Size matters.

Matt Melton does a statistical take on something that should make you slap your head and wonder why nobody else has done the same thing before (and apologies if you’re that someone and I’ve failed to find you):

… I decided to take a look at how crowd size affected home field advantage in 2010. I didn’t look at any particular telling stat (penalties, yards, touchdowns, etc.), I simply looked at whether or not the home team won, and then what the crowd size for that game was. I then, very unscientifically I might add, divided the crowd attendance into ranges of 10,000 and calculated the winning percentage for each range. The results are summarized in the table below.

Now before you start jumping up and down, Matt acknowledges there’s a certain chicken-and-egg aspect to this.  The better programs tend to be bigger draws.  So that you see winning percentages climb in almost-virtual lockstep as home crowd size increases may be attributed to either factor.  (There’s also sample size as an issue, something that’s often a problem when you’re looking at a statistical base being generated by only 120 participants.)

But here’s where it gets interesting.  Matt then filters his results through the point spread and finds that there’s correlation up to a point:  “As the crowd size inched north of 50,000 or so, the added fans provided no real boost to the home team’s hopes of covering the spread.”  And it’s his last chart that’s worth a look.

That’s only the data from one season, but if that’s a breakdown that plays out as a trend, he’s right.  Appreciative gamblers everywhere ought to tip their caps in his direction.

About these ads

8 Comments

Filed under Stats Geek!

8 responses to “Size matters.

  1. TennesseeDawg

    Man, it’s been a long offseason. :)

  2. Go Dawgs!

    “As the crowd size inched north of 50,000 or so, the added fans provided no real boost to the home team’s hopes of covering the spread.”

    Georgia Tech sighs in relief.

    • FisheriesDawg

      Speaking of Georgia Tech, is there any way to dig deeper into programs that have nearly half opposing fans in the stadium when they actually fill it?

  3. Joe

    Boy, UT, PSU and Mich sure hurt the over 100K stat. Add OSU’s vacated wins very soon and that could be a bad number!

    • Macallanlover

      Vacated wins….right. We saw how that worked with GT (and I hope you aren’t holding your breath with The aU from last season.) NCAA has allowed the horse to leave the barn when it comes to respect.

  4. Mayor of Dawgtown

    In a related development scientists have also confirmed that water is wet.

  5. TrboDawg

    I wonder if we could see the same stats as a percentage of capacity. IE if the stadium can hold 50,000, but only 20,000 show up…