The factors you are trying to measure changed over time For example Submarine COs aggressiveness changed with ther replacment of combat experienced commanders. There are books already published on the subject. (I'll get back to you with the title after the election) Changed for the Americans with experience in effective AASW etc. What I am telling you is that historians have alredy got better data than lumping whole war data into one study. The statistical theorists are going to throw out your measures of effectiveness (as expressed here) as inappropriate statistical measures. You should get a good book on statistics if you are serious about it though a more pleasant read is probably the old text "Operations Research." Pleqse don't get me wrong I want a meaningful end for your work. I share your interest but I want it to be new and relevant, especially since I know how much time it will take.
You should study the AWSORG records groups at the National archieves as a starting point.