A good observation. The reason why OptimalF reduces from 0.25 to 0.23 in uncorrelated events is that you toss the two coins at the same time. If you would toss them a time apart, the capital change from the first toss would affect the investment in the second toss and therefore the OptimalF factor would be again 0.25.

In Workshop 6, a coin toss is equivalent to opening a trade. Since trades are normally opened at different times, it's a situation more equivalent to uncorrelated single coin tosses than to multiple simultaneous coin tosses. The analogy is not perfect because trades overlap, but due to the 0.5 factor we don't overrisk.