Hi Bruce,
It is very difficult to determine in real time what kind of bear market will develop in the near future.
E.g. the COVID crash was short and steep with a V bottom reversal. The Stock Filter being ON through Apr - June 2020 would have greatly delayed re-entry back into the market and significantly reduced the ensuing profits.
The 2022 bear market was long and gradual. The Stock Filter being ON in research showed a significant enough reduction in drawdown for this period because the delayed signal entries meant trades were filtered out. Ultimately, all filters delay or prohibit entry.
If a system has an edge over a large sample of researched trades through many different and varying up & down market conditions, then that edge is what a trader has to place their trust in.
The problem that all traders have to some degree is that the uniqueness, and hence uncertainty, of EACH bear market causes a degree of discomfort that hovers over them all the time. Even while the market is rising and going gangbusters as they try to anticipate what kind of bear the next one will be.
As you know, this discomfort is psychological in nature. Most traders try to solve this discomfort in advance through some analysis technique they can apply now that will hopefully reduce (most subconsciously try to remove) that anticipated discomfort.
As you know from your own trading experience and The 4-Week Training Intensive, it is impossible to analyse away discomfort. The first step in the solution is to completely accept the risks that are inherent in the market.
This requires a reshaping on how a trader thinks. When we truly think from the market’s perspective we come to completely accept ALL the risks, including each unique type of drawdown that we’ll experience in the future. This is what makes the discomfort go away. It is truly liberating.
That said, we do need to determine, set, apply and trust rules that govern our actions so that we are objective, confident and consistent. When to apply the Stock Filter and when not can be one of those rules.
This can be monitored by running research from time to time on new out-of-sample data as it emerges. How much data? Enough to produce a large enough sample of trades to be statistically significant and that will provide an observational difference. Which is what we did last month with around 18 months of data.
The analytical problem of course is that there will always be a delay.