So You’ve Decided to Intentionally Walk Shohei Ohtani… Again
John Schneider loves intentional walks. He intentionally walks Aaron Judge more than anyone else. He intentionally walked Cal Raleigh with the bases empty in the ALCS. So he must have felt very strange when the first two games of the World Series passed without a single intentional walk of Shohei Ohtani, a man who has been intentionally walked repeatedly this postseason even though he was mired in a deep slump early on. Now that he’s hotter than the sun, Schneider was no doubt ready to go to his preferred tactic as soon as the situation presented itself.
And oh, did it present itself! Yesterday, the Dodgers and Blue Jays played 18 innings to settle Game 3. Ohtani opened the game with a double, a home run, another double, and another home run, the last hit a game-tying solo shot in the seventh. That second homer set up a perfect storm. Extra innings with the Dodgers batting in the home half meant that Ohtani represented the tying run every time he came to the plate the rest of the way, and you don’t have to roll out the red carpet for Schneider; he’s always ready to deploy some tactics. Ohtani had five more plate appearances in the game; Schneider intentionally walked him in four of them. I did the math to see whether those were good decisions, and how much they affected the outcome of the game.
Intentional Walk Number One
The Situation: No one on, one out, bottom nine, tie game, Jeff Hoffman pitching
The first intentional walk of the night came after Ohtani had worn out Blue Jays pitchers for eight innings. By the time he came to the plate with a chance to walk it off, Schneider was tired of it. He waved Ohtani down to first base, preferring instead to get the platoon matchup and face Mookie Betts.
I plugged Ohtani, Betts, and Hoffman into my matchup model to come up with estimates for each batter/pitcher matchup. Perhaps unsurprisingly, Ohtani/Hoffman looks very good for Shohei, to the tune of a wOBA around .400 and home runs in 7.3% of plate appearances. With an average batter at the plate, our WPA Inquirer gave the Dodgers a 58% chance at a victory. Plug in Ohtani, and that number climbs to 60.1%.
The problem is that with a runner on first and one out, good outcomes for Betts get much better. An extra-base hit would now often end the game, while a walk would put Ohtani in scoring position. Even with a meaningfully worse projected batting line – .265/.355/.578 for Ohtani versus .241/.354/.449 for Betts – the Dodgers’ chances of winning rose to 63.9% after the intentional walk.
Conclusion: Intentionally walking Ohtani with one out is a bad plan, even if he gets caught stealing later in the inning.
Intentional Walk Number Two
The Situation: No one on, two outs, bottom 11, tie game, Braydon Fisher pitching
The second intentional walk felt automatic; if Schneider was going to intentionally walk Ohtani with no one on base and one out, he was certainly going to do it with no one on and two outs. There was a righty with muted platoon splits on the mound again, but that didn’t seem like an important part of Schneider’s calculations.
This time, Ohtani projected to do extremely well against Fisher; .270/.379/.549, in fact, as compared to a .257/.393/.447 line for Betts. Once again, though, putting a runner on base and letting Betts end the game with an extra-base hit looks bad for Toronto’s chances when you run the math. With Ohtani at the plate, the Dodgers had a 55.3% chance at a win. When Schneider walked him to bring up Betts, that winning percentage rose to 57.6%, even though the Betts/Fisher matchup was clearly better for Toronto. Putting a speedy runner on first, for free, just has too many ways it can go wrong.
Conclusion: Intentionally walking Ohtani with the bases empty is a bad plan.
Intentional Walk Number Three
The Situation: Runner on third, two outs, bottom 13, tie game, Eric Lauer pitching
The third intentional walk? A gimme. If you’re going to walk Ohtani with no one on base, surely you’ll walk him with the winning run 90 feet away. There was one complication, though; Lauer is a lefty, so the matchup with Betts wouldn’t be good. Schneider had the answer: He walked Ohtani and Betts, loading the bases for Freddie Freeman.
It’s really hard to make the math look good when you intentionally walk the bases loaded. The other team can now score a run with a walk. Freeman has a career .386 OBP; Ohtani has a career batting average of .282. You can see the issue here. Indeed, with Ohtani batting, my spreadsheet says that the Dodgers had a 64% chance of winning. After walking two batters to load the bases for Freeman, it sees Los Angeles with a 67.1% chance of prevailing.
Conclusion: Intentionally walking Ohtani and Betts to load the bases for Freeman is a bad plan.
Intentional Walk Number Four
The Situation: No one on, one out, bottom 15, tie game, Eric Lauer pitching
Well, clearly this one isn’t gonna be good. We already know that walking Ohtani to get the platoon advantage doesn’t make sense with no one on base and one out. Obviously, then, walking Ohtani to have Lauer face Betts and lose the platoon advantage won’t look good here.
