written as percentages of the high-low range.
Upper-shadow construction treats a zero gap between the high and the real-body high as a shaven top, while a short non-zero upper shadow can still pass if it stays inside an adjustable percentage limit.
Session polarity
Session polarity is constructed from last minus open. A positive difference is a white real body, a negative difference is a black real body, and a zero difference is a doji.
Seven nested tests and one stringency setting
A fully qualified hammer is defined as passing seven nested tests, including a recency check that the low occurred inside a chosen lookback such as five bars and a final check for a white real body.
A stringency parameter grades those nested tests. A larger value lets fewer tests pass, a smaller value requires more tests, and a default of 6 keeps only the bearish-trend and upper-range real-body conditions.
All readings on this track · 8 readings
- 1992Constructing candlestick reversals from body, shadow, and trend location
- 2002A staged reading from long-shadow hammer to engulfing
- 2003Constructing a hammer from a failed breakdown
- 2004Hammer and hanging-man bars still need context and confirmation
- 2004Strength-rated hammer candlestick construction
- 2004Treat every paper umbrella as a timed experiment
- 2008Weekly engulfing, hammer, and soldier reversals as a control test
- 2018RSI, hammer, and moving-average gates for range hypotheses