keyboard_arrow_uptop

About a week ago, I contributed a portion of an article published on this site that was centered around a simple conceit: Sam Miller and I would each design and manage an All-Star roster optimized to win a single game, and a single game only. In the introduction to my section of the piece, I noted that I found the whole exercise a little foolish:

To design a system that simultaneously maximizes the odds of winning one—and only one—baseball game and comports with some recognizable form of baseball reality requires selecting from a pool of players (in this case, big-leaguers) whose talents have been heretofore understood and described in terms of their ability to contribute to wins over the rather large (but still impossibly small) sample of 162 games.

A week later, I no longer think the exercise was particularly foolish. “Curious,” is perhaps a better word, and “quixotic” is perhaps an even better one. I am convinced, however, that my participation in the proceedings was foolish in the extreme, and for a very simple reason: I got smoked.

I’m not here to argue that I deserved to win. Meg Rowley is as infallible as she is enamored of Kyle Seager (extremely, in both cases) and I acknowledge that I deserved to lose the competition, and to feel bad about it besides. But I do want to take the opportunity to defend and substantiate one particular portion of my strategy: hitter selection. Here’s what I said:

I’ve given it some thought, and I’ve decided that I want—all other things being equal—hitters who see a lot of pitches, hitters who make a lot of contact when they’ve had enough of seeing pitches, hitters who hit a lot of line drives when they make contact, and hitters who play good defense when they’re not hitting.

My thinking (shamefully unsubstantiated, in the piece) was that Sam’s pitchers—in general, a parade of high-strikeout, low-home-run-allowing relievers—would be best countered by hitters fitting the profile described above, and not necessarily always by the best hitters in the game, if those hitters tended to rely for their offensive production on both (a) prodigious power accompanied by (b) a lot of strikeouts—which two features are often found together.

Some, in the wake of my piece, suggested that it was not, and that (in particular) high-strikeout hitters might not be at any particular disadvantage against high-strikeout pitchers, and high-contact (etc.) hitters not at any particular advantage. So—is that true?

Here’s one very simple way of getting towards an answer. There’s probably more complicated ones out there, but I don’t know the math to do them well. So here’s what I did: I asked BP’s Stats Team to identify, for each of the years 2006-2016 (thus covering portions of 11 seasons) the pool of relievers who ranked in the top 20 percent of the league in both strikeout rate and home run rate allowed. These relievers, I supposed, were of the elite type—high-K, low-dinger—who might face my hitters in the fictional All-Star game. They were, therefore, the pitchers against whom I wanted to test my assumptions.

Here, for example, is a portion of a table:

Pitch Count effects, 2006-2016, vs. t20 percent relievers

High-PC

.188

.296

.274

REDACTED FOR THE SAKE OF SUSPENSE

The row of the table displayed above records the mean triple-slash line—again, against only the pool of elite relievers—of hitters who ranked in the top 10 percent of the league in pitches seen per plate appearance. That line, you will note, was .188/.296/.274. It’s pretty bad, which is to be expected: the relievers in our sample are very good. The real question is, then: How did the other guys do?

Others

.205

.283

.280

Delta

-.017

.014

-.006

% Delta

-8%

5%

-2%

The top line of this table records the triple-slash line of the other guys—the bottom 90 percent of the league (in pitches seen per plate appearance), against the same pool of relievers. Their line is .205/.283/.280, and that—this is the important part—is better than what the patient guys did against the same guys. The trend isn’t totally consistent—patient hitters had an OBP 5 percent higher than the bottom 90 percent, for example—but in general the data suggest that there’s no particularly compelling reason to select for patience when going up against elite relievers.

So, in at least this respect, I was wrong, and the haters and losers were right. Sad! But patience wasn’t the only thing I was looking for. I also looked for contact rate:

Contact Rate effects, 2006-2016, vs. t20 percent relievers

High-Contact

.228

.307

.294

Others

.196

.280

.275

Delta

.032

.027

.018

% Delta

16%

10%

7%

Here, I think, my case was a little stronger. Once again, nobody did particularly well against elite relievers, but hitters with high contact rates (top row) had a batting average 16 percent higher, an on-base percentage 10 percent higher, and a slugging percentage 7 percent higher than the rest of the league. This suggests that there is, in short, reason to select—ceteris paribus—for hitters with strong contact ability against elite relievers.

