For now, I just size based on the probability that I am wrong
This is IMO the best approach. The goal of my money management is to ensure my max DD target has a very low (<5% per year) probability of being exceeded. Hit rate is pretty much all that matters here, because noteworthy drawdowns by definition occur when you are having long streaks of consecutive losers.
I will double the bet size a couple of times a year when a fat pitch comes along, where market action, sentiment, and fundamentals all line up perfectly and the payoff is highly skewed.