Algorithm Performance: 08/09/2022
Performance Rankings
FeelGoodInc: +0.05%
The Neural Net: -0.02%
Long Term Portfolio: -0.08%
Trading Bot: -0.12%
The Market: -0.21%
Base Algorithm: -0.23%
Kokomo: -0.47%
GoodVibrations: -0.54%
What Happened And Why?
I’ll be up-front here. I just had this whole report written out, but my computer crashed before I could get it posted. It’s nearly midnight, and I don’t have time to re-run the programs I need to recreate the graphs and data I was referencing, so I’ll have to give a quick summary:
Today was a bad day for us. Our 3 main algorithms underperformed the market, and our “safer” algorithms underperformed harder
One day doesn’t a failed system make, but it’s worth looking into
If we break returns down by signal, Kokomo and GoodVibrations lost money trading with Pivot Points and Heiken-Ashi candles, whereas the Base Algo made money with both. This suggests to me that these signals have their power diluted when applied to price history controlled for outside factors, as opposed to raw price history
FGI’s good performance today supports this hypothesis, since it (like the base) uses raw price information to calculate trading signals
I’d like to do a backtest segmented by signal, to see if this persists over periods of time, as opposed to today being a fluke. This test will probably have to wait for the weekend though
Tomorrow’s Outlook
The full algorithm reports will be published tomorrow morning, once Allen has had a chance to vet its recommendations. In the meantime, here are our tentative exposures for the trading day tomorrow:
Thank god I at least had this one saved before things crashed
That’s about it for tonight. Let’s hope for better luck tomorrow. In the interest of full disclosure, Allen and I intend to sit out tomorrow, due to CPI numbers coming out. I think our hedging systems will withstand the news well, but it’s tough to say. Good luck trading everyone, and thanks for reading.
-Asher