More specifically, I had the Dodgers at a 60.5% chance of winning before the walk, as compared to a 64.5% chance of winning after the walk. That’s a huge increase in winning percentage, and it makes sense to me; Ohtani is worse against lefties, Lauer is better against lefties, and we’ve already run the math on one-out, bases-empty walks. This one just had to be worse.
Conclusion: Yeah, don’t do this!
Add everything up, and Schneider cost the Blue Jays an aggregate 0.132 WPA with his intentional walk decisions; in other words, he lowered their chances of winning by 13.2 percentage points. That’s enormous! Teoscar Hernández went 4-for-8 with a home run and only racked up 0.131 WPA. Freeman went 2-for-7 with two walks and hit a walk-off home run in the bottom of the 18th, the biggest play of the game by a mile, and that performance was worth 0.118 WPA. Ohtani was the best player on the field Monday night. Schneider did more to advance Los Angeles’ chances of winning than any other Dodgers hitter, though. That’s not what you want as a manager. Sure, maybe Ohtani would have hit a third, game-sealing home run in one of these at-bats. Maybe he would have ripped a triple down the line. But maybe he would have made an out. He does that more than half of the time! It wasn’t what ended the game, but Schneider’s unwillingness to find out what was in the mystery box made the Blue Jays’ road meaningfully harder last night.
Ben is a writer at FanGraphs. He can be found on Bluesky @benclemens.
Great article! Small correction: ‘Extra innings with the Dodgers batting in the home half meant that Ohtani represented the tying run every time he came to the plate the rest of the way’. He represented the winning run every time.
What if the Blue Jays scored 2 runs in top of the 10th (or any other extra inning) and Ohtani lead off the bottom half? Ohtani is neither the tying nor the winning run.
The blue jays weren’t going to score in extras if the game went 100 innings
What! They had Vladimir Guerrero and…Daulton Varsho! Ernie Clement! The hot dog vendor! Some guy they dragged out of the Dodger stadium men’s room! A Jonas brother!
This doesn’t refute this fun article but – the Dodgers issued the second most intentional walks this year (33, behind only the Angels with 44) – Toronto only 25 (6th most in baseball).
As much as Schneider loves him some IBB, Ron Washington could not get enough.
Pretty crazy that it ended up being correct enough in each case.
I think it’s fairer to say it ended up “working out” in each case, rather than correct
If Schneider had walked Ohtani the first 4 times Jays would have won in regulation.
That was the real blunder
Can you imagine Scherzer’s reaction if, to open the bottom of the 1st, Schneider had just held up four fingers?
Maybe we get to see Bieber’s reaction, as Schneider has said he’s just going to intentionally walk Ohtani for the rest of the World Series. I assume he’s bluffing/playing mind games there.
I’m laughing
I like imagining Scherzer’s reaction to just about anything.
“Other toll booth lane moving faster” is a particular favorite.
Getting to McDonald’s at 10:32 and they stopped serving breakfast two minutes earlier.
‘Sorry man, I know you’ve been waiting here with your groceries for a long time, but this checkout is closing now. Maybe try the one down there?’
Or it could be like the time that Gerrit Cole, on his own, decided he’d had enough of Raffy Devers for one lifetime, and waved the four fingers himself.
I don’t know what they told Seranthony, but I can’t imagine it was “challenge him with middle middle heaters.” I think after that dinger Schneider just decided it was too hard to pitch him tough
It was funny watching the pitching have that serious discussion with him and then the first pitch he throws is a fastball right down the middle.
In a vacuum, yes, I agree with your gripes on all 4. But Shohei is hotter than Hades and Mookie has apparently been body snatched by the ghost of Julio Lugo. I think the 2nd walk at least was justifiable. The first and 4th were borderline.
I will agree that walking Shohei AND Mookie to get the bases loaded for Freddie was completely insane when a hit, a walk, a HBP, or a wild pitch ends the game.
Just going to note that a hit or wild pitch (barring a weird bounce right back to the catcher) ends the game regardless of whether you IBB Ohtani and/or Betts
I’d rather face Ohtani with the platoon advantage than Betts without the platoon advantage, but I’d REALLY rather not face Ohtani.
If you’re going to walk Ohtani, which you are, you walk Betts too. Made total sense to me and I was surprised when Betts got most of the way through his routine before getting sent.
