Premium Market Analysis, Trader Education, Software, and Trading Strategies. Thirty Years Of Skin In The Game

Technical Analysis

Stationary Win Rate Examples

Visual chartists, also known as noise traders, in the process of defending a method with a low win rate claim that high win rate strategies do not exist. When presented with evidence to the contrary, they argue that a high win rate is non-stationary. In the article, we present examples that refute this claim.

In another article, we presented three examples of trading strategies with high win rates: two short-term and one longer-term. One of the strategies has been in the public domain since at least 2003.

The main response on Twitter by some revolved around the non-stationarity of the win rate. This is a valid argument because if the win rate is non-stationary, then any associated strategy may not be robust. However, low win rates can also vary widely, and this is more of a problem with related trading strategies such as visual charting and naïve trend-following. For example, although a chart pattern trader expects 2 winners in 4 trades on average, a long streak of consecutive losers can occur, and the win rate can drop to 25% or even lower. In this case, the risk of ruin increases substantially.

Below is a chart of how the win rate of the popular RIS2 strategy (first published by Cesar Alvarez and Larry Connors) has evolved in SPY ETF daily data since its inception.

blank

After a period of fluctuations, apparently at higher levels, since 2003, the RSI2 win rate has stabilized around 75% and never fallen below 72%. This is true for a strategy that has been in the public domain for more than 15 years and has several variants that are widely traded.

Below is a chart of the win rate of our proprietary PSI5 strategy in SPY ETF daily data since inception.  Note that this strategy is based on a formula from a graduate text in probability theory and is not data-mined.

blank

The win rate of the PSI5 strategy also fluctuated at higher levels above 80% in the 1990s and reached a low of about 65% in 2003. Since then, it has stabilized close to 70%.

Therefore, as was demonstrated above with two specific examples, a high win rate can be fairly stationary for extended periods. The argument made by low-win rate promoters against stationarity is, in many cases, motivated by ignorance. Note that no condition is permanent in the markets due to changing dynamics. However, non-stationarity should be a more serious problem for low-win rate strategy promoters because the impact on profitability can be more adverse.

If you found this article interesting, I invite you to follow this blog via any of these methods: RSS or Email, or follow us on Twitter

If you have any questions or comments, happy to connect on Twitter: @mikeharrisNY

Charting and backtesting program: Amibroker

Disclaimer

Technical and quantitative analysis of Dow-30 stocks and 30 popular ETFs is included in our Weekly Premium Report. Market signals for longer-term traders are offered by our premium Market Signals service. Mean-reversion signals for short-term SPY traders are provided in our Mean Reversion report.

Copyright Notice