World Series time! Enjoy Premium-level access to most features through the end of the Series!
July 30, 2003
Lies, Damned Lies
One of the perks of traveling for work--I've been doing a lot of that lately--is the USA Today planted in front of your hotel room door. Sure, for the most part, McPaper's articles are about as substantive as the "continental breakfast" you're likely to eat while reading it--but now and then, in its own glossy, Technicolor way, USA Today stumbles across something significant.
Last Wednesday's sports page featured a headline on leadoff hitters--it seems that there aren't very many good ones these days. As the article pointed out, none of the league's leadoff hitters are among the top 30 players in OBP. Among qualified players, the highest-ranking leadoff hitter is Ichiro Suzuki, 39th as of this writing (Jason Kendall, who has occupied the leadoff spot in Pittsburgh since the departure of Kenny Lofton, ranks 31st). And it's not as if Suzuki or Kendall are walking machines in the mold of Rickey Henderson--Ichiro is a fine player who can hit .340 consistently, but his walk rate is well below league average, while Kendall's OBP is boosted in part by his fearless desire to lean into pitches.
Then again, players of the Rickey/Tim Raines profile have never been terribly common. It also doesn't help when teams insist on placing mediocrities like Eric Young or Endy Chavez in the one-hole. Is anything going on here, apart from a one-year fluke?
The problem, in fact, is widespread, and it hasn't been limited to this season. In the table and graphic below, I've presented OBP by batting order position for each of three periods: (i) 1982-1989, sort of the Henderson/Raines prime years; (ii) 1999-2000, the last days of irrationally exuberant offense, and (iii) 2001-2002.
OBP by Batting Order Position
Order # 1982-1989 1999-2000 2001-2002 1 .336 .349 .332 2 .333 .346 .331 3 .349 .384 .379 4 .345 .375 .368 5 .329 .356 .338 6 .322 .345 .327 7 .315 .326 .318 8 .308 .329 .312You'll notice how much the No. 3 and No. 4 slots have come to monopolize OBP. Although those lineup positions also accumulated higher OBPs than the leadoff slot during the 1980s, the gap is much wider now, on the order of 40 or 50 points instead of 10 or 15.
Moreover, the gap is growing wider. While the new strike zone introduced in 2001 decreased OBPs across the board, the power hitters that occupy the No. 3 and No. 4 spots were relatively insulated from the trend, declining by five points and seven points, respectively. By contrast, leadoff hitters lost 18 points off their collective OBP (the second, fifth, sixth and eighth slots were similarly affected).
"The strike zone is the very heart of a baseball game. An inch in the strike zone means far more than ten yards in the outfield." --Bill James, 1988 Baseball Abstract (p. 23)
In the 1988 Baseball Abstract, Bill James ran a study of the 1963 strike zone expansion, which also reduced offense significantly. James sought to identify how different sorts of hitters were impacted by the new strike zone. Although we won't go into as much nuance as James did, let's look at the impact of the 2001 strike zone expansion, grouping players by one particularly important characteristic: their isolated power (ISO). In the table below, I've sorted all batting title qualifiers in 2000 into rough quartiles based on their ISO in 2000, and compared their average strikeout and walk rates before and after the introduction of the new strike zone.
ISO BB Rate K Rate Quartile Range n 2000 2001 Delta 2000 2001 Delta ---------------------------------------------------------------------- 1st .000-.140 36 8.6% 6.6% -24.0% 12.8% 14.4% +12.0% 2nd .140-.180 46 8.6% 8.0% -7.5% 14.2% 15.4% +8.4% 3rd .180-.230 38 8.9% 7.7% -13.6% 16.5% 17.3% +5.0% 4th .230+ 43 11.4% 10.5% -7.6% 16.1% 17.0% +5.8%Players with poor isolated power suffered the most from the expanded zone. Their walk rates declined by 24 percent, compared to a composite nine percent among the other groups. Their strikeout rates also increased more substantially than other types of hitters. Since these hitters almost by definition rely on OBP to have any value, this is a potentially devastating trend for them.
I checked to see whether the same phenomenon occurred in seasons in which the strike zone did not change--could it be that slap hitters are always vulnerable to such a decline in their in walk rate? Quick-and-dirty answer: they aren't. Walk rates are one of the more stable statistics in the game, and the decline these hitters experienced in 2001 seems to be directly related to the redefined zone.
These trends are interesting in light of James' findings for a couple reasons. First, James discovered that much of the change in offense that occurred between 1962 and 1963 was attributable not to a change in walk rates, but to a decline in isolated power. He suggested that this was a result of compensation on the part of the league's pitching staffs; under the new conditions, power pitchers (high-K, high-BB) were able to prosper, and they occupied a higher percentage of the league's innings pitched.
