When you scale out at a .10c move, are you moving your stop (physical or mental) to break even? I ask because I changed scale-out strategies in August (SIM based) and took a huge hit (30% winrate, -$4000 on the month). I wanted to "let the runners run" and wait until it got near my price target. I was able to review the data at the end of August and noticed what was going on (what wasn't working). Like you, my numbers showed that my avg loser was way bigger than my avg winner, per trade expectancy was negative. I think the biggest (good) changes were the scale-out and the risk-based share sizing.
Obviously, each person needs to develop their own strategy, but my current strategy is this:
1) Risk Based Share Sizing (see dynamic calculator in DAS area of forum) factored for confidence (e.g. if I'm not confident, I use smaller size than my max-risk allowable). My share size is based upon the stop distance.
2) When the stock reaches 1R (if my stop distance is $0.13 away, it'd need to move in my favor $0.13), I sell 25% and move my mental stop to break-even. I repurposed the Fibonacci tool (since I don't use it) to be a 1R, 2R, 3R levels indicator.
3) During the trade, I scale out some more where I see resistance forming or near known resistance levels (moving averages, half dollar, whole dollar), or at 2R, 3R, or the price target. Using my sell 25% hotkey.
4) My mental stop tends to float, so if a moving average that has been respected recently (bounced off of) is above my break-even point, I'll watch it as an indication for a change of direction (it can go past it a little bit to allow variance) and use that as a trailing stop.
5) Once my shares get down below 100, I usually measure the last two pullback distances, add a few cents to it (as long as it doesn't place it below the break-even point), and set it as a trailing stop order. Move on to the next trade.
What really helped me refine this to my personality was a few data points that I record with every trade:
- Highest Price in Trade (this is price related, not direction related, so highest price seen for either long or short)
- Lowest Price in Trade
Those aren't prices you executed at, but prices that the stock reached while your trade was open. From there, I can calculate the average R-movement, updraw%, and downdraw% for every trade. When I reviewed August's losses, the data told me that: 78% of my picks moved in my favor. I was simply not scaling appropriately and letting winners run against me. My position sizing and $risk was all over the place.
For September, I implemented the new strategy listed above. Winrate is 76%. It has a lot of "small" wins.
I'm working on releasing a few custom tools to the community to help people narrow down their strategy and refine their edge. One is a data-focused journal (does most of the work for you) and the other is a backtester (allowing you to automatically replay all of your trades with different scale out approaches you want to test).