keyboard_arrow_uptop
Baseball Prospectus is looking for a Public Data Services Director. Read the description here.
Image credit: © Kim Klement-USA TODAY Sports

Welcome to Ohtani Week: a celebration of, well, Shohei Ohtani. There’s been no player more fascinating or exhilarating since Ohtani graced our shores in 2018. Over time, the initial curiosity and excitement surrounding MLB’s first true two-way player in a century morphed into something more: Pure, uncut awe derived a superstar breaking barriers previously thought unreachable. All week, we’ll be talking about the most lovable—and possibly most talented—man in baseball. So throw on your Angels cap, grab your laptop charger, and dig in.

Earlier this year, I had a conversation with former Baseball Prospectus editor-in-chief Ben Lindbergh. We briefly discussed Shohei Ohtani and how Ohtani’s value seems hard to capture. “Ohtani is gonna break WAR,” I told Ben.

Everyone understands that Ohtani is something special, at once being one of the best power hitters in the game while at the same time acquitting himself just fine as a Major League starting pitcher for the Los Angeles, California Angels of Anaheim, California, which is near Los Angeles in California. There have been players who have played two ways. In fact, Michael Lorenzen of the Cincinnati Reds has played a few games in the outfield in addition to pitching relief. But for the most part, two-way players have been good at one half of the game and barely adequate at the other. In that sense, Ohtani is a marvel and has quickly become the go-to answer for just about any baseball hypothetical that could double as the plot for a sci-fi movie. It’s now a race for second place to see which player I would clone 26 times.

Ohtani is breaking barriers and will probably break the arbitration system (how do you even find a recent comp for a player who did… this?) But does Shohei Ohtani break WAR?

It’s not a matter of measuring on-field value. We know how to assess pitching performance. We know how to assess hitting (and baserunning). We know how to add two numbers together. Is there more to it than that? Well, there’s the roster spot that Ohtani saves the Angels by being a two-way player. They can use that spot to employ another player to… do something. That has to have some real value, right?

Venerable Sabermetrician Tom Tango suggested that the value of the roster spot is “a rounding error” in the grand scheme of things, but I’m not so sure I buy that. Maybe WAR was already a little broken and Ohtani was just the drop of cold water that fractures the windshield, but I think that we need to appreciate what the two-way nature of Ohtani’s abilities has meant to the Angels, and what it could mean to other teams.

And how WAR is completely unprepared to deal with that.

***

WAR is based on a simple premise. What would happen if a player had vanished before the season began and the team had to make do with whoever else happened to be already in their organization? Had Ohtani decided to walk away before the 2021 season, someone would have made those starts and taken those plate appearances.

But we also don’t want to give the player any credit (or blame) for their teammates. And so we use a bunch of #GoryMath to create an artificially context-free, neutral environment for evaluating players. WAR assumes that if the Angels had lost Ohtani the pitcher, they would simply call up someone from the minors or the waiver wire and have them take over those innings. Similarly, if the Angels lost Ohtani the hitter, they would probably promote the best of their bench hitters into Ohtani’s customary DH role and call up someone from AAA to sit on the end of the bench.

When we look at how replacement-level pitchers actually pitch though, we see that they don’t actually end up covering all those innings. They aren’t as good at recording outs (that’s why they’re depth signings and fringy prospects) and so they don’t pitch as deep into games as actual good pitchers do. It means the team would need to pull more out of their bullpen both in the game and as part of their overall seasonal workload, which is going to have knock-on effects. WAR just doesn’t do a very good job with those sorts of real-world wrinkles. It’s good as an abstract way to look at individual player value, but it can trip up when we want to know a player’s value within a specific context.

What would the Angels do if Ohtani were to go on the injured list? Well… in the case of the 2021 Angels, they’d be in a tough spot. The Angels have played Ohtani as part of a six-member starting rotation, which has granted some of their other starting pitchers more rest than they would otherwise have gotten. It’s hard to know what the effect of that is/was on Andrew Heaney or Dylan Bundy or Griffin Channing, but one has to assume it’s positive. The Angels could simply revert to a five-member rotation and call up a hitter to take the place of their injured DH. Or they could call up another starter to maintain the six-spot shuffle, but then they’d be playing with a shorter bench than usual.

But that’s the 2021 Angels. WAR doesn’t want to give the player credit (or blame) for who else is on the team and what they do around the player. Another team might have played Ohtani as a DH/pitcher, but in a five-member rotation and used the extra roster spot for another hitter or another reliever. Or maybe an extra “clubhouse influencer.” There are a lot of different things that Ohtani’s wonderous double talent could allow for. So if Ohtani were to disappear, in addition to losing Ohtani’s performance, the Angels (or any other team that employed Ohtani) would lose the flexibility that Ohtani brings. We don’t have a word in baseball that describes that value. 

If I might simplify the Ohtani problem a bit: Suppose that a team had a regular starting third baseman who was fully capable of handling shortstop and doing just fine there. The team just already happened to have a good shortstop. Well, the team now has an interesting bit of flexibility. When trying to figure out their utility infield situation, they could realistically look at a player who had a better bat and could play at third a time or two per week, but wasn’t a good idea as a shortstop. Maybe that player doesn’t end up being out there and available, and they end up signing a more traditional former shortstop who can still hit .240, and that third baseman might never step foot into the short spot. But they had the option, and that option was given to them by the versatility of the third baseman. There’s value in not having choices foreclosed on you before you even have the chance to make them.

So let’s come back to reality. The Angels used Ohtani’s flexibility to implement a six-member starting rotation, partly because that was Ohtani’s own preferred schedule. They had other options on what to do with that spot, and we could probably do some #GoryMath about the relative value to the average team of going to a six-spot rotation vs. an extra platoon hitter vs. a designated pinch runner vs. yet another reliever. And we’d probably come to the conclusion that the options all had about the same value and that it wasn’t much, and so as a result, the average value that Ohani’s flexibility would add to an average team would be “a rounding error.” After all, the player who would be brought up to this “extra” roster space is likely to be a marginal player, who lives in that shadowy “replacement level” zone. 

And this is where the abstract, context-free vacuum that defines WAR comes back to haunt us. I think that there has developed an idea that all players who are on or around “the bubble” of an MLB roster are all one undifferentiated mass of “replacement level” performance. That’s how WAR views it, but WAR isn’t real life. It’s an abstract, context-free vacuum.

Let’s make that a little more real. In 2019, there were 451 players who had at least 100 plate appearances. Using BP WARP, from the number one spot (Ohtani’s teammate, Mike Trout