Betts, Carpenter, and the Evolution of the Leadoff Hitter

Historically, leadoff hitters don’t possess much power. Historically, they have served as table-setters, players who get on base so that more powerful hitters down the lineup can drive them in. Historically, all that’s true.

A look at this year’s home-run leaderboards, however, reveals Mookie Betts, Matt Carpenter, and Francisco Lindor all among the top 15. Betts is having a year that rivals Rickey Henderson’s 1990 campaign as the greatest leadoff season of all time, and Carpenter leads the National League in homers and WAR. Lindor, meanwhile, is just a lone dinger away from his second straight 30-homer season. That all bat leadoff for their respective clubs.

Because it receives the most plate appearances, the leadoff spot is, by definition, relevant to a club’s run-scoring efforts. Despite its importance, teams have generally failed to place one of their best hitters in that position. In 2002, leadoff hitters put up a 93 wRC+ overall, behind the marks posted by Nos. 3 through 6 in the lineup and virtually even with second and seventh. That’s just one year, but it’s representative of teams’ reluctance to place their best, or even second-best, hitters in the leadoff spot. The graph below shows a five-year rolling OPS+ for the leadoff spot, with data from Baseball-Reference.

For the most part, leadoff hitters have been roughly league-average hitters. They were a bit better than that in the late 1960s, when pitchers dominated everyone, and they had a great run in the late 80s and early 90s, too, when Hall of Famers Rickey Henderson and Paul Molitor were putting up great seasons at the top of the lineup. It’s possible teams spent the rest of 90s and early 2000s looking for Rickeys and Raineses and that, when they couldn’t find speedsters who got on base a ton and hit for extra bases, they merely settled on players who possessed the first of those traits. The result was suboptimal lineups that left runs on the table by giving too many plate appearances to players who weren’t among the best hitters on the team.

There’s been a noticeable change over the last decade, though, leadoff hitters have been above-average hitters for the most part. Below we see all players with at least 250 leadoff plate appearances this season.

Leadoff Hitters in 2018
Name Team PA OBP SLG wRC+
Mookie Betts BOS 460 .441 .673 195
Matt Carpenter STL 348 .405 .644 175
Francisco Lindor CLE 530 .368 .556 147
Shin-Soo Choo TEX 360 .393 .507 142
Brandon Nimmo NYM 275 .371 .478 135
Whit Merrifield KCR 310 .369 .429 118
Lorenzo Cain MIL 325 .366 .443 118
George Springer HOU 483 .337 .438 116
Curtis Granderson TOR 281 .343 .440 114
Yoan Moncada CHW 379 .325 .446 112
Derek Dietrich MIA 253 .356 .406 112
David Peralta ARI 266 .337 .440 109
Trea Turner WSN 271 .343 .431 109
Chris Taylor LAD 311 .328 .427 108
Cesar Hernandez PHI 513 .374 .371 107
Adam Eaton WSN 212 .367 .364 104
Charlie Blackmon COL 250 .344 .464 103
Marcus Semien OAK 281 .317 .418 102
Jose Peraza CIN 215 .335 .414 100
Jon Jay 2 Tms 426 .348 .369 99
Brett Gardner NYY 428 .339 .377 99
Trey Mancini BAL 232 .328 .392 98
Travis Jankowski SDP 263 .344 .335 93
Leonys Martin DET 314 .315 .389 91
Brian Dozier 2 Tms 259 .317 .370 89
Joe Mauer MIN 252 .317 .349 80
DJ LeMahieu COL 285 .310 .418 79
Dee Gordon SEA 450 .302 .346 79
Ian Kinsler LAA 244 .262 .371 74
Ender Inciarte ATL 254 .292 .266 56

There are still some disappointments down towards the bottom of that table, but nearly two-thirds of baseball teams are employing at least average hitters at leadoff this season. The Cubs, Giants, and Pirates, Rays haven’t landed on a leadoff hitter this season, though the Cubs’ experiment with Anthony Rizzo says they are willing to try one of their best hitters at the spot. The same is true for Brian Dozier in Minnesota even if the results weren’t there. That there are four hitters with a wRC+ of at least 140 is pretty remarkable. The table below contains all seasons since 2002 in which a hitter recorded at least 500 leadoff plate appearances and a batting line at least 40% better than average.

