Browns’ Shaun Rogers is a Life-Saver
July 22, 2010While We’re Waiting… Paying College Athletes, Bentley Sues Browns, and Rogers 911 Audio
July 23, 2010Before the season started, we looked in depth at the logic of a well-constructed batting order. There were plenty of conclusions, but perhaps the most striking was how little batting orders really matter. To wit:
“What I find most interesting is that even with the worst possible lineup, we’re only costing ourselves, 0.31 runs per game over the best possible lineup. Yes, that’s a cost of 50 runs over the course of season, or about 5 wins, but we’re talking about THE WORST LINEUP against the BEST. No manager would bat Sizemore eighth and Marson fourth, but even if you did, you’re talking about fractions of a run per game!”
Indeed. Sizemore and Marson. Those were the days, huh?
Anyway, I recommend you read the whole piece if you haven’t, because there really are some interesting takeaways about what matters and what doesn’t when trying to maximize run scoring. In fact, I got to thinking about batting orders again the other night when Rick asked a question on Twitter:
@WFNYScott @WFNYJon @WFNYTD What’s the line-up Friday when Choo comes back? Santana third and Choo fourth?
We chatted back and forth about who would bat third and fourth upon Choo’s return, and TD touched on it this morning in his recap, but, as always, we should have just turned to Castrovince:
“But it doesn’t appear Cabrera will return to the leadoff spot. Acta said he’s leaning toward having Cabrera at No. 2 and Carlos Santana at cleanup, which means Choo would slot in at No. 3. Suddenly, that’s a much more potent punch early in the order, and Santana […] might begin to see some more pitches to hit.”
I have to say, I like the logic here. Let’s uncover it before running some simulations.
Most of what’s at play here is the issue of “protection.” It’s sort of accepted (if faulty) wisdom that a strong cleanup hitter will help a #3 hitter get more pitches to hit. The thinking goes something like this: a pitcher doesn’t want to face the strong cleanup guy with men on base, so the #3 hitter will see more strikes than balls, because the pitcher is afraid of walking him. A similarly accepted (and faulty) notion suggests that the cleanup hitter is “protected” by having a good on-base guy in front of him in the lineup. For example, once your #3 hitter gets on-base, your cleanup hitter will likely see more strikes, especially if the #3 guy can steal a bag here and there.
Despite what I know about protection in a lineup—by and large it doesn’t exist to the degree we’d like to believe—I still like to think about little tweaks like this. And I think Acta has it largely right: by slotting Choo in the third spot and Santana cleanup, the Indians can leverage Choo’s speed, forcing more fastballs to Santana to control the running threat.
On the other hand, of course, if Choo steals second, why in the world would the opposing team throw Santana a strike when something called a “Jhonny Peralta” might be waiting on deck? But that’ll have to wait for another day.
It’s time for the simulations. If you’ll remember, there’s a calculator out there that will give you the average number of runs any given lineup will score per game. The inputs are on-base percentage (how often a batter makes an out) and slugging percentage (what he does when he doesn’t make an out); this means that speed and baserunning don’t count in this simulation, so if need be, we can use that as a tie-breaker. For these simulations, I’m going to use the following players and numbers:
OBP | SLG | Notes | |
Brantley | 0.349 | 0.363 | CHONE Projection |
Cabrera | 0.350 | 0.404 | Career Averages |
Choo | 0.387 | 0.487 | Career Averages |
Donald | 0.324 | 0.420 | 2010 MLB |
Hafner | 0.360 | 0.410 | 2010 MLB |
Kearns | 0.352 | 0.425 | Career Averages |
LaPorta | 0.338 | 0.401 | 2010 MLB |
Peralta | 0.316 | 0.400 | 2010 MLB |
Santana | 0.431 | 0.547 | 2010 MLB |
And here are the two lineups I ran (notice, only the three and four spots change since we’re trying to isolate the effect of switching Santana and Choo):
Lineup 1 | Lineup 2 |
Brantley | Brantley |
Cabrera | Cabrera |
Choo | Santana |
Santana | Choo |
Hafner | Hafner |
Kearns | Kearns |
LaPorta | LaPorta |
Peralta | Peralta |
Donald | Donald |
The results? Lineup 1, with Choo third and Santana fourth, would generate 5.270 runs per game. Lineup 2, with those two players reversed, would generate 5.256 runs. Pretty darn close, though the edge goes to Choo third and Santana fourth. Throw in the fact that Choo is likely a better a baserunner than Santana, and I think Acta’s decision is the wise one.*
For those who remember the previous piece on batting orders, the first, second, and fourth spots are the most valuable run-producing spots in the batting order, so it’s not surprising that placing our best hitter (Santana) fourth rather than third would increase run-production. Interestingly, the MOST valuable lineups all had Santana leading off in order to maximize his plate appearances per game, but I think it’s safe to say that ain’t gonna happen anytime soon. But remember that batting Santana fourth will cause him to come up fewer times per game than he would batting higher in the order.
