Two Managerial Decisions and Another Questionable Intentional Walk

As you may already know, I’m something of an intentional walk connoisseur here at FanGraphs. When questionable ones occur, particularly in the playoffs, I like to delve into the specifics to figure out which ones are good decisions, which ones are close calls, and which ones are just plain silly.

Earlier this week, I wrote about Joe Maddon’s bases-loaded intentional walk, which was about as far on the silly end of the spectrum as you can get. Today, I’m going to cover the other notable intentional walk of the week: the Yankees giving Miguel Cabrera a free pass on Thursday. Then, as a bonus, I want to talk about Cardinals manager Oli Marmol and a clever thing he did that might escape notice if you aren’t watching closely.

Miguel Cabrera

Let’s set the stage. Runners on second and third, two outs in the bottom of the eighth, Yankees trail by a run. New York had Lucas Luetge on the mound, who got the previous batter, Jeimer Candelario, to ground into a bases-loaded double play. That left him in a matchup with Cabrera. Luetge is a lefty; Cabrera is a righty, and Austin Meadows, the next hitter up, bats left-handed. Aaron Boone called for an intentional walk.

My snap surface-level decision: I didn’t hate it. Getting the handedness advantage is a big deal. Cabrera’s run doesn’t matter much; if he scores, the Yankees are in a world of hurt anyway. The only downside is loading the bases so that a walk would force in another run. Cabrera is much better against lefties, Meadows has a meaningful platoon split, and Luetge has been far better against lefties than righties in his career. Easy peasy!

As I think about it more, though, I’m less certain it was the right move. Cabrera has a meaningful platoon split, and over a huge sample size; he’s roughly 6% better against lefties than righties. He’s a worse hitter than Meadows at this point in his career, but you could see the numbers working out, particularly given that Luetge has allowed a .339 wOBA to righties and .238 wOBA to lefties, albeit in a limited sample. But wait! Look under the hood a bit longer, and I don’t buy it.

Cabrera’s career platoon split is meaningful, but it’s almost entirely due to walks. Take a look at some stats that don’t involve walks:

Cabrera has walked 15.7% of the time against lefties and only 9.4% against righties. It’s a huge tailwind to his overall line, but that doesn’t matter in this situation. An unintentional walk here shouldn’t affect the decision at all; if you’re deciding between purposefully walking someone or not, well, “they might walk otherwise” isn’t a good reason to do it.

That’s not to say that there’s no reason to issue the walk. Luetge has been meaningfully worse across the board against righties, and his cutter/slider pitch mix generally seems like a bad plan for attacking opposite-handed hitters. Meadows strikes out far more often against left-handed pitching. Avoiding Luetge’s weaknesses and highlighting Meadows’ shortcomings is a worthy goal.

Does it all make sense? I plugged the particulars of each player into a simple batting simulation to find out. I regressed everyone to league average based on how many batters they’ve faced; we can be more sure about Cabrera’s true-talent splits than either of the other two players and more sure about Luetge than Meadows because pitcher platoon splits become reliable more quickly.

The Meadows/Luetge matchup marginally favors the Yankees. Per my calculations, which leave room for error, Meadows projects for a .282 OBP against Luetge. OBP is what matters most here, because pretty much any run makes a Yankees comeback quite difficult. But Meadows has power; he still gets his fair share of doubles and homers even with that poor OBP mark. In aggregate, it works out to the Tigers winning the game 88.2% of the time if Meadows faces Luetge with the bases loaded, quite close to the 88.5% our game logs showed.

So, that’s the likelihood if the Yankees put Cabrera on intentionally. If they don’t, they’ll still end up with that outcome some portion of the time thanks to an unintentional walk. I repeated the same outcome-by-outcome math with Cabrera against Luetge to estimate the team’s likelihood of winning there. Cabrera projects for a solid .318 OBP against Luetge, but OBP doesn’t matter for him in the way it would for Meadows. The bases weren’t loaded, so a walk wouldn’t do much. Walks aside, Cabrera isn’t a terrible batter to face at this point in his career. We project him for a batting average in the .250s, even with the boost of facing a lefty. It’s not as simple as comparing Cabrera’s batting average to Meadows’ OBP, because when Cabrera walks, the Tigers get to send Meadows to the plate, but even those situations result in outs more than 70% of the time. Not having the bases loaded, it turns out, is a big deal.

In aggregate, the Tigers stood to win the game 87% of the time when Cabrera stepped to the plate. Their odds of winning increased when the Yankees walked him. It’s not so much that Meadows is a better hitter than Cabrera even with the platoon disadvantage, though that’s likely true. The bigger issue: don’t walk the bases loaded! Pitchers walk people sometimes. When you really can’t afford to give up a run, adding a situation where you give up a run is asking for trouble.

The sum total of all of this? The Yankees cost themselves a sliver of a win (probably) by declining to face Cabrera. Meadows punished them with a double that pushed Detroit’s lead to 3–0. No one got to see Cabrera go for his 3,000th hit. Everyone lost, except Cabrera, who was loving it. “No, my on-base percentage went up,” he said when asked whether he was upset about the walk. At least someone had fun!

Oli Marmol

This is just going to be a quick hitter, but I had to say something about a managerial sleight of hand that I actually love. Blink, and you might have missed it. With a 2–0 lead over the Brewers headed into the sixth inning last Saturday, Marmol brought in Génesis Cabrera, a lefty, to face a trio of right-handed hitters. Keeping score at home, it’s easy to say that’s bad. Lefty to face righties? What a joker. But, as Marmol told The Athletic’s Katie Woo after the game, it wasn’t so simple.

“Their best shot there is to unload their bench, they’re itching to do it. Our job going into tonight’s game was to keep those four guys (Tellez, Yelich, Narvaáz, Peterson) on the bench… We thought it’d be a better place for Cabrera and keep the lefties on the bench and then hand it over to Giovanny Gallegos for their top three guys, because they aren’t going to pinch-hit for them.”

Honestly, no notes. It’s not a huge thing, but making small decisions like that is how managers accrue value. Cabrera has reverse splits in his career and is likely a low-split pitcher based on his excellent changeup. The bottom of the Brewers’ lineup on this day was full of light-hitting righties. By starting with a lefty on the mound, Marmol gave Craig Counsell two bad choices. Leave your righties in to get the platoon advantage, and you’re wasting some valuable bench bats. Bring in the lefties, and you’re giving the Cardinals a lefty/lefty matchup. In the end, only one of Milwaukee’s four lefties batted in the game, and the Cardinals held on to win.

It’s great! I love little decisions like this, whether the manager explains them or not. Managing isn’t always easy, and it’s full of judgment calls. Sometimes you think that the little edges that are hard to calculate from the outside turn what looks like a bad decision into a good one. Sometimes which pinch-hitters are available is more important than who’s batting. Just don’t leave a future Hall of Famer sitting short of a milestone for marginal gain, please; that was no fun.