Best Leadoff Hitters Since 2002
Name Season PA AVG OBP SLG wRAA wRC+
Mookie Betts 2018 460 .352 .441 .673 54.7 195
Matt Carpenter 2018 348 .295 .405 .644 32.8 175
Mike Trout 2012 639 .326 .399 .564 48.2 167
Shin-Soo Choo 2013 669 .294 .432 .481 47.0 158
Hanley Ramirez 2007 526 .345 .405 .596 42.0 157
Jacoby Ellsbury 2011 689 .327 .381 .556 47.7 152
Hanley Ramirez 2008 622 .304 .400 .559 41.8 149
Matt Carpenter 2013 632 .323 .398 .483 34.8 148
Francisco Lindor 2018 530 .290 .368 .556 32.3 147
Matt Carpenter 2016 511 .276 .386 .527 28.3 143
Shin-Soo Choo 2018 360 .275 .393 .507 20.8 142
Jose Reyes 2011 584 .336 .383 .493 27.8 142
George Springer 2017 625 .284 .369 .525 30.3 141
Alfonso Soriano 2006 610 .294 .368 .588 33.1 141
Charlie Blackmon 2017 716 .329 .397 .602 56.1 141

In the 16 seasons prior to this one, only 11 players met the criteria above. With the current season approaching its final month, there are still four hitters who meet those criteria for 2018. The top two marks on the list have been produced by players this season: Betts and Carpenter are in a position to surpass Mike Trout at something, which is always a big accomplishment. Betts is actually in the middle of potentially the greatest leadoff season of all time. A few years ago, I wrote about Matt Carpenter’s hot start and discussed the top leadoff seasons ever. Only eight times has a leadoff hitter amassed a qualifying number of plate appearances at that position and had a wRC+ at least 50% better than league average. Rickey Henderson authored three such seasons seasons. Here are the rest with their full-season totals — i.e. with PA from other spots in the lineup.

Best Leadoff Seasons Ever
Player Year wRC+ BsR OFF
Mookie Betts 2018 192 5.5 56.6
Rickey Henderson 1990 190 8.8 69.4
Mike Trout 2012 167 14.1 64.2
Matt Carpenter 2018 159 -0.2 35.8
Rickey Henderson 1985 159 11.7 58.3
Pete Rose 1969 154 -2.0 42.7
Rickey Henderson 1993 151 7.5 46
Don Buford 1971 151 0.6 31.6
Jacoby Ellsbury 2011 150 5.1 47.5
Shin-Soo Choo 2013 150 2.6 42.8

The projections have Betts producing a 180 wRC+ and about 67 offensive runs this year, which would both fall just short of Rickey Henderson, whose runs total likely sells him short as BsR did not contain runs on double plays or any other runs on the bases aside from steals. Carpenter is currently projected to end up with a 154 wRC+, even with Pete Rose’s 1969 season.

As for the evolution of the leadoff hitter, it has received considerably less attention than that of the No. 2 hitter. There’s good reason for that — namely, that No. 2 hitters have gotten much better over the last decade. The graph below shows the splits from every lineup position by year since the 2002 season.

A decade ago, the No. 2 spot in the lineup produced runs at a rate close to leadoff and the sixth spot in the lineup, a little bit ahead of the seven hole. As a greater understanding has emerged regarding the potential runs gained by optimizing the lineup — particularly from The Book — the No. 2 spot in the lineup has seen steady growth to the point where it is nearly even with the cleanup spot. The leadoff spot has seen growth by distancing itself from seventh, bypassing sixth, and coming even with fifth, but it really should be higher than that. The value of a home run is slightly reduced at the leadoff spot, but that doesn’t justify putting a hitter not in a team’s top three at that spot just because they hit homers.

The gains from the second spot in the lineup seem to have come at the expense of the fourth spot in the lineup. While it might garner admiration in sabermetric circles — including from myself — to bat a better hitter second, if that hitter is simply moving from the fourth spot, there isn’t a great advantage gained. The authors of The Book found, regarding lineup construction, that it is to bat the best hitters first, second, and fourth, while switching the second and fourth hitters makes little difference.

Switching the numbers for the leadoff and third hitter above might only amount to one run over the course of the season, but given that the gains in the No. 2 spot have come at the expense of the four hole with the third spot remaining on top, it’s fair to argue whether the discourse regarding lineup optimization has made much actual impact.

I should note that the extent of the impact it could make is probably limited. The question of lineup construction is mostly moot, so long as better hitters are jumbled near the top of the lineup. Nevertheless, it’s interesting to examine the achievements of this year’s leadoff hitters in context. The offenses of the Cardinals, Indians, and Red Sox have certainly benefited. It’s a great year for leadoff hitters, and more teams should consider starting their lineup with a bang.

Craig Edwards can be found on twitter @craigjedwards.

Newest Most Voted
Inline Feedbacks
View all comments
5 years ago

It’s interesting, because by The Book (IIRC, it’s been a while since I read it) you’re supposed to bat your best hit/power combo 2nd, and that’s clearly Betts. That would probably put Benintendi in the leadoff spot, and he would be a very good leadoff hitter (.380 OBP, well above average baserunning) Betts’ OBP is so much higher he’s the best option at whatever spot you deem most important.

I don’t think The Book assumes that you have a guy with a .317 ISO and a .440 OBP, because how many of those do we have? Betts, Trout, and Ramirez are three guys who just provide an embarrassment of riches.

