What's in a Preseason?

I, for one, actually like watching preseason games. I like the chance to see new guys. To see how guys look after the offseason. To watch a game being played where the end result doesn't matter takes away that element of... I don't know, stress? and let's me focus on just enjoying the fun of the game. Plus it's always nice to be able to feed the beast after months of not getting to watch my local guys.

I got to thinking though, and wondered just how much you can judge a team's quality based on their preaseson performance? Is there any relation between how well a team does in preseason and how well a team does over the course of the true season?

Short answer: No. Long answer? Noooooooooo.

Here's how I looked at it.

First, I tallied up all the results for each team in the 2013 preseason against all opponents and compared their average result to their average results for the regular season. There was absolutely no correlation whatsoever. Here's a graph to illustrate.



So surficialy there seems to be no relation between preseason and regular season results for the 2013 season. Is this because preseason matches are padded with the likes of playing against foes like the USMNT U17s and FC Tucson? It doesn't seem to be. I pulled out only results for MLS vs. MLS teams and the results look basically the same. Here's that chart.



Is there maybe a looser connection? Could an "above average" preseason correlate with an "above average" regular season? I checked to see if preseason and regular season above-or-below-averageness correlated. That is to say, how many teams finished the regular season above average and also had an above average preseason? And vice-versa.

In 2013, 8 teams had correlated preseasons and regular seasons regarding above or below averageness. In 2012, that number was 9. So out of 19 teams, that's about the number we'd expect if there was truly no relation. This held true for both all results and MLS only results.

How about if we simply look at games played? Do teams playing a higher number of games do better than teams playing a lower number of games? Perhaps at this point, unsurprisingly, the answer is, no, there does not appear to be any correlation between the number of preseason games played and the results of the regular season.

These results were the same for both the 2012 and 2013 seasons, both considering all preseason matches played for all MLS teams and only MLS team vs. other MLS team results. I've been trying to think of other ways to dice it up, but I think I've more or less convinced myself that, while preseason matches are an important means for a team to gain fitness and build chemistry, as well as try out new things and sort out who makes the final roster, the actual results of these games means very little. One should probably be careful before getting too excited, or deflated, based only on preseason form.

Maybe that's not surprising, but I thought it was interesting to see just how little evidence there is that preseason games can be used as any sort of predictor for the regular season.