What are they using for the Betts projection? The first walk says – 241/.354/.449. But he’s hitting 250/338/350 in the playoffs. They can assume regression, but he slugged 406 in the regular season. Expecting a 449 slugging here or a 350+ OBP seems like the bigger problem in this analysis.
yeah, using context neutral projections for this analysis isn’t “wrong” per se, but the reason teams have stat guys to put helpful probabilities on dugout iPads AND still hire biased, instinctive human managers is that the guys on the field are humans, not RNGs. You really want to tell, for example, Rookie™ Braydon Fisher, who’s had bad results in the playoffs, to go get God out with the platoon disadvantage when Mookie’s looked bad all night? Even knowing that averages matter and that meat computers do crazy things, getting unquantifiable buy in and confidence from the players matters a lot.
I’m old enough to remember when Ohtani’s Ks were the story!
Remember back when Smart Baseball People were convinced he was never going to hit in MLB, and that he should just pitch? That makes me feel old.
I love this! Finally someone who makes an argument with actual math and numbers. I was beginning to think I was on an island making these arguments. Great job Ben!
Even with this knowledge, I would have done it each of those times.
I’m still stunned the guy reached base 9!!! times in one game without making an out; that is surreal.
Everyone on this site gets the math and is no fan of the IBB in 99% of cases, but gee it’s hard to argue with the results. Ohtani was absolutely smashing any mistake to some far outer reach of Dodger stadium. Do you trust your 4th, 5th best reliever to not groove one to him? I know the stats account for pitcher/hitter match ups, etc. etc, but when you see a guy rip 4 consecutive XBH, I don’t know, you’ve got to consider that maybe this guy is just seeing the ball well and hitting any mistakes at like 110mph each time.
Yep. I’m of the belief that him and Judge are in their own category in the sense that when they’re white-hot, sometimes you just have to avoid them, regardless what the math says.
“gee it’s hard to argue with the results”
Not really, imo.
A major underpinning of the analytic approach espoused by this site, and like-minded sites/individuals (including essentially every MLB front office at this point), is that you can’t judge the ex ante wisdom of a baseball decision based on a handful of ex post outcomes.
This is why I think it’s a little silly that I’m seeing people say MLB needs to change the rules to penalize intentionally walking a superstar.
There is already a massive penalty for intentionally walking Shohei Ohtani with the bases empty: intentionally putting the winning run on base!
Well as with most teams, there are plenty of silly Dodger fans
All this makes me wonder what would happen if it became a universal practice in baseball to routinely issue intentional walks to the best sluggers like Ohtani, Judge, etc., such that they almost never got to swing and potentially put the ball in play.
Would the MLB permit this to go on indefinitely, or would someone (e.g., the Commissioner, Congress, Trump) step in and try to Make Baseball Great Again by eliminating intentional walks?
Would it even be possible to solve this problem, without taking extreme measures? Eliminating intentional walks wouldn’t really work because pitchers could still issue unintentional intentional walks, and the only way to eliminate unintentional intentional walks would seemingly be to eliminate walks altogether.
But that wouldn’t really be fair to hitters unless you were to also eliminate strikeouts, and dictate that plate appearances could only end when the batter puts the ball in play or is hit by a pitch…
I mean, there is one notable historical example we can look to of it being a universal practice to routinely intentionally walk the best slugger in the world. It kinda resolved itself when teams figured out that it usually didn’t help them.
I mean, Bonds led the league in IBB 7 straight years — a streak that ended after the 1998 season.
Amusingly, in Bonds’ 73 HR season, he was walked intentionally fewer times than Aaron Judge was this year (36 to 35).
You could make walks the hitter’s option to accept or decline? Would be funny to see a pitcher throw 40 pitches out of the zone, after each one the batter just shakes his head, gives a thumbs down, and steps back in. TEST OF WILLS!
I ran the numbers several years ago (in 2016), and my calculation was that a hitter who walked every PA would have .691 wOBA. (For comparison, Judge’s highest single-season wOBA was .476, in 2024.)
Over a full season of PA’s, I calculated that a player who always walks would generate ~22 WAR from batting value. That’s before any adjustments for position/defense or baserunning value.
So that’s the mathematical challenge to overcome if issuing intentional walks routinely. By linear weights, an IBB represents an opponent making the decision to gift a batter an outcome that would make said batter the most valuable hitter ever to play MLB – by a very wide margin – if he could get that result in every PA.
So it takes a lot of context, such as the run represented by the hitter mattering very little to the game’s outcome, for an IBB to be a good decision.
In the last 5 games Shohei’s had 10 hits in 19 at-bats.
All but 1 have been for extra bases.
His ISO is over a thousand.
If you let him swing you’re risking losing the entire game to gain 1 out with basically coin flip odds.
That’s insane.
He’s averaging 1.7 bases per At-Bat!!!
This is undoubtedly a great article.
But as I was reading, one thought kept coming back to me:
if this really was a bad plan, then why did it end up working out so well — at least for the Blue Jays, from a results standpoint?
It makes me wonder: is baseball ultimately more about people, or about data?