5 years ago
Reply to  sadtrombone

One thing Cora has been talking about a lot is the psychological value of putting your best hitter in the very first at bat of the game. Given that the sabermetric question is effectively a toss-up (Betts’ ISO loses some value, but his OBP and BsR gain quite a bit, so… draw?), I tend to give the benefit of the doubt to whatever “soft factor” edge they feel they can get by putting Betts in the 1 hole.

Also, batting leadoff really does seem to agree with him. I haven’t done a rigorous study of it, but honestly he seems to have produced the best numbers (indicating a higher degree of comfort) in the leadoff spot.

5 years ago
Reply to  kbn

I think you have a good point. Because no one has defined a metric for synergy, doesn’t mean it doesn’t exist.

5 years ago

I agree with kbn as well. Cora said in the spring when he declared Mookie the leadoff hitter that he wanted to set the tone, the synergy of which you speak, which is exactly what he has done.

5 years ago
Reply to  sadtrombone

I agree that batting Benintendi leadoff and Betts second makes the most sense in theory, but considering how the Red Sox are performing as a whole this season, there’s no point at all in even pondering a switch (not saying that you’re suggesting that; I agree with you!) The comfort of Mookie and Benintendi matter more during such a successful season than *maybe* gaining some small benefit in production by switching them. If Mookie likes leadoff, he’ll bat there for as long as he pleases, Cora must be thinking, and he’s right!

5 years ago
Reply to  sadtrombone

Whatever you lose in “wasting” his power leading off you probably gain by however many extra at bats he gets throughout the season by hitting 1st. Not many, probably, but maybe 5-10 over the course of a season? Now, of course, for him to gain an extra at bat means that Betts is making the last out of the game, but still.

5 years ago
Reply to  dl80

The question becomes do those extra at bats with nobody on make up for the fact the #2 is likely to come to the plate more often in bigger situations? I’d rather have a few less AB in general, but the AB Betts has come in higher leverage or bigger spots in the game.

For instance, with RISP Betts has the 9th most PA on the Red Sox. JD Martinez (142), Benitendi (134) lead the club, and even guys like JBJ (107), Nunez (105) and Brock (90) come to the plate in bigger spots than Betts (88). If you switch to looking at high leverage, Betts has the 6th most PA on the Sox.

Meanwhile Betts leads the club with the most PA with nobody on.

I guess I’d much rather have Betts hitting in tougher situations, and with more runners on base where he can make a bigger impact. So I’d definitely bat Betts second if it was up to me.

5 years ago
Reply to  Twitchy

This is where I think I would lean too. My understanding is that if you have a guy who is the best at everything (or close to it) he should be batting 2nd for the reasons you describe. That said, he’s so, so good at getting on base that it’s tempting to bat him leadoff anyway. Having a guy like Betts/Trout/Ramirez means you’re so awesome anyway it’s hard to make a wrong decision.

5 years ago
Reply to  sadtrombone

Yeah I mean it’s hard to call it a bad decision leading off. I just think that shifting 15-20 at bats to someone else (who is still a great hitter in his own right), while giving your best bat more opportunities to drive in runs makes you slightly better.

5 years ago
Reply to  Twitchy

One of the reasons that Benintendi and JD have more at bats with RISP is that Mookie is batting ahead of them. Put a lesser hitter there, and those numbers would decrease dramatically.

5 years ago
Reply to  Gregg

If that lesser hitter was Eduardo Nunez, sure. If it were Benintendi not so much

5 years ago
Reply to  Gregg

Almost everybody on the team has more PA with RISP than Betts. It’s not like JBJ is coming up very often with Betts at the plate, yet he has more opportunities with RISP.

Yes, Betts OBP helps the guys behind him, but it’s also a situational thing. Batting second would allow him to still get a significant # of at bats, while giving him more at bats in critical spots.

5 years ago
Reply to  Gregg

I agree with this. There are so many factors that come into play when you switch hitters around in order to create opportunity. I think the best direction is for Mookie to lead off; get your best player to the plate as often as you can.

5 years ago
Reply to  sadtrombone

Very approximately, every spot in the lineup is worth 18 PAs over the course of the season (if all spots are equally likely to get the last out, 162/9 = 18). Give those 18 PAs to Betts, and he’s likely to get on base 8 times and hit one HR. With those same 18, Benintendi gets on base 7 times and hits 0.5 homers. If you’re always getting solo HRs, that’s about 1.5 runs in value that Betts gains over Benintendi over the season.

Of course, extrapolate those HR rates to the full season, and you see that Betts is going to have about 9 first-inning HRs compared to Benintendi’s 4.5. If Benintendi is on base for over a third of those “extra” HRs that Betts would hit in the #2 spot, you make up the 1.5 runs, but barely. This is all very rough math, but I’m pretty confident in saying that the difference between Betts-Benintendi and Benintendi-Betts is less worth at most a couple of runs over the course of the season.