The question of an edge | Traders Hideout


futures.io - futures trading strategies, market news, trading charts and platforms


Traders Hideout


Discuss day trading practices and futures trading strategies on this forum for all markets. This forum is also for discussing and reviews for brokers, data feeds, and commercial or third party add-ons




 

The question of an edge

  #37 (permalink)

North Carolina
 
Trading Experience: Beginner
Platform: NinjaTrader, Tradestation
Favorite Futures: es
 
Posts: 644 since Nov 2011

Let me summarize why I think might be why there are different views of expectancy. If you trade a system, trades are generated based on a consistent set of rules and taken in a consistent way. As such, expectancy is used both as as the predictive statistic (an extrapolation of what the future will look like) and as a descriptive statistic of the past (what actually happened).

However, attempting to trade like a system for a discretionary trader would be both nearly impossible and also nonsensical because to do that you would need to generate your fixed set of rules and then you would need to trade them in exactly the same way with no variance. If you were to do both of those activities though, you are now just trading a system and there is no longer any discretionary component. As such, expectancy takes on a different meaning for the discretionary trader. The rules are no longer fixed: so it is not a validation of the rules but rather of the cognition of the trader. As many more factors can effect cognition, those other factors can become more relevant. However, market cognition is still probably a better place to invest your energies then in general psychology.

Mathematically, it is known that models with more variables are more likely to capture complex phenomena like markets exhibit. However, introducing more variables, into traditional systems most often results in two problems (1) inability to understand the model and (2) over-fitting. We know that the discretionary trader's model should be more complex. As such, if the discretionary trader sees a verifiable truth in markets (such as markets trend or markets trade in a range) then they assume that they are trading closer to reality. A method closer to reality is more likely to be robust. It is more likely to capture the truth of the markets. The cost for trading a more realistic method though, a more robust method, is that it becomes more difficult to bound the strategy at the extents. This translates into not knowing how much to bet and not having as much clarity into the model. Simplifying the model to produce a consistent set of rules, will produce a characterization of reality but the benefit is that we know exactly how the model behaved in the past. We can now gain insight into how the strategy is working and, at least, we can figure out under given scenarios the historical risks of betting varying amounts. Unfortunately, the cost is that because the model has been simplified, the model may not work as well going forward and while we know our historical estimates of risk: they could be wrong. Ironically, most discretionary traders focus on consistency but rationally we should expect both greater variance in discretionary trading and enhanced robustness. On the other hand, we should expect trading systems to exhibit greater consistency but higher rates of failure.

Reply With Quote