_____________________________________________________
I sort of knew going into this analysis that it didn’t matter all that much which order Choo and Santana batted in the lineup. The difference between 5.270 and 5.256 is 0.014 runs per game. Over the course of 70 games (or about what’s left this season) that’s only .98 runs! No big deal, really, and unlikely to cost us a win.
But there was another thing that caught my eye in that Castrovince piece about batting orders: namely, that Cabrera might not lead off for much longer.
For awhile now, Michael Brantley has been thought of as the lead-off man of the future. A large part of that thinking was to get Sizemore into the heart of the lineup, something that seemed anathema to Eric Wedge.
But now we have two “viable” leadoff men in Cabrera and Brantley, and the rest of this season will go a long way toward establishing that role going forward. While the leadoff position is a valuable one—whoever leads off will see more plate appearances than anyone else on the team—the value is largely dependent on a player’s ability to get on base.
So when comparing Brantley and Cabrera as leadoff candidates, what we’re most worried about is their respective ability to not make outs. Any power production (i.e. slugging percentage) is largely a bonus, as is speed on the basepaths: a lead off hitter’s primary job is to get on base. And getting on base is largely a factor of controlling the strikezone, taking walks, and not striking out (a batted ball is more likely to become a hit than a strikeout).
Let’s look at some numbers. First, 2010 totals:
OBP | BB/K | BB% | |
Cabrera | 0.316 | 0.36 | 5.1% |
Brantley | 0.220 | 0.54 | 6.9% |
MLB Average | 0.329 | 0.48 | 8.7% |
Hmm. So this season, both of our leadoff men are below average at getting on base. Not below average compared to leadoff men, but below average compared to EVERYONE. That’s not good. It looks as though Brantley is slightly above average in BB/K ratio (he walks about half as often as he strikes out), whereas Cabrera has way too many Ks for a leadoff guy (he’s struck out nearly three times as often as he’s walked). Both, though, are below average this season at drawing walks, and for leadoff men, that’s just unacceptable.
So why should these two be the ones to lead off? After all, Brantley’s numbers are worse than Lou Marson’s, and Cabrera is well below average at just about everything that matters for a leadoff hitter. Let’s look at their career stats and throw Sizemore in there, just for a comparison (I used Brantley’s AAA numbers to avoid terribly small sample sizes):
OBP | BB/K | BB% | |
Cabrera | 0.350 | 0.53 | 8.6% |
Brantley | 0.364 | 1.17 | 10.9% |
Sizemore | 0.363 | 0.56 | 11.0% |
Now you see why all three are viable lead off options, though Cabrera is certainly the least-suited to the role. Each guy is above average at getting on base—MLB average is right around .330—but Brantley is the best at not striking out (granted, against AAA pitching). Both Brantley and Sizemore draw a similar number of walks, and considering Brantley’s speed, youth, and paucity of strikeouts, he’d be the obvious lead off man moving forward, except for the fact that he can’t seem to do any of these things at the MLB level.
Luckily for the Indians, Brantley’s only 23, and has plenty of time to adjust to MLB pitching. But come Friday, I’m going to be less concerned with the three and four spots in the lineup: I’ll be paying attention to who’s leading off, and why.
11 Comments
Brilliant stuff here. Ideally, I’d like to see Brantley lead off with Cabrera hitting second. Brantley hasn’t done much to this point, but having some consistent, high-level talent hitting 2-3-4 for the rest of the year should give us a good idea of what he could be capable of.
