Joe Maddon’s Other Curious Decision

Andrew Miller, for once, didn’t look invincible. After relieving Corey Kluber in the top of the seventh inning, he walked Kyle Schwarber — who answered all of the questions about rust and timing in that fantastic at-bat — and then gave up a single to Javier Baez, loading the bases with nobody out. Down 3-0, this was the Cubs shot at winning Game One, and potentially running away with the series; if Cleveland couldn’t win the home game where Kluber dominated on full rest, they weren’t going to have an easy time winning four more without that ideal setup.

But Miller, being the excellent pitcher that he is, got Willson Contreras to fly out to shallow center field, leaving the bases loaded. Then Addison Russell struck out, and Miller was one out away from getting out of the jam. The final at-bat of the seventh inning seemed like the Cubs last shot to win; a big hit in the gap would tie the game — or a home run would even give them the lead — but an out would end the rally, leaving the team down three with only six outs to go against Miller and the looming Cody Allen.

So when David Ross stepped up to the plate to take his chances against Miller, I was pretty surprised, to say the least.

Ross, by all accounts, is a great dude. Everyone loves him. He’s an excellent defensive catcher, and he’s been an underrated player his whole career. But Ross is also a 39-year-old third string catcher who has hit .188/.291/.356, good for a 70 wRC+, in 558 plate appearances over the last three years. Given his age and overall performance record, Steamer projects Ross for a 71 wRC+ at this point, the kind of mark that is only acceptable from excellent defenders and veteran leaders who add value in ways other than hitting.

This is not the guy you want at the plate with the bases loaded in the World Series, which is why I fully expected to see a pinch hitter in that spot. But Maddon stuck with Ross, Miller struck Ross out, the rally ended, and the Indians went on to win 6-0.

Now, to be clear, this isn’t why the Indians won last night. They won because Corey Kluber was amazing — August Fagerstrom will have more on his dominance later this morning — and their offense put some runs on the board to support his dominance. Andrew Miller is so good that any hitter would have been likely to make an out there. This wasn’t letting Zach Britton sit in the bullpen while your team gets eliminated. And this game wasn’t a must-win for the Cubs, who now will likely be the favorites to win every game from here on out, given that Kluber will likely be starting on short-rest in his next two outings.

So, like the decision to start Chris Coghlan, this is really more of a thing to wonder about than any kind of pointed criticism. But I do have to wonder, if not for that kind of situation, what is Jorge Soler’s point of even being on the roster?

I know Soler has been up-and-down since his scintillating debut in 2014, and it can be frustrating watching a guy with obvious natural abilities fail to live up to early expectations. But Soler, disappointment and all, is a 24-year-old who has hit .258/.328/.434 as a big leaguer over the last three years; that’s a career wRC+ of 106. That’s not amazing or anything, but as a right-handed hitter who does damage on contact, it’s hard to imagine a situation that calls for his skillset more than being down three with the bases loaded and a left-hander on the mound.

During our live blog of the game, the general defense of the move was to point to Ross’ line against LHPs this year; he hit .283/.418/.528 against them, good for a 148 wRC+. But like quoting Coghlan’s numbers since getting traded back to Chicago, this is the kind of small-sample data that misleads more than it informs. For one, we’re talking about 69 plate appearances; anyone can do anything in less than 70 PAs. And despite Ross’ success against lefties this year, looking at the larger picture paints different story.

You know what David Ross’ wRC+ against lefties was last year? 3. Yes, 3. His career wRC+ of 100 against LHPs is basically exactly what we’d expect from a right-handed hitter with a career wRC+ of 93. There’s just no real reason to believe that Ross has special lefty-mashing powers, and like almost any right-handed hitter, a reasonable expectation of his performance against a lefty can be found by taking his current overall projection and adjusting up by about 10 percent. Except that, as a 39-year-old catcher who has been an average hitter over his career, he’s well below average the plate now, so even with the platoon advantage, Ross isn’t a very good option for the Cubs at the plate.

Soler, of course, has the same platoon advantage Ross does, since both are right-handed hitters. And Soler is simply a much better hitter at this point in his career than Ross is, especially when he makes contact. Both hitters have strikeout problems, but when Soler puts the bat on the ball, good things happen. Here are their relative lines when putting the ball in play over the last three years.

