In March of 2013, Coco Crisp spoke to Grantland about his base-stealing philosophy:

“A lot of guys like to see what kind of pitches a pitcher will throw, for hitting. Since I don’t care about that [laughs] … I just like to go up there, see the ball, hit the ball. I don’t like too much information for hitting. But for base stealing, I can never have too much information. So I like to go in there and see what their first movement is, check on their times [to home]. With base stealing, the biggest thing is knowing what time you can steal off of a pitcher, and being patient with that. I just try to collect as much information as possible so I have a higher-percentage chance to steal.”

Coco then went on to hit a career-high 22 home runs in 2013, but posted the lowest stolen bases number of his A’s career (21).  I don’t know why he ran so little in 2013; maybe his legs weren’t fully healthy all year, maybe age started to take its toll, or maybe he gave away too much strategy in the Grantland article.  But I have a sneaking suspicion that he was able to hit so many homers not by sheer luck or statistical anomaly, but by applying a basestealer’s mindset to home run hitting.  He’s not hitting homers, he’s stealing home runs.

Coco has been a remarkable base stealer over the course of his career, and his success rate has improved with age.  In his four seasons with the A’s, he has posted a 87% success rate (better than Carlos Beltran’s career rate).  Here’s another quote from the Grantland article:

“Q:  That whole thing about not making the first or last out of an inning at third base … will you never, or rarely, run if you’re on second with nobody out, or two outs, because of fear of getting caught?
A:  Well I’m always trying to make sure I don’t get caught. But even more so, I guess, in those situations. So I want to make sure that it’s 100 percent that I’m going to make it.”

For Coco, it’s not about attempting a steal every time he gets on base to rack up his total steals count.  Rather, it’s a case-by-case risk analysis.  If there are enough factors leaning in his favor (slow delivery, high leg kick, pitch type, game situation) he’ll go.  It seems to me he’s approaching home runs in the same way.  Here’s how he does it:

1.  Directional Hitting

Have you ever noticed that when reporters interview a guy after he hits a big home run, he often says something like, “I wasn’t trying to hit a home run, I was just trying to hit the ball hard”?  Very rarely do I ever see someone say, “Yeah, I was trying to dent the freaking moon.”  I kind of feel like this is one of those canned interview answers, like “both teams deserved to win” or “it’s not about me, it’s about the team” or “I only took steroids to recover from an injury”, but I also feel like this response has some truth to it.  Sometimes, a ballplayer really is just trying to hit a line drive, but accidentally hits a home run instead.

Coco Crisp is not one of these players, because he’s not strong enough.  (I’m speaking in relative terms.  Obviously Coco Crisp could beat the crap out of me; I’m just saying he doesn’t have the brute strength of a Miguel Cabrera, who probably often tries to park a double in the gap but instead hits a semi in the parking lot.)  Being the self-aware guy he is, he compensates for his lack of pure power by aiming down the lines.  This is where each of Coco’s homers landed in 2013:

You can see from the chart that Coco is an extreme pull hitter when it comes to home runs; if you chart his non-homers, he exhibits only slight pull tendencies and sprays the ball to all fields.  Out of his 22 homers, 21 were pulled strongly, many of them down the line, none were hit to center or the alleys, and one was hit to the opposite field.

What I see from this chart is intent.  When Coco decides to try for a home run “steal”, he maximizes his chances for success by taking advantage of (1) the shorter distance to the wall from the plate at the corners, and (2) the faster bat speed generated by pulling the ball.

2.  Situational Hitting

Using his pull-only approach, Coco will look for a particular pitch (seems to me like he guesses fastball or breaking ball about 50/50 on successful home runs) and if he gets that pitch on the inside half of the plate, he’ll let it fly.  But are there particular game situations where he is more likely to attempt a HR steal?

Let’s take a closer look at the differences in Coco’s approach when there are runners on base and when they are empty:

Bases empty:  8.1% BB, 13.2% K, 3.0% HR, 3.83 pitches/PA
Men on base:  9.9% BB, 8.8% K, 1.5% HR, 3.53 pitches/PA

These splits are over the last three seasons combined, but the differences in the ratios generally hold true for any of those three individual seasons.  In 2013, for example, he posted an ISO of .200 with the bases empty and .149 with men on.  16 of his 22 homers were solo homers (and 15 of the first 16).

Coco strikes out 50% more often and hits homers 100% more often with the bases empty.  This Jekyll-and-Hyde result appears to be due to a deliberate willingness to sacrifice contact for power with the bases empty.  With men on base, he appears to be much more of the self-proclaimed “see the ball, hit the ball” player.  I don’t think Coco is the only player to show a consistent situational split like this, but his does seem to be rather pronounced.

Why is this?  I’m not sure.  I think much of this has to do with the base-stealer’s mentality.  The element of surprise is a huge part of stealing bases, one that greatly increases the odds of success.  By that logic, it may seem to Coco that a game-changing home run is much more on a pitcher’s mind when there are runners on base than when the bases are empty.  Alternatively or in addition, it may be about risk for Coco.  While there is a greater reward for hitting a homer with men on base, there is also a greater penalty for striking out and failing to advance the runners or extend a rally.  I think that additional negative effect is hard to stomach for a base stealer that wants to be 100% sure he’s going to be successful on a stolen base attempt.

Of the players that hit 18 or more homers in 2013, Coco ranks dead last in average home run distance, and it’s not even close.   There was an article on FanGraphs last fall that questioned Coco’s ability to repeat his 2013 homer binge and suggested that “Crisp will have a difficult time replicating this season’s power numbers because in comparison to other home run hitters, he simply hasn’t hit his fly balls very far.”

I disagree, because Coco simply isn’t like other home run hitters.  He’s not a home run hitter at all, he’s a home run stealer.  I’d rather look at this the way we look at stolen bases: how many times did Coco attempt to hit a home run in 2013 and how many times was he successful?  Only Coco knows the answer to the first part of that question, but I think that Coco’s increase in homers in 2013 was driven more by an increase in home run attempts (i.e. more situations where Coco identified optimal go-for-it conditions) than an increase in success rate.  I would expect his rate of success would remain relatively constant or even rise as he becomes wiser with age (just as his SB success rate and BB/K rates have increased).  The real question then is how many times will Coco attempt to hit a homer in 2014?  If the answer to that unknowable question is something similar to 2013, I would give Coco a great shot to hit, nay, steal 20+ homers again in 2014.