February was pretty slow and I decided to paper trade for almost all of it so I could work on the algo some more. Small caps were so incredibly dead that it was rarely watching more than 15 stocks by 10:30 AM. I decided to increase the market cap from 1B up to 10B and eventually 15B just to see what it would trade.
A lot more trades were made due to this and I found a few problems.
- Low volatility stocks were being traded even though the ATR was good on all the candles. I added a percent ATR filter which seems to have helped.
- Risk/reward was poor on many setups so I added a R/R filter based on the stop and profit target. This actually filtered a lot of losing trades out. I decided on a 1.25:1 reward/risk. 2:1 or higher is ideal but sometimes you just can’t get that based on the stop and target. There are some where 1:1 is the best you can do and they work out perfectly.
- Since this algo only trades flags, the high candle needs to be above most of the previous 5-6 candles. I previously had checks in place for this but they were not functioning the way I wanted. I made a small change to skip the last 3 candles instead of 2 which seems to have worked. This is very difficult to test because I do not have many examples.
- Lastly, new exit indicators. I have double tops and topping candle checks, but those only cover certain scenarios. I added a new double top check that goes back multiple candles instead of only checking the last two. Another scenario is what happens when the stock misses the target by just a few pennies without blowing through it. It should probably sell it, right? That happened on $ZSAN on the last day of February (although it was not traded due to the bad R/R).
I am not a registered financial adviser/broker/anything. Use this information for entertainment/informational purposes only. Any tickers mentioned are not recommendations to buy/sell/or sell short. They are used as examples only.