Wednesday, December 12, 2012

Blood Bowl Manager

Blood Bowl Manager is an interesting stand alone program Sceadeau linked me to which basically works as a log parser for online Blood Bowl games. It then spits out a bunch of interesting stats and saves all the information online. If you're really into things you can even download logs for other people's games and replay them in the client. That's not really something I've been terribly interested in doing thus far but I can see using it in scheduled league play to scout out the opposition. That it's an option at all is pretty sweet, though.

One thing that shows up in the stat recaps in the manager is a 'chance' stat associated with each type of die roll. It can be positive or negative and my guess was that it shows how lucky or unlucky you were in terms of what you were trying to roll vs what you actually rolled. The first example we saw is shown in the screenshot below:

You may need to enlarge the picture to see. I was the Elveseseses team and I rolled 1d6 28 times in the match against Sceadeau. My distribution of rolls was pretty much as smooth as you could expect but my chance number comes in at -8.57. Sceadeau on the other hand rolled very much on the high end of the spectrum and his chance number was 10.26. My theory at the time was that the chance number being negative was indicating that even though my rolls were average when they came up wasn't optimal for me. If I needed a 4, I rolled a 3. If I needed a 2, I rolled a 6. That sort of thing. So while the end result in a 'snake and ladders' world view where only the sum matters would be about average it was actually bad for me in a Blood Bowl action sense. I want to put that theory to the test though, so I'm going to break down those 28 rolls according to what I needed to roll to see...

Need At LeastQuantityNumber of FailuresEV(Failures)
2701.17
31785.67
4321.5
511.67

So on the 28 rolls I actually took I expected to fail 9 of them and actually failed 11 of them. I was making relatively high percentage plays (I needed 2+ or 3+ on most of those rolls) so with my average distribution of results it should have been pretty good for me. But I only hit exactly the number I needed 3 times out of 28. So I was getting lots of overshooting and undershooting which resulted in fewer successes than expected. 

I still don't know precisely what the chance number means (the help files are in French and I couldn't find any relevant information on the program forums) but I do feel like my intuition for what it should mean was about right. My number was negative because I failed more often than I 'should' have. But that's what small numbers of dice are going to do for you!

No comments: