Testing scenarios

I have been doing a great deal of testing of the scenarios that I have recently updated for the First Anglo-Dutch War. The most dramatic change is for the Battle of the Kentish Knock. There, the simulation plays out more like the real battle, so that the Dutch are competitive. The worst outcome was for the Battle of Portland. For whatever reason, George Monck's squadron is able to easily sail upwind to engage the Dutch. In reality, that was much more difficult and time consuming. I need to go back and study why the simulation is different from the battle.