Tuesday, March 15, 2011

What Does Spring Training Mean?

Every year, baseball teams pack up and head south for the spring. Every year, fans across the country dream about what could be--that young prospect who could bust out, the crafty veteran who could pull the team together. And, every year, a few teams have springs bad enough that those same fans are ready to give up entirely before the season even starts! So, that got us thinking here at Sports, Stats & Stuff about what the preseason really means, if it means anything. Do playoff teams perform better in the spring than non-playoff teams? Which teams tend to under-perform, and which teams over-perform? At what point should I begin pulling my hair out? If these questions interest you (and if they don't, why are you still reading?) hit the jump!

As it shockingly turns out, over the last five years, the average change in winning percentage from preseason to regular season is....0! Of course, the average of any set of opposing positive and negative numbers are going to turn out to be zero, so the next obvious point to go to was: how much does the average team change? As it turns out, when you take the absolute values of the percentage change, the difference comes out to 8.7%, or 14 games. In other words, a team that performs at .500 in the spring (extrapolating to an 81-81 record) could instead easily go 95-67...or 67-95. In fact, not only is it possible, that's the average.

This is basically how the White Sox's springs go
Some teams have had even larger discrepancies between pre-season and regular season performance.  The Chicago White Sox have had the largest such difference, with a 13.6% average change from Spring Training to regular season winning percentage.  That's a difference of more than 22 games!  A .500 White Sox team in Spring Training could on average go anywhere between 103-59 and 59-103.  What's amazing though, is that 13.6% average isn't an absolute value, it's a real one. The Sox are pulling a .379 spring winning percentage, which would translate to only 61 wins. Instead, the Sox have won an average of 83 games in the same time span. On the flipside is the Sox' division rival Minnesota Twins.  Their average pre/regular season difference is only 4.3%.  As such, a .500 Twins team in Fort Myers, FL would only likely end up between 74-88 and 88-74 - a fairly small window compared to other teams.  So the lesson for Twins fans is the opposite of the lesson for the South Side Sox: your spring training record matters, at least a little bit.

False hope, meet "too good to be true"
When you take out absolute values, we do see that a few teams have had bad springs and good summers regularly over the last five years. The Phillies have improved by that same 8% on average each year, going from a .475 spring club (77 wins) to a .563 summer one, winning 91 games a year. The Dodgers improved by a slightly larger margin of 8.8%, also a 14-game discrepancy. We already discussed which team has the most deceptively poor spring (see: Chicago White Sox, one paragraph ago), but there is one squad that gives more false hope than any other: the Kansas City Royals. Each spring, the Royals have averaged a .525 clip--on pace for a nice 85-win season. No, it wouldn't have been enough for a division title, but staying competitive can do wonders for attendance, which helps entice free agents, which brings more wins (See: Philadelphia Phillies, 2004-2008). Instead, the Royals went out and averaged around 68 wins a year (.417 winning percentage), and they're exactly where they were five years ago. Other teams that tend to overachieve during the spring? Cleveland (-6.6%, or 11 games worse), and the St. Louis Cardinals (-5.4%, or 8 games worse).

Wherever your passions lie, we do have one ironclad piece of advice for you: extremely positive or extremely negative results during Spring Training are almost always a fluke.  In fact, while the standard deviation of all Spring Training winning percentages since 2006 in 0.106, the standard deviation of regular season winning percentages is little more than half as much: 0.064.  This means that winning percentages are much more likely to tend towards the extremes than towards the average (.500) during Spring Training compared to the 6 best months of the year.  For example, the 2009 Angels had the highest Spring Training percentage of any team in any year since 2006, at .765; however, the highest regular season winning percentage for any team during that same time period was .636 by the 2009 Yankees.  This discrepancy amounts to a 21-game difference in regular season wins., meaning that incredible pre-season results can hardly translate into the regular season.

This same tendency exists on the low end, as well.  The 2010 Pittsburgh Pirates had the worst Spring Training since 2006, at .250, and they also had the worst regular season, at .352.  Though the 2010 Pirates were abysmal both in both March and in April-September, their regular season winning percentage was a 16-17 game improvement over their Spring Training one.  What all these numbers mean is this: if your team does incredibly well or horrifically bad during Spring Training, don't celebrate/pout yet.  Regular season numbers tend towards the mean (.500), and your teams' actual performance likely won't even be close to that good/bad.

So folks, with about 2 weeks to go in the spring, we say this loud and clear: DON'T PANIC. If you're a Phillies or Dodgers or White Sox fan, you should feel especially good...and if you're a Kansas City fan, well, good luck with that this year. Then again, they're only averages...

No comments:

Post a Comment