PnL-sharpe-based risk allocation is a step ahead of PnL-based risk allocation. It uses the average PnLs normalized by historic standard deviation of returns to penalize trading strategies that have large PnL swings, also known as very high volatility returns.

This allocation method solves the problem of avoiding the construction of a high-volatility portfolio. But it still does not account for the correlation of returns between different trading strategies, which can still end up causing us to construct a portfolio where the individual trading strategies have good risk-adjusted PnLs but the portfolio as a whole is highly volatile.

The trading strategy with the best performance still makes the most money, similar to what we saw in the individual PnL-based allocation. However the other trading strategies still get a decent portion of the total allocation amount. This is because when we factor for risk in our allocation method, even strategies that make a lot of money don't necessarily end up with large allocations because the volatility in their returns also increases with their PnLs.