Agree with you that the leadoff spot is just as interesting as the middle of the order. I don’t feel like we should be wasting at bats on Brantley at the top as of right now. Power may not matter in the leadoff spot much but when all you hit are singles, your walk rate is going to plummet because pitchers aren’t afraid to throw you fastballs over the plate all day long. I would like to see Brantley hit in the 9 hole for now, and if he ever gets on base the same guys will be batting behind him anyways. Donald’s numbers are horrible at all spots outside of 9 but I think can handle a promotion to 8.
And while I realize the speed thing is really not a big enough factor to show any statistical difference, but there is another train of thought out there that says you should hit the speedier player behind the better hitter. The Brewers have experimented with hitting Braun behind Fielder at times to make better use of Braun’s relative speed. The thought being that you would not be trying to steal bases (risking outs on the basepaths) with your best hitter up at the plate. Plus as you mentioned in the article, in any crucial game situation with Choo on 2nd, Santana up and Peralta (or Hafner, or a struggling Kearns, or anyone not as good as the Best Catcher on the Planet) on deck, Santana is just going to be walked anyways. I find it hard to believe Choo would be running much with Santana at the plate.
Then again splitting up Choo and Hafner (and in the future Grady) might be better for the the late inning at bats against bullpen (LHP v. RHP).
2011 Opening Day?
Cabrera
Choo
Santana
Grady
LaPorta
Hafner
Goedert
Donald
Brantley
great read for a sleepy july thursday.
my 2011 projected lineup (in June cuz I want The Chis)
Brantley
Cabrera
Choo
Santana
LaPorta
Sizemore
The Chis
Hafner
Donald/Phelps? Keeping 2B warm for Kipnis in ’12
Good points Tommy. Know that most of these conversations are largely the hair-splitting type, since batting order isn’t as big a deal as we’d like to believe.
But I still say we take this year to evaluate Brantley. By and large, we know that Cabrera is a .340-.350 OBP guy. If Brantley can find some success against MLB pitching, he’ll likely be better than that (Cabrera’s MiLB OBP was .351 compared to Brantley’s .387–and Brantley just doesn’t strike out nearly as much). It might take some time for him to make the adjustment, but isn’t that what 2010 is all about? I tend to think long-term that Brantley is more of a leadoff type than Cabrera, and so I’d probably switch Cabrera and Brantley in your 2011 lineup if I had my way.
Again, small differences. And ones that probably matter more as philosophy than actual outcomes.
You would drop Cabrera all the way to 9?
And I agree that 2010 is all about “finding out what we have”, but can’t we evaluate Brantley just fine from the 9th spot too?
Taking the pressure off of Donald at the bottom seems to have helped him, although I don’t know if that had anything to do with it. I do know that baseball is hugely a mental game and I think we should do whatever it takes to put our guys in the best position to succeed.
@3
Any lineup projections for June 2011 might need to include Weglarz too.
@Tommy – good point.
Not to be an instigator, but the simulations assumes that each player will hit certain numbers regardless of the spot they are placed in correct? What about the psychological factor? Not that you would have any insight into Santana’s mind, but wouldn’t putting him 4th in front of “a Jhonny Peralta” create more pressure on a rookie than a veteran like Choo?
@Rick: yeah, the simulations assume consistency no matter where you put them. I haven’t a clue as to the psychological effects of moving around the lineup or how they might play into this, but you make a good point, especially considering that numbers-wise, it makes very little difference as to who bats third and who bats fourth. We’re mostly talking about tie-breakers, like speed, psychology, and comfort-level.
I should also say that when you first posed the question, my gut feeling was to agree with you about Santana staying in the #3 spot since he seems comfortable there.
@Tommy: yeah, Cabrera seems to be a decent #9 hitter in that he could “turn the lineup over” the way Tony LaRussa does in St. Louis (when the pitcher bats eighth). Just an idea, and probably not one that the Indians can afford right now. Brantley can be evaluated in the #9 hole, but as Rick points out (and as many point out all the time), there’s something unique about each spot in the lineup, and if Brantley needs to lead off in the future, he might want to start learning/adjusting now, IMO.
@8 – Santana (like a perfect vacuum) only feels the absence of pressure.
@mgbode,
Good man. I knew somebody could come up with a nice Santana quip related to Rick’s post.