The dangers of playing second base

I don't know how often I've heard a variation on the following phrase, when parsing the host of baseball articles I've read recently:

Second base is a more "dangerous" position than [position X].

OR

[Player X] wouldn't be moved to second base, because that position presents more of an injury risk.

The basic idea behind these quotes -- the idea that second base is a more dangerous position than, say, third base or left field -- has been around for years. From a common-sense standpoint, I'm not sure it's wrong. I mean, the act of turning a double play and presenting yourself in position to receive take-out slides seems pretty dangerous.

So today, we'll start to run some early data, and see what turns up.

My goal here is to see if second basemen spent more time on the DL, or had more DL stints than other positions in or around the infield. Because I am incredibly lazy limited in the amount of time I can spend on this, I'm going to work with a simple two-year sample of DL stints: 2010 and 2011. All my data is coming from Jeff Zimmerman's great databases (2010 and 2011), so big thanks to Jeff for making this information available. Jeff's 2010 data doesn't specify which position other than IF or OF for most players, so I went through the data and gave the player credit for whatever position they spent most time at during 2010.

Here's the data I came up with.

While I'd like to pose this caveat -- that this data, like most injury data -- is a little murky, due to the subjectivity of injury data, I think this is an okay data set. We lack some specific, useful information (such as the frequency of injuries that took place while fielding the position, as opposed to under other circumstances), but let's work with what we've got.

So, if we start with 2010, what do we see? During that particular season, second basemen seemed to have the same number of DL stints as shortstops, and just two more than third basemen. Catchers and left fielders both had more stints on the DL. When it comes to days on the DL, we see that second basemen have quite a bit more than their infield brethren, with 1223 total days, and no other non-catcher infield position has them beat. Left fielders and catchers had more DL days, and center fielders are pretty close. Verdict: For 2010 second basemen were injured for a longer period than 1B, SS, 3B, and RF, are pretty close to CF, and were injured less often than LF and C.

In 2011, things get a little bunched up. In regards to DL stints, third basemen had more, and shortstops had only two fewer. Also catchers, center fielders, and left fielders also had more stints on the DL than second basemen. As for DL days, second basemen lead third basemen and center fielders, but the numbers are awfully close. Left fielders aren't too far off, while catchers dwarf all positions, and everyone else is lower down the injury totem pole. Verdict: For 2011, second basemen were injured for a longer period than 1B, SS, and RF, are pretty close to 3B, CF, and LF, and were injured less often than C.

Finally, the overall numbers over our two-season sample show that second baseman had the exact same number of DL stints as third basemen, and only one fewer than shortstops. Center fielders were pretty close too, while catchers and left fielders had more stints. In regards to overall time on the DL, second basemen have everyone beat but catchers, but left fielders and center fielders are pretty close, and third basemen aren't exactly a world away. Verdict: For both seasons, second basemen were injured for a longer period than 1B, SS, 3B, and RF, are pretty close to CF and LF, and were injured less often than C.

Again, let's stress that this is a small sample. But given that this is currently all I have to work with, let's see if there's any big takeaways to pull from this.

-- It appears that the DL stint numbers for 2B, 3B, and SS are very similar over this period, but shortstops recover the fastest, then third basemen, then second basemen. This could be due to the nature of the injuries sustained, due to the increased athleticism of players in certain positions, we don't know.

-- It appears that playing left field and center field was quite a bit dicier than playing right field. Again, we're not quite sure why yet. We could hypothesize that maybe it is because center fielders have to do the most work, or because left field is often a dumping ground for some of the more older / injury-prone / clumsier players on the field. But we don't know.

-- Catcher appears to be the most hazardous position on the field, by the data. This passes all the smell tests, in my eyes. The squatting, the collisions, the foul tips, the wear ... it makes intuitive sense, for whatever that is worth.

-- Injury-wise, playing second base was comparable to playing left field or center field. Again, this doesn't necessarily mean that playing those positions definitely makes you more likely to get injured, but it doesn't seem to help.

So, I guess I'd like to leave things off with a more substantive conclusion, but I can't in this instance. Maybe second base really is more dangerous. Maybe it's not. Instead of wrapping this up in a neat little bow, let's think about how we can move this discussion forward. Here are a few talking points:

• Should we try to identify only defensive plays where a fielder is injured during the run of play? Are we best served in trying to figure out whether it is the act of fielding the position that causes injury, rather than the consistent playing of the position? I can see arguments both for and against.
• Should we count positions differently, perhaps rate a percentage for players who play multiple positions? Say, for example, Ben Zobrist spends 60% of the season playing second base, and 40% playing right field. Zobrist injures his oblique while swinging a bat in batting practice. Do we account for this as 60% of a second base injury and 40% of a right field injury? Do we count it as either? Both?
• What value can we ascribe to this information, if we do figure it out? Let's say our methodology is sound, and we determine that playing second base is, for example, 1.2x as likely to result in injury as playing shortstop. What would this data be worth? How could apply it?

Maybe if we can hammer out a decent methodology, and a decent idea of where we want to go, we can get somewhere with this. Injuries are one of the wild frontiers in baseball, and let's see if we can explore it a little.