Statistics in sports I am trying to come up with a topic for my bachelor's thesis and I was wondering if anyone wrote a similar piece of work. My idea is to describe various uses of statistics in sport(s). I really liked movie Moneyball and a topic based on this connection of Maths and a sport is exactly, what I am aiming for. Any suggestions or already written piece of work on this subject?
 A: Here is an outline of one idea among very many possibilities:  
Theoretical: Assume all teams in a league are equally good, so that any one team has a 50-50 chance of winning each game. Find the avg number of 'runs' of wins in an MLB season for a team in such a hypothetical league. (Standard result, look it up.) 
Also, find the expected length of the longest run of wins in a season. (May have to simulate this. In R, there is a pre-programmed function rle, for Run Length Encoding, that would be helpful.) 
Applied: Get actual MLB data for runs of wins and losses; compare with theoretical.
Of course, variations are possible. I believe the version described to be
feasible. It is important to get a project that can actually be completed
in the available time.
It seems to me that a satisfying bachelor's thesis on sports should have
a mix of (accessible) theoretical results and (available) real sports data.
Also, some statistics magazines (e.g., 'Chance' and 'Significance') have
articles that may suggest projects. Journal of the American Statistical Association (JASA) sometimes has sections on sports statistics. Student
memberships to ASA (www.amstat.org), Institute of Mathematical Statistics (IMS) (www.imstat.org), and the Bernoulli
Society are quite affordable and provide access to magazines and journals.

Teaser code -- One team, one season:
 > n = 162;  WL = rbinom(n, 1, .5)
 > WL.rle = rle(WL)
 > WL.rle
 Run Length Encoding
   lengths: int [1:78] 3 3 1 1 1 2 1 6 1 2 ...
   values : num [1:78] 0 1 0 1 0 1 0 1 0 1 ...
 mean(WL.rle$lengths[WL.rle$values==1])
 [1] 2.153846
 > max(WL.rle$lengths[WL.rle$values==1])
 [1] 9

