Randomly skipping some trades

Trading provides one of the last great frontiers of opportunity in our economy. It is one of the very few ways in which an individual can start with a relatively small bankroll and actually become a multimillionaire - Jack D. Schwager.

I was having a discussion with tech/a over at ASF the other day. He uses discretion (eyeballing the chart) to choose which candidates to buy from the stocks that have triggered. I thought that's fine -- until I found out that sometimes he doesn't trade at all because the candidates don't pass his filter.

Initially I was thinking that this was playing with fire because such a strategy cannot be backtested with TradeSim. I mean, having 10 candidates and picking any 4 because you have capital to take 4 trades is fine. Discretion can be applied here and shouldn't matter which 4 we pick as this is what monte analysis takes into account -- the different portfolio combinations and permutations.

It has been brought to me attention that TradeSim actually can test for this. This post was inspired by some good work by Stevo. You can actually program a code as part of the EntryTrigger and tell TradeSim to randomly only take X% of the trades that trigger.

So I did just that, to see if my system held up even if I missed some trades. And it does quite nicely. That said, I don't plan to ever do this in practice. You never know which trade will be the big winner.

For the above test, market orders were used to account for slippage and 200 monte carlo simulations were run each time, with the average return tabulated and shown in the above chart.