This wasn't the case in 2001. Isolated power hardly declined at all, and most of the offensive downturn was realized on the BA and OBP side of the ledger. As Joe Sheehan has speculated, modern pitchers are already exploiting their ability to generate strikeouts to the maximum possible extent. A bigger strike zone doesn't induce any sort of "compensation"--it just permits even more strikeouts, without a corresponding increase in walks.
In a related item, James also discovered that the 1963 change had a relatively greater effect on taller players--behemoths like Frank Howard (6'7") suffered, while shrimps like Albie Pearson (5'5") were largely unaffected. Because ISO is correlated with height, this conclusion might be relevant in light of our earlier findings. No need to be lazy here; let's run the same analysis of changes in strikeout and walk rates, this time based on a player's height.
BB Rate K Rate Height n 2000 2001 Delta 2000 2001 Delta ------------------------------------------------------------------- 5'11" or shorter 42 9.5% 7.8% -18.2% 13.1% 14.2% +8.5% 6'0", 6'1" 60 9.2% 8.1% -11.2% 15.2% 16.3% +7.7% 6'2", 6'3" 43 9.3% 8.5% -8.9% 15.2% 16.6% +9.2% 6'4" or taller 15 10.5% 9.7% -8.2% 18.1% 18.2% +0.7%Although the strikeout rates don't exhibit a clear pattern, there's a pretty strong relationship between walk rates and height--the new strike zone has been especially harmful to shorter players. And that makes sense. The 1963 strike zone expansion consisted of adding strikes in the lower part of the hitting zone, and taller hitters suffered. The 2001 strike zone expansion was the opposite case--it consisted of reinstating the high strike. Players who are best equipped to handle that pitch--taller players, and power hitters with good bat speed and uppercut swings--have adapted best to the change.
We've strayed pretty far from the original question--whatever happened to leadoff hitters?--but I trust you'll see where I'm going with this. At least in the traditional reckoning, the ideal leadoff hitter is a guy who does an exceptional job of getting on base, but doesn't have a lot of "wasted" power that could be put to better use later in the order. (We'll take a mulligan on the matter of baserunning ability for the time being). You're looking, in other words, for a high-OBP, low-ISO player. Those guys are increasingly difficult to find in today's game.
One more little study: let's look at the correlation between isolated power and unintentional walk rate for each year since 1955 (the first year that IBBs were recorded separately). I've weighted the correlation by plate appearances, so as not to distort the results with small sample sizes.
Walks and power have always been related skills. While the usual stathead take is that a good batting eye facilitates power numbers by allowing a hitter to wait for a pitch he can drive, the reverse is also true: power facilitates walks. Unless you're talking about Bobby Jenks or someone, pitchers are rational beings, and the incentive to pitch a guy carefully comes when he's capable of getting an extra-base hit--you'll trade-off an increase in the chance that a guy reaches first for a decrease in the chance that he'll hit a long one. (That's especially true when there are runners on base, which is why leadoff hitters actually face more of an uphill battle in drawing walks than anyone else. If you're looking for another testament to Rickey Henderson's greatness, this is it).
As a result of the new strike zone, the correlation between walks and power took a huge jump upward in 2001, one that shows no signs of abating. Increasingly, hitters aren't differentiated by having particular profiles of skills--Tony Gwynn on the one hand, Rob Deer on the other--but simply by being good or bad. Specialization is passť: the good hitters tend to do everything well, while the lesser hitters do none of it.
Arguably, a trend like that ought to produce a bit of a reprioritization in terms of player development. We're in the midst of a running debate within the BP authors group about high-OBP, low-SLG minor leaguers--think someone like Esteban German. As much as we believe that drawing walks is an essential skill, there's some question about how well walk rates will hold up in the major leagues when a guy has so little power that pitchers have no disincentive to throw him anything but strikes. A hitter like Kevin Youkilis, who can loft the ball and has demonstrated doubles power, is probably OK, but many hitters in the German mold struggle upon reaching the big leagues (take a look at his PECOTA comparables). Given the new strike zone, the odds of someone like German making a successful go of it in the majors are slimmer still. A player with plate discipline that accompanies other skills is an elite prospect, but "empty" walks may be something to be wary of, just like empty batting averages. Isolated power may be to hitting prospects what strikeout rate is to pitching prospects.
The matter of what to do about leadoff hitters is simpler. If you don't have someone like Ichiro or Kendall or Luis Castillo--and most teams don't--it isn't worth faking it. Giving additional plate appearances to a .280/.340/.370 player, simply to avoid "wasting" power, isn't terribly beneficial. While the vintage model of leadoff hitter might have gone the way of the Pinto, power hitters are as common as SUVs, and it's time to start revving up the offense early and often, gas mileage be damned.