January 20, 2006
The Case for Contact
"Lankford, who was recently benched then traded by the Cardinals, had an OPS of 841 for them this season and a career OPS of 848. While these numbers do not make him an All-Star, they certainly make him a serviceable major-league hitter. But Lankford was benched in favor of the light-hitting Placido Polanco. (Albert Pujols was moved to LF from 3B, taking Lankford's spot and making room for Polanco.) Polanco, who is having the best year of his career, has a 782 OPS this season and a 738 mark for his career. So what could possibly be the reason for such a move? Simple: Lankford strikes out too much.I went on to argue that when you add up all the situations where contact advances a runner, and subtract all the situations where contact results in a double play (a double play is to an advanced runner what a caught stealing is to a successful stolen base, i.e., the out is more than twice as bad as the advanced base is good), Lankford came out way ahead.
In essence, my conclusion was that benching a guy who strikes out too much (and Lankford struck out a ton) for a contact guy with questionable on-base skills and little power (Polanco has since improved a bit) was a mistake--at least on the offensive side. The team would net less runs as a result and presumably win less games.
But is that necessarily the case? Could it be that a contact guy has hidden advantages over the power hitter who strikes out and walks a lot?
Over the course of the season in terms of total runs created and OPS, the answer is no. Adam Dunn and a healthy Jim Thome will, of course, outproduce Ichiro and Juan Pierre by those metrics nearly every year. That they strike out frequently doesn't matter because they make far fewer outs per plate appearance, thanks to their many free passes. In other words, against the entire league, which includes fourth and fifth starters, middle relievers just called up from the minors and the entire Kansas City Royals' rotation, Dunn, Thome, Jason Giambi and their ilk are on-base and power machines. But is their approach to hitting--looking at a lot of pitches, getting ahead in the count and swinging from their heels--as successful against the top pitchers in the league, say those with a 2.5 to 1 K/BB ratio and a K/IP ratio of 7.5 or better?
The best pitchers in the league make hitters swing at more difficult pitches, both by getting ahead and locating well. The first hittable pitch you see might be the best one you're ever going to see in that at-bat. If you take it for a strike, you might not ever recover. But a contact hitter like a Juan Pierre or an Ichiro isn't as likely to get behind because they'll swing at the first good offering and often put it in play. Once the ball is in play, then the quality of the pitcher is largely removed from the equation. So while the top pitchers allow very low batting averages against--because of their many strikeouts--the contact guys aren't as subject to that because they don't miss much. And the extra out saved by the Dunns and Giambis by walking doesn't occur nearly as much against pitchers who rarely walk hitters.
Or at least that's my hypothesis. I haven't done a study, but I would love to see how Ichiro and Pierre, two extreme contact hitters, fare against the top 10-20 percent of pitchers (choose your favorite metric to say who those pitchers are) versus how the power-hitting/strikeout-walk guys fare. And if, as I hypothesize, the contact guys experience far less of a dropoff against the top competition, wouldn't that affect their value to a team? After all, a run against Pedro Martinez or Roy Oswalt has got to be worth more than a run against Aaron Sele or Ryan Franklin. If a pitcher only allows three runs per game, scoring a third of the expected output against him has got to be more important than scoring a sixth of the expected output against a lesser pitcher. If the contact guys don't fall off as much as the walk-strikeout-power guys as the competition gets tougher, then they're going to contribute a disproportionate amount of runs relative to their overall stats against the better pitchers.
Moreover, in the postseason, rotations shorten--you'll only see the fourth starter once a series at most, and scrubby middle relievers are only around for mop up duty in blowouts. No, you're seeing the big three, the primary setups and the stopper, most of whom presumably are better than the garden variety pitcher you see all year. And in the postseason, you're facing winning teams, so presumably their rotations and bullpens, all things being equal, are better than league average to begin with. Could this help explain why the A's of the aughts haven't won a playoff series?
Don't get me wrong--the traditional sabermetric approach favoring walks and strikeouts in equal proportion over a hit three times out of 10 is still correct--the issue is whether the K/BB ratio (as well as the home-run rate) can be maintained against the top pitchers. If it can't, (and surely it can't in most cases), then the question is whether the rate of drop-off for the big power/walking/low-contact guys is more severe than it is for the high-contact, low-power guys. If it is, and we posit that a run against a tougher pitcher is more valuable than a run against a weaker one, then we need to adjust our relative valuations of these different kinds of players, particularly in the postseason. Were we to do this, it would have major implications across the board from salary arbitration hearings to Hall of Fame voting. While there might not be proof of clutch hitting in baseball, a hitter whose skill set makes him disproportionately likely to come through against top competition might exist. Is this what old-school types like LaRussa instinctively sense when they worry about an otherwise productive hitter's strikeouts? Or is it, as I concluded five years ago, that they just find strikeouts embarrassing?