Are extra-innings contests evenly matched? (Mets Game 14) April 21, 2016
Posted by tomflesher in Baseball, Economics.Tags: extra innings, Mets game 14, probability, statistics
trackback
The Mets lost to the Phillies in 11 innings last night. That was a surprising result – based on the run scoring in the first two games, the Pythagorean expectation for the same Mets team facing the same Phillies team would have been around 95.5%. Even going into extra innings seemed to be a stretch with Bartolo Colon pitching. Plus, the Phillies were in the bottom of the league in extra innings last year.
Addison Reed blew his first save of the year when he allowed a single to Peter Bourjos that scored David Lough. Despite strong performances from Antonio Bastardo and Jim Henderson, Hansel Robles allowed a double, a wild pitch, and a single that brought Freddy Galvis home.
Once we hit the tenth inning, it’s evidence that the teams are evenly matched, right? Not necessarily. in 2015, there were 212 extra-innings games. The home team won 111 of them, about 52.4%. That’s obviously higher than expected, but keep in mind that if this were a fifty-fifty coin flip we’d expect at least 111 wins around 22.5% of the time. Where it gets interesting is that the home team has (with the exception of 2014) consistently won over half those games, but that the more games that are played, the better visitors do. Since 2006, 2144 extra-innings games have been played with teams winning 1130 of them for a .527 winning percentage; that’s something that, if this truly is a 50-50 proposal, should only happen by chance 0.6% of the time.
Year | G | W | L | perc |
2006 | 185 | 105 | 80 | 0.568 |
2007 | 220 | 117 | 103 | 0.532 |
2008 | 208 | 108 | 100 | 0.519 |
2009 | 195 | 106 | 89 | 0.544 |
2010 | 220 | 116 | 104 | 0.527 |
2011 | 237 | 134 | 103 | 0.565 |
2012 | 192 | 96 | 96 | 0.500 |
2013 | 243 | 125 | 118 | 0.514 |
2014 | 232 | 112 | 120 | 0.483 |
2015 | 212 | 111 | 101 | 0.524 |
Total | 2144 | 1130 | 1014 | 0.527 |
One other result gives us pause: from 2006-2015, 24297 games were played and the home team won 13171 of them. That’s a considerable home field advantage, since all teams play half their games on the road and half at home. That corresponds to a .542 win probability for any home team. If that, rather than .500, is the expected win rate for a home team, then teams perform significantly worse in extra innings.
In other words, though the home team still has an advantage, that advantage shrinks once we hit the tenth inning.
The Mets are idle tonight. They’ll pick up in Atlanta on Friday.
Comments»
No comments yet — be the first.