Skepticism Toward Stock Candle Sticks

The author of this piece is bias against candle stick patterns. I had difficulty with the idea that a technique for gauging rice prices in 18th century Japan could be used to predict stock movement.

In light of this, I attempted to objectively research the subject. Before turning to results from Google Scholar, here’s a disclaimer: this is not financial advice, and I am not a financial adviser.

TLDR: Academia’s view on candlestick patterns is mixed and pattern-dependent but mostly negative.

Article selection criteria:

  • Article was published later than 1995.
  • Article commented on the profitability of candlestick patterns, not just whether or not they could be found.
  • Article did not study theoretical candlestick patterns. It researched patterns that traders use in practice.

Articles

Roy, Partha. (2012). Fuzzy candlestick approach to trade S&P CNX NIFTY 50 index using engulfing patterns. International Journal of Hybrid Information Technology. 5. 57–66.

  • Tested S&P CNX Nifty 50 Index of India with real-time 2012 data.
  • Achieved “good returns” with 2-day engulfing pattern and a modified version of the pattern they call the “U-Turn.”
“U-Turn” candlestick pattern
  • Not sure how to interpret “good returns,” so scoring this as a “weak win for a few specific patterns and loss for other one day patterns”

G. Caginalp & H. Laurent (1998) The predictive power of price patterns, Applied Mathematical Finance, 5:3–4, 181–205, DOI: 10.1080/135048698334637

  • Tested S&P 500 with data from 1992–1996.
  • Found a profit of 1% possible over a two-day holding period (think about that annually).
  • Found Three Inside Up, Three Outside Down, and Evening Star patterns most profitable. Three-day patterns were more profitable in general.
  • Strong win for specific three-day patterns but loss for others.

Lu, Tsung-Hsun, Yung-Ming Shiu and Tsung-Chi Liu. “Profitable candlestick trading strategies — The evidence from a new perspective.” (2012).

  • Tested Taiwan Top 50 Tracker Fund with data from 2002–2008.
  • Reviewed two-day patterns:
From Lu, Shiu, and Liu (2012)
  • Found 9% returns with piercing pattern.
  • Found support for bull engulfing pattern during a bull market, and support for the bear engulfing patter during a bear market, but this was largely eliminated when put through a more rigorous mathematical process called “bootstrapping.”
  • Scoring this a win for specific patterns but a loss for others.

Marshall, B.R., Young, M.R. & Cahan, R. Rev Quant Finan Acc (2008) 31: 191. https://doi.org/10.1007/s11156-007-0068-1

  • Tested Japanese equity market data from 1975 to 2004.
  • Covered most patterns used by traders and found them not to be profitable.
  • Loss for candlesticks.

Wu, Y., Lu, J., & Qi, A. (2017). Profitability of Candlestick Charting Patterns in the Stock Exchange of Thailand. Advances in Mechanical Engineering. https://doi.org/10.1177/1687814018819562

  • Tested Thai Set 50 Index with data from 2006–2016.
  • Tested most patterns traders use and found them not to be profitable.
  • Loss for candlesticks.

Prado, Hércules & Ferneda, Edilson & Morais, Luis & Luiz, Alfredo & Matsura, Eduardo. (2013). On the Effectiveness of Candlestick Chart Analysis for the Brazilian Stock Market. Procedia Computer Science. 22. 10.1016/j.procs.2013.09.200.

  • Tested Ibovespa, a Sao Paulo Stock Exchange Index, with data from 2005–2009.
  • Covered most patterns used by traders and found them not to be profitable.
  • Loss for candlesticks.

Zhu, Min & Atri, Said & Yegen, Eyub. (2015). Are candlestick trading strategies effective in certain stocks with distinct features?. Pacific-Basin Finance Journal. 37. 10.1016/j.pacfin.2015.10.007.

  • Tested all stocks traded on the Shanghai and Shenzhen exchanges with data from 1999–2008.
  • Examined two-day candlestick patterns.
  • Found that, “Bearish harami and cross patterns do well in predicting the turning points of upward trends while bullish piercing, engulfing, and harami profitably predict the turning points of downward swings” (Zhu, Atri, and Yegen).
  • Best results from holding one day for bullish patterns and five days for bearish patterns.
  • Suggest candlestick patterns may be more profitable in exchanges with more individual investors, such as the young Chinese exchanges, rather than ones with strong institutional investors, such as US markets.
  • Win for two-day candlestick patterns in non-institution dominated exchanges.

Tsung-Hsun Lu & Yung-Ming Shiu (2016) Can 1-day candlestick patterns be profitable on the 30 component stocks of the DJIA?, Applied Economics, 48:35, 3345–3354, DOI: 10.1080/00036846.2015.1137553

  • Tested 30 stocks from the Dow Jones Industrial Average over data from 1974 to 2009.
  • Best results from Gravestone Doji, which had 1.17% returns over a one-day period and 1.71% returns over a five-day holding period.
  • Win for specific candlestick pattern. Loss for others.
One day candlestick patterns from Lu & Shiu

Horton, Marshall. (2009). Stars, crows, and doji: The use of candlesticks in stock selection. The Quarterly Review of Economics and Finance. 49. 283–294. 10.1016/j.qref.2007.10.005.

  • Tested 349 randomly selected stocks with representation of all major industries from the Value Line database with data from 1992–1996.
  • Also tested the S&P 500 with data from 1992–1996.
  • Covers most patterns traders use.
  • Found no impact — loss for all candlestick patterns.

Tharavanij, P., Siraprapasiri, V., & Rajchamaha, K. (2017). Profitability of Candlestick Charting Patterns in the Stock Exchange of Thailand. SAGE Open. https://doi.org/10.1177/2158244017736799

  • Tested Dow Jones Industrial Average over data from 1992 to 2002.
  • Covered most known patterns (actually went through authoritative Japanese literature on the subject).
  • Found no impact — loss for all candlestick patterns.

Overall, the findings are mostly negative for candlestick patterns. However, there may be enough pattern-specific evidence to validate a trader’s individual experience.

I personally find fascinating Zhu, Atri, and Yegen’s suggestion that candlestick patterns may be more effective in environments with fewer institutions.

If you have more articles, a different interpretation of the data, or other thoughts, eager to hear them below!

Until then, while some patterns had scattered support, most of these candles are burnt to the wick.

Interested in too many things.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store