Attendance Analysis (Pt. 5): The Power of Success

Man, I'm such a slacker.  It's been awhile since my last post on attendance (which I know you are all waiting for with baited breath), but thankfully PlayOnWords and thebaddancingraysfan have stepped up and raised some very good questions in their respective Fanposts (here and here).  How much of an impact will Obama's high speed rail have on the Rays?  How much of the Rays' attendance issues are to blame on the stadium's location and the lack of quick, accessible public transportation?  I'm not from the area and can't speak from experience, so please check out the discussions in their posts if you're interested.  In my opinion, though, I believe that the Trop's relative inaccessibility does make it tough for fans to attend weekday games and it probably limits the number of Tampa businesses that have season tickets, which the Rays seem to understand themselves already.  To what extent?  That's tough to say, but I'd argue that it's an important factor but not a crippling one.  It's not the be-all-end-all solution that people in support of a new stadium seem to be claiming, but I do think it's having an impact nonetheless and shouldn't be overlooked.  And if a new stadium doesn't seem to be in the cards anytime soon, why don't the Rays pursue improving the Tampa-St. Pete public transportation system?  That's just a discussion idea, so have at it.

Anyway, down to what I promised I'd talk about next: winning.  How much of an effect does winning have on attendance figures?  I mentioned this in an earlier post, but team attendance figures (as measured by Att+) correlate strongly with team winning percentage.  Here's the breakdown:

Okay, that's a bit confusing, so let me explain.

I've taken a league-wide sample and included every season back to 1961, trying to determine how much of an effect a team's winning percentage in past seasons can have on a team's attendance this season.  What you're seeing are the correlations between a team's winning percentage in Year n, n-1, n-2, etc. and that team's Att+ figures in Year n.  For example, say Year n is 2009.  The correlation between 2009 Att+ and 2009 winning percentage is .523 (strong), the correlation between Att+ in 2009 and winning percentages in 2008 is .449 (medium-strong), the correlation between Att+ in 2009 and winning percentages in 2007 is .344 (medium-to-weak) - and so on.  In other words, the farther back in time you go, the less and less it matters how a team has performed.  Fans do have memories, but mostly only of the past two seasons.

That seems like really good news for the Rays, right?  We've been competitive for the past two seasons, so maybe that means fans will have forgotten about the past and we'll see an even better turnout this year.  That very well might be the case; I don't really know.  The most important thing to take away from this data, though, is that the best predictor of attendance is how well the team is performing now.  I tried creating a weighted Win% average of the past couple of seasons that would correlate even stronger with a team's Att+ than their current Win%, but I couldn't do it.  Maybe I was simply screwing up (which is entirely possible), but no combination that I could come up with correlated any stronger with Att+ than this season's winning percentage.  If the Rays want to draw even more fans this upcoming season, they need to win and they need to be in the race for the playoffs.

Of course, that's common sense.  Fans support a winner!  But in this case, common sense is actually right.