How about line drive rate?

Line Drive Rate effects, 2006-2016, vs. t20 percent relievers

High-LD

.213

.303

.288

Others

.199

.281

.277

Delta

.014

.022

.011

% Delta

7%

8%

4%

The effect is less pronounced, but it’s still there: the hitters who were best at producing line drives saw gains across the board versus their normally-abled brethren. I’m thus, once again, confident that it’s useful, ceteris paribus, to look for hitters who’re good at hitting the ball on a line. They do perform better against high-strikeout pitchers than their peers.

I’m equally confident, more to the point, that high-strikeout hitters need to hit for a lot of power to be worth selecting over hitters with less power, but more of the characteristics I discussed above. Here are the charts for hitters with an especially high strikeout rate:

High-K effects, 2006-2016, vs. t20 percent relievers

High-K

.167

.259

.244

Others

.212

.293

.288

Delta

-.045

-.034

-.045

% Delta

-21%

-12%

-16%

Woof: 16 percent is a big dropoff in slugging, and 21 percent is a bigger one in batting average (though I care about that less).

Note that the correct conclusion to draw from this is not that it’s not ever worth picking up a hitter with a high strikeout rate for an exercise that Sam and I went through. It’s just that, if they’re someone like Adam Dunn (who was definitely in the top 10 percent in strikeout rate, for most of those years) they have to have a slugging percentage that is at least (approximately) 16 percent higher than the lower-strikeout rate guy just to break even, in terms of their value to your team. Adam Dunn meets that criteria. Not many others would. And therein lies the rub.

The exercise Sam and I engaged in was silly. It was fun. And I lost it, fair and square. My abominable pitcher usage strategy probably had a lot to do with that, but on the basis of these numbers, I’m not so sure my hitter selection strategy did: high-contact, high-line drive hitters are useful, at a premium of between 4 and 7 percent of slugging, and high-strikeout hitters are harmful, at a discount of about 16 percent of slugging, when facing elite pitching. Is that the difference between Kris Bryant (who I left off the team) and Christian Yelich (who I kept on)? Maybe. Maybe not. But I think the tradeoff I considered was the right one to use, and so I’ll take the loss standing up.

Addendum: In response to a question by Ben Lindbergh, in the comments below, I've run the same tables as you see above, but against all pitchers rather than just elite pitchers. I then compared the percent-deltas, to see if the effect was exaggerated against elite pitchers. You can see the results here, but the short version is that yes, they're very exaggerated. This, happily, lends even more credence to my strategy.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
bornyank1
7/08
Do you have the average slash lines of the high/low-contact and high/low-LD groups against the league as a whole? Wondering if one or the other tends to be better against everyone, not just against elite relievers.

(I don't doubt that there is some real effect here--I once tried to detect one myself--but I can't tell whether these tables settle it one way or another.)
rianwatt
7/08
That's a good point, Ben. I'll see if Rob might be willing to run those.
rianwatt
7/08
New data here:

https://docs.google.com/spreadsheets/d/13z4T1f5vR_AzV-QlB-svQG9AsaXDw-B8Sz4VyqiP_Uo/edit?usp=sharing
lichtman
7/09
Not getting this at all. According to your data:

Your top 10% in line drive rate lose 42 points in OBP versus "elite relievers" (as opposed to all pitchers).

All other hitters lose 43 points. Virtually identical results.

Your top 10% in line drive rate lose 143 points in SLG
versus "elite relievers" (as opposed to all pitchers).

All other hitters lose 131 points. Virtually identical results although yours do a little worse.

Your top 10% PC guys lose 52 points in OBP versus elite relievers and "the rest" only lose 41 points.

Your top 10% in PC lose 148 points in SLG and "the rest" lose 130.

Doesn't look like your theory holds any water to me whatsoever.