*This is an update to a previous study on this topic last month. The original results were forwarded to Ben Lindbergh of The Ringer, who kindly provided his own insights and suggestions to help improve our study. If you would like to view our initial findings, you can find them here. You can also check out Ben’s article at The Ringer.
Each year, hundreds of players from the MLB and other foreign leagues depart from their clubs during spring training to participate in the World Baseball Classic. As a tournament that takes place every four years, the players surely see it as an incredible opportunity to represent their country and play alongside other world-class caliber players from the same nation.
However, their MLB teams simply see it as a month where they don’t have full control over their players, and as a high-intensity environment ripe with opportunity for injury. Spring Training is a time for players to ease back into competitive play. Hitters play a handful of innings each day before minor leaguers replace them, and pitchers rarely exhibit maximum effort in their shortened spring stints.
On the other hand, in the most recent edition of the World Baseball Classic, we saw pitchers hit velocity marks they’ve never come close to, and relievers being pushed with 2-inning outings and high-leverage appearances on back-to-back days.
The 2017 edition of the tournament isn’t even a month old, and teams are already feeling the repercussions of their players’ WBC participation. Drew Smyly, Seth Lugo, and Didi Gregorius have all hit the DL with injuries of questionable relation to the WBC itself.
The question that is on the mind of every fan and every employee of a Major League franchise remains: how big of a risk is participating in the WBC?
To begin to answer this question, we first must acknowledge that we’re working with a unique group of participating players. In order to be selected to compete in the WBC, a player likely meets these two criteria:
1. Player ended previous season in full health
2. Player has demonstrated a level of performance that justifies selection to national team
These two criteria alone rule out a significant portion of MLB players. In the 2016 MLB season, there were 31,662 days lost to the disabled list, and that statistic has been steadily rising for the last ten years (Conte Analytics).
The first criterion is the most important for our purposes, as it implies that our sample of players is, generally speaking, healthier than any other random sample of MLB players. That being said, we can’t simply compare players who participated in the WBC to those who didn’t. Instead, we must compare these WBC players against themselves.
The crucial year for injury evaluations is the one immediately following the WBC tournament. Any substantial conclusions that we come to must find that there is a spike in injuries in this season compared to other seasons. To determine this, we compared a player’s “WBC Season” to the 3 years preceding their “WBC Season” and the 3 years following their “WBC Season.”
As an example, Robinson Cano was a participant (and the MVP) in the 2013 WBC. For Cano, we gathered data on his injury history in the 3 years preceding 2013 (2010-2012), 2013 itself, and the 3 years following 2013 (2014-2016). With Ben Lindbergh’s help, we were able to upgrade our previous study – one only focusing on DL stints – using a more comprehensive database from Baseball Injury Consultants. Baseball IC provides data on every bump and bruise a player suffered, whether it be in the majors, minors, or spring training.
While our original study removed injuries classified as “freak” (think breaks, sprains, etc.), Rob Arthur of FiveThirtyEight has found these sorts of injuries not to be as random as one might think. Including them in this updated study also removes the subjectivity of deciding which injuries to classify as such.
For any study that involves injuries, it’s important to recognize how aging affects a player’s likelihood of being injured. To take this into consideration, we assumed than a given player, when younger during their pre-WBC seasons, would miss less time due to injury. On the flip side, that same, now older player in their post-WBC seasons would miss more time due to injury. In our research, we operated under the belief that these pre-WBC “healthier” days would be offset by a player’s post-WBC “less healthy” days for a number comparable to the WBC season on its own. In other words, an average of seasons 1-3 and 5-7 would generate a number on par with season 4 in terms of aging.
Putting that all together, we compared the average “WBC Season Days Missed” to an average of the “Non-WBC Season Days Missed,” which was calculated as shown below
Right off the bat, it appears our original suspicions have been confirmed. WBC participants, regardless of position, miss 2.35 more days due to injury on average following their WBC season.
Where it gets really serious, though, is when you narrow it down to just look at pitchers. In a non-WBC season, an average WBC pitcher misses 23.55 days per season from injury – a pretty low number when you think about all of the Tommy John surgeries and mysterious dead arm periods. In that WBC season itself, the number jumps to 27.62 days missed per season from injury. Here’s a visual look at the 4.07-day difference:
Delving a little bit deeper into individual WBCs, we see the 2009 tournament as a huge culprit for much of our results. In 2009, pitchers missed an average of 10.08 more days due to injury, while hitters found themselves missing an average of 5.69 more days. For a tournament by tournament visualization, see below:
More often than not, a WBC participant found himself more likely to suffer an injury in the season following the tournament. How does this compare, though, to a non-WBC participant? We already mentioned that players selected to the WBC are generally much healthier than an average MLB player, so what happens if we were to take a control group of healthy MLB players?
Lindbergh provided a list of almost 200 players who spent between 10-20 days injured in the season before a WBC. The results for these players were particularly stunning. See below:
This casts a huge shadow of doubt on the whole study, but we believe to have an explanation for the findings. The given control group only takes into account a single season before the WBC. The players included in this group exhibited excellent health – 10-20 days missed due to injury – in that single season. However, we’re comparing them to WBC players who, on average, missed 22.41 days per non-WBC season over six total seasons (3 before, 3 after the WBC).
What I’m getting at here is the (not so far-fetched) idea that perhaps the control group is comprised of players experiencing a season of outlier health in the pre-WBC year, while the WBC participants experience their season of outlier health in the WBC year. Perhaps the control group is simply regressing to the mean, which wouldn’t be all that surprising given their short track record of good health.
With all that said, it will be interesting to take a look at this 2017 WBC with 20-20 hindsight. Just how indicative were the injuries to Smyly, Lugo, and others of the risks of the WBC? Our findings, while nowhere near conclusive, may raise some eyebrows in front offices around the league. Come 2021, we’ll see if this risk can outweigh the immeasurable risk of upsetting a star player by denying him the right to represent his country in the next edition of the World Baseball Classic.
Research assistance by Ben Lindbergh