Soler and Ross, On Contact
Player PA BA SLG wOBA wRC+
Soler 479 0.374 0.628 0.421 169
Ross 305 0.309 0.564 0.363 129

For comparison, Anthony Rizzo has put up a .425 wOBA on balls in play over the last three years. The difference between Soler and Rizzo, at this point, is entirely control of the strike zone; when they hit the ball, similar things happen. Soler is the kind of guy who, if he got ahold of a mistake from Miller, could have easily tied the game with a ball in the gap. Or given the team the lead with a grand slam, if Miller really hung one.

Ross could have done that too, of course, but it was less likely, even after accounting for the pinch-hitting penalty that Soler would have faced. And it’s not like Ross was more likely to put the ball in play; his 67% contact rate from 2014-2016 is worse than Soler’s 69% mark. His only real advantage is that he’s a bit more selective, and he had a slightly higher chance of drawing a walk, but while a walk would have been nice, you’d still be down two at that point. Hoping for Ross to draw a walk so that Fowler could try and get a game-tying base hit against the best reliever in baseball is really stretching things, even if he did look mortal that inning.

The difference between having a below-average hitter and an average hitter up at the plate in one at-bat isn’t so large that the difference shows that Joe Maddon made a colossal mistake here or anything, but Soler is almost certainly a better hitter than Ross, and if he’s not going to pinch hit for Ross in the highest leverage situation you can imagine with a lefty on the mound, is there a scenario in which he’s going to be useful to the Cubs this series? That at-bat was exactly why you have a guy like Soler on the team, but instead, the Cubs sent their third string catcher up to try and tie the game against the best reliever on the planet.

Most likely, this will all be forgotten in a few hours. The Cubs get to throw Jake Arrieta against Trevor Bauer tonight, and given that they forced Miller to throw 46 pitches last night, they’re not likely to see much of him in Game Two. The Cubs remain the best team in baseball, and they still have a very good shot of winning the World Series.

And Jorge Soler probably would have struck out against Miller too. But given the alternative, I’m not really sure why we didn’t get to see him try.





Dave is the Managing Editor of FanGraphs.

57 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
bbmoneymember
6 years ago

I don’t know that’s a nice number of plate appearances Ross had v lefties this year.

MPRox
6 years ago
Reply to  bbmoney

I didn’t know Rob Gronkowski read Fangraphs!

Bat
6 years ago
Reply to  bbmoney

Almost certainly mentioned above and I missed it, but Ross was 3 for 5 against Miller before last night’s at bat so this may have factored into Maddon’s decision as well?

Obviously 5 at bats spread out over a period of years – some of them when before Andrew Miller became Andrew Miller – isn’t a good indicator of the outcome of this particular at bat.

But I’m sure Maddon knew that Ross was 3 for 5 against Miller in his career and that may have impacted his decision.

Apologies if this was mentioned in the article and I didn’t see it.

bbmoneymember
6 years ago
Reply to  Bat

All of them were before Andrew Miller became ANDREW MILLER. All from 2010, when Miller posted an ERA north of 8 and a WHIP of 2.3+. Not to mention, 5 at bats. It wasn’t mentioned, because it wasn’t worth mentioning I’d wager.

Bat
6 years ago
Reply to  bbmoney

Yeah, I don’t know bbmoney.

I think it wasn’t worth mentioning for Cameron, and I’d agree with his SSS / ancient history reasoning.

But Maddon definitely would have known about the 3 for 5, and I’m not so sure it didn’t impact his thinking even if he didn’t say anything about this postgame, or at least I don’t think he did.

Managers tend to think about these match up / game strategy things a bit differently than stats guys. See Showalter, Buck and Britton, Zach.

Dave T
6 years ago
Reply to  Bat

I highly recommend reading this post – http://www.fangraphs.com/blogs/when-you-should-ignore-the-data/

TLDR summary: “Batter/Pitcher match-up data has been shown to have no predictive value. ” Even when a batter has extremely dominated a pitcher over 20-30 PA’s (or vice versa), the actual future results have been basically average.

Tom Jitterbug
6 years ago
Reply to  Dave T

I don’t think Bat was suggesting that the batter/pitchup matchup history was worth considering, just that Maddon may have used this (flawed) statistic when making his decision.

Bat
6 years ago
Reply to  Tom Jitterbug

Exactly – thanks tomjakubowski.