August 28, 2017

Flu-Like Symptoms

How Designated Hitters Lost Their Groove

by Rob Mains

Last week, I hosted a chat here. The first question, from John in San Francisco, was about Oakland DH/1B/3B Ryon Healy:

What do you do with a player like Ryon Healy if you're the A's? He showed real promise last season shooting through the minors and in a half season in the majors last year. This year, he continues to show power, but everything else—defense, on-base ability, baserunning—is lacking. With so many other DHs on the team or on the horizon, what do the A's do with a guy like Healy? He seemingly hasn't made adjustments now that pitchers have film of him. Is .250/.280/.480 a major leaguer?

I answered the question about the 2016 Full Vogelsong Player of the Year by pointing out that his .259 TAv at the time (it’s now .261) was slightly below average, but ranked him fourth among designated hitters (trailing only Nelson Cruz, Edwin Encarnacion, and Hanley Ramirez, assigning all of each player’s plate appearances to his time at DH).

This struck me as odd, so I decided to do some more digging about designated hitters. To aid my digging, I relied a lot on TAv. I’m guessing that if you came to this site and clicked on this article, you know what TAv is. If you do, skip to the next two paragraphs. But if you don’t, TAv is short for True Average. It’s our all-in measure of hitting skill. It’s adjusted for park and league, and scaled to .260, which means that every season, the league average TAv is .260. Think of TAv the way you’d think of batting average—.260 is average, .300 is good, .220 isn’t, that sort of thing.

I like TAv because it adjusts for the run environment and the ballpark in which players play, so we can compare, say, 1996 Dante Bichette (pre-humidor Coors Field, NL teams averaged 4.68 runs per game, .313/.359/.531 equates to a .261 TAv) with 1967 Jimmy Wynn (Astrodome, NL teams averaged 3.84 runs per game, .249/.331/.495 equates to a .332 TAv). You might be familiar with Baseball Reference’s OPS+, which is also park- and season-adjusted; it’s scaled to an average of 100. OPS+ is a useful measure, but I prefer TAv, because I feel it better weights the components of offensive performance.

As you probably know, scoring in baseball hit a nadir in 1968, when teams scored just 3.42 runs per game, the second-fewest in history, barely above the 3.38 average in 1908. In response, MLB lowered the pitching mound and shrunk the strike zone. Scoring rebounded to 4.07 runs per game in 1969 and 4.34 in 1970. But then it began marching downward again, especially in the American League. Here’s a very small chart:

By 1972, American League teams were scoring 3.47 runs per game, barely above 1968’s 3.41. So the following winter, American League owners voted to implement the designated hitter rule. American League scoring has never dipped below 4.00 runs per game since the DH began.

In its first years, the DH was dominated by old guys, mostly with bad wheels. In 1973, of the eight DHs with the most plate appearances, seven were 34 or older. As the position evolved, DHs trended younger and more athletic, as great hitters who just weren’t good in the field took over: Hal McRae, Paul Molitor, Edgar Martinez, Frank Thomas, Travis Hafner, David Ortiz. And as good—not just old—sluggers gravitated toward the position, DHs became an offensive force:

I added that solid black line at .260 because it represents the league average. As you can see, DHs have been above-average hitters, every single season. That shouldn’t be a surprise, of course. Your average DH had better out-hit your average catcher or middle infielder. One of the tenets of Bill James’ defensive spectrum is that there are a limited number of players who can play the challenging defensive positions at the right end of the spectrum (left to right: 2B, SS, C) while a lot can play at the left end (DH, 1B, LF). So it should be easy to find good hitters who can be DHs.

But I cut that chart off at 1998 for a reason. The 1998 season, it turns out, represented the high-water mark for DH TAv. That year, DHs had a .283 TAv. Martinez hit .320/.426/.561 to lead the way, and in the American League, only first basemen, at .289 (Carlos Delgado, Mo Vaughn, Jim Thome, and Rafael Palmiero all had TAvs above .315), out-hit DHs.

Since 1998, DH production has faded:

Look at that last point on the graph: Yeah. So far this year, designated hitters are well below-average hitters. That’s crazy, right?

Let me show you a few more graphs to explain how we got to this point. Here’s one that shows TAv by position in 1973, the first year of the DH. I’m going to include only American League teams, given that it’s the league that has the DH, and thus has roster decisions and options different from those in the Senior Circuit.

If nothing else, that’s pretty colorful, right?

The solid black line at .260, once again, represents the league average. As you can see, in the DH’s inaugural season, DHs were only slightly above average.

Fast forward 11 years, to 1984. Now the DH is better established, and teams have moved away from recruiting AARP members as their DH. Here’s the breakdown by position, using the same colors as before:

(Technical note: I know that I’m playing around with the scale here, and I’m not supposed to be truncating the scale for bar graphs in any case. Guilty as charged. The key here, I think, isn’t the absolute number, or how it’s scaled relative to the other graphs, but rather the rank and the difference from the solid black line.)

American League right fielders—Dwight Evans, Dave Winfield, Harold Baines—had a big year in 1984. But DHs were good, too; their .275 TAv compares to .263 in 1973.

Moving forward another 11 years, to 1995, a (slightly) strike-shortened season:

American League first basemen had some crazy years in 1995. Mark McGwire was still with Oakland and Frank Thomas played mostly in the field, and they had OPS totals above 1.000. Palmiero, Vaughn, and Tino Martinez had big seasons too. But DHs were second in production, with a .280 TAv.

Eleven years later, 2006, the shape of the graph changed:

DHs still had the second-highest TAv among American League positions. But there was a significant leveling that occurred between 1995 and 2006. The high end got lower and the low end got higher. Maybe the shorter 1995 season created more outliers. Maybe it was the introduction of the Joint Drug Agreement in 2005. Maybe it was noise. But DH TAv fell from .280 in 1995 to .269 in 2006.

Perhaps that’s what’s driving the decline in DH TAv. Maybe it’s not that DHs are getting worse, it’s just that everything’s getting more even. The offensive production from the left end of the defensive spectrum doesn’t exceed the right end as much as it used to. The top 30 players so far this year in TAv (minimum 350 plate appearances) include right-side-of-the-spectrum Anthony Rendon, Jose Altuve, Carlos Correa, Zack Cozart, Corey Seager, Buster Posey, Daniel Murphy, and Eugenio Suarez. Might it just be a big leveling?

In a word, no.

Yes, the ranges have flattened. The difference between the best- and worst-hitting positions isn’t as large as it once was. But designated hitters are qualitatively worse. American League first basemen are hitting better than DHs, which isn’t remarkable. But so is every single outfield position. And third basemen. American League second basemen are generating more offense than DHs, for crying out loud. DHs have a collective TAv that’s only two points higher than shortstops. DHs just aren’t providing the offense they used to.

Having illustrated that DHs are contributing less on offense, I’ll explain why I think it’s happening in my next article. See you Thursday.

Rob Mains is an author of Baseball Prospectus. Follow @Cran_Boy
Click here to see Rob's other articles. You can contact Rob by clicking here