I’ve heard that the Yankees have built one of the largest analytics departments in MLB — perhaps the biggest, or close to it.
Yet looking at what’s happened to the team over the past few years, it does make you think.
It worked out well because the sample size was 4 and not 10,000
Expanding on Francis’ very good point about sample size:
it worked out well because, after those 4 intentional walks, the remaining Dodger hitters in those innings went 1-6, with the 1 hit being a single. (There was also an intentional walk to Betts.)
Ohtani also made an out on a caught stealing after one of those 4 intentional walks.
A slash line of .167/.167/.167 (excluding the intentional walk to Betts) is obviously well below the true talent level of Betts and Freeman.
This isn’t refuting the article’s argument whatsoever. The proof is in the data shown. If Schneider managed by the numbers, he shouldn’t have walked Shohei in these situations.
Taking a bigger perspective for the game as a whole, however, what is ultimately going to be better for the game of baseball: a game in which a player who is so immensely talented that the opposing team would rather walk him 5 straight times to face two other back-to-back hall of famers, or a game in which a player is only intentionally walked based on the fractional WPA? To me, the answer is easily the former.
I can’t imagine that it’s better for the game to have the most exciting player in the league have the bat taken out of his hands repeatedly.
Every Yankee fan can attest to this, or late 90s/early 00s Giants fan, etc.
For a single game, sure, I could see how this plays out. The reality is that this scenario doesn’t actually occur over the course of an entire series, let alone a season. Bonds still averaged 150 hits and over 50 HRs a year when he was averaging 200 walks a year. Judge is still getting 175 hits and well over 50 HRs over the past few years. They still put up better counting stats than anyone else, which is the reason they’re being pitched around in the first place.
We know that the math never works out to make IBB a good decision using projections. However, this method assumes that baseball events are independent, but we also know that the events are not completely independent. The models work over large sample sizes, but they are not actually independent. I propose a bet: Given the projection for Ohtani and some small margin on either side of that projection, on that night in those PA would you bet the over, under, or push on those projections? I’d bet the over, and I don’t think it is close. Would I bet his actual project in those moments (which we have no way to calculate) is high enough to make the math work out? I don’t know, but I believe very strongly that his actual talent on that night was significantly higher than his season stats are.
If you have a model that works over small sample sizes, you likely would have better things to do with your time than post here.
If you’re looking at a bases-empty IBB, the math is pretty simple: what is the likelihood that Ohtani gets an XBH? A single, HBP, ROE, C interference or unintentional walk is functionally a push (there’s a very small penalty because that reliever threw a few more pitches in anger), and obviously, any out is a win.
For simplicity, the XBH estimates ignore the chance of a multi-base ROE.
With bases empty/1 out:
A double moves the win pct needle about 6% over the IBB; a triple about 19%. A HR is obviously game over so it flushes whatever your current win percentage is away (approx. 37%). The out nets you about 10% (and since the bases are empty, the type of out is irrelevant).
Which boils down to the IBB makes sense in context of the single game if:
(2B/PA estimate)*(0.06)+(3B/PA est)*(0.19)+(HR/PA est)*(0.37) > (Outs/PA estimate)*(0.1)
You also do have to account for series WPCT%, and that’s where it starts moving more toward issuing the walk. Whatever the delta between being up 2-1 and down 2-1 is basically the “value” of the game.
I was nodding along with your very good analysis, until the last paragraph. Where I’m either overlooking something, or disagree.
I don’t see how series win % impacts the IBB decision. Either the Dodgers or Blue Jays will win Game 3, and I don’t see any quantifiable impact on future games from deciding whether or not to IBB Ohtani in Game 3.
I can see how series win % impacts a decision such as reliever usage. In that case, series win % is relevant for determining a situation’s leverage in the context of winning the series (the overall goal, of course). But the IBB isn’t that type of decision: it’s a discrete tactical decision.
Interesting. Watching the game though, I have no doubt Shohei would’ve homered if they pitched to him. Betts they have handled easily. I think Schneid’s instincts were right here.
I love analytics, but come on, WPA be damned. Did you see how Ohtani was smashing the ball? Do you know how bad Betts has been? Would YOU have pitched to Ohtani in extras there? The WPA argument is a large sample size argument – i.e. an IBB is the wrong statistical call over the course of however many sample iterations. But we’re not playing the full season here. It’s one game in the World Series. Ohtani is hotter than July. Would you give him the chance to beat you there, or take your chances with someone else?
This is a situation in which, even though analytics has improved the game, the eye test is still valuable. I know I’ll probably get a lot of rabid downvotes for this, but really, Schneider’s decision to IBB Ohtani so much make sense. Can you imagine if he let Ohtani hit 4 times in extras?