Spring Training Statistics Mean Nothing

Spring is a season full of optimism. The greening of the grass, the warm rays of sunshine against your face, and the sound of that familiar crack of the bat. These indelible reminders are what make spring such a great time of the year.  Unfortunately, people mistake this time for optimism as a poor justification for reckless aspirations. It continues to amaze me how people fall for the same old song and dance every spring. Perhaps I shouldn’t be surprised. Perhaps it is just my inherent cynicism. But every year, people mistake spring training success as a predictor for imminent achievement and accomplishment.

March 31, 2012; Port Charlotte, FL, USA; Boston Red Sox right fielder Darnell McDonald (54) singles in the third inning against the Tampa Bay Rays during their spring training game at Charlotte Sports Park. Mandatory Credit: Kim Klement-USA TODAY Sports

Unfortunately, the general public puts way too much stock in spring training statistics when assessing a player’s future performance. I understand it, I really do. People simply want their team’s players to do well, and spring training success acts as a confirmation bias of sorts.

For example, a player on your favorite team hits .400 and suddenly, he’s primed for a breakout season. Or maybe a career journeyman catches fire, and is now expected to be  a legitimate starter for a club with playoff aspirations. Perhaps a strong spring justifies the big contract given out to a player the previous winter. Yet, when the games actually begin to MATTER, some players fail to live up to the pre-season hype. It’s not that the players succumb to the pressure of playing major league ball. These guys have been playing baseball their entire life; playing games that matter shouldn’t alter their psyche or approach. It’s the idea that expectations based on incredibly small sample sizes are meaningful at all.

In fact, all of the above examples were real cases of 2012 spring training success stories but regular season disappointment. Eric Hosmer, following a superb rookie campaign, hit .398 and led the circuit with 29 RBI in the spring. When it mattered, he hit .232 and saw his power disappear, hitting only 14 home runs and slugged .359. Longtime journeyman Darnell McDonald tried to make his spring performance a launch pad for his career, hitting .427/.512/.816 in eighteen games. However, once the regular season began, McDonald remained the same old McDonald, and was later designated for assignment in June. Additionally, Albert Pujols hit .385 and had a league leading 7 home runs following the inking of his mega-contract signed earlier that offseason. While Pujols was still productive, it was the worst season of his career and certainly did not match the expectations produced from his contract and productive spring.

Incredibly, year after year, people continue to fall for the farce of spring training statistics. Statistically speaking, there is practically no evidence linking spring training numbers to regular season production. Using the coefficient of determination (r²), a number’s correlation to other data can be determined. It may seem a bit math-y, but it is really quite simple.  An r² value close to zero (0) represents data that has practically no correlation while values closer to 1.0 show data that greatly supports and correlates with the data. I know in Earl’s piece, he stated there was an r² value of .21 for spring training statistics, but more recent data suggests even less correlation. In a piece from Beyond the Box Score, Lance Rinker found that a hitter’s average, on-base, and slugging numbers had r² values of .034, .002, and .052, respectively. This correlation is as close to meaningless as possible, and further proves the lack of predicative value spring training numbers provide.

Just as Earl perfectly detailed in his piece, spring training is full of fallacies. One such example was that of the young phenom, who is prematurely labeled as baseball’s next big thing. Sure, while there are guys like Mike Trout and Bryce Harper, who blow past every expectation imaginable and truly live up to the hype, most players are not Trout and Harper. Unsurprisingly, following the seasons put together by last year’s rookies of the year, people are already clamoring for baseball’s next young star.

To many, that next young star is Jackie Bradley Jr. Bradley has hit a robust .484/.590/.645 this spring with plenty claiming he should start the season in the majors. Don’t get me wrong, I am plenty excited for Bradley to be in a Sox uniform. It’s just that rushing him to the Majors because of a couple of spring training at-bats is short-sighted and lazy. First off, the Sox have no nowhere for him to play. A little more seasoning never hurt anyone and would almost certainly benefit from playing every day. Furthermore, it is unfair and ridiculous to think that just because Bradley has crushed spring pitching, it is a foregone conclusion that he will perform at the Major League level. It is common knowledge that pitchers specifically work on things in spring training and quite honestly just do not give 100% in effort to stay healthy. Lastly, the sample sizes seen in spring training numbers are way too minute to credibly give any definite conclusion regarding a player. Just like a coin flipped twice might land twice on heads, it doesn’t mean there is an 100% chance of getting heads; it is simply too small of a sample.

So please, stop quoting spring training statistics and act like they mean something. They don’t.