Hi Kevin, just wanted to share something with you from my monthly trading reviews, it contains a few key lessons using your approach to monitoring.
See the attached file (shown below), it’s the equity curve for one of my Mean Reversion strategies for the ASX.
All live trades, from April 2015. The big drop in equity is Dec 2015, 2 large losses that were both bigger than any in the 10+ years of backtest history. At the time I was thinking the strategy was starting to break down.
On Jan 1st 2016 I decided to put my trade history into your spreadsheets from the workshop and boy did I learn a whole bunch of stuff, it was very eye-opening:
- I didn’t realise this strategy had actually been performing much better than expected initially (those numbers are correct – I triple checked them!) so I could have expected at some stage it would revert to the mean rather than expecting it to continue how it was,
- Those big losses in Dec brought the performance back to expected average so may not necessarily be a cause for concern, and after a few months the strategy has recovered to average expected,
- The gap between actual and perfect performance is widening, in some cases slippage is much higher than expected, some trades no longer show up in my backtests due to data adjustments plus my broker may have execution issues I’m still investigating so it’s interesting to notice those problems and try to resolve,
- My previous monitoring approach was very basic and didn’t give me any context in relation to expected average, personally I think looking at this stuff in context is huge!
Anyway, I thought it was really interesting and your worksheets have taken my performance monitoring to the next level, so thanks!
If you have comments, please send them via e-mail to me. Right now, all blog comments are stuck in approval, until my web host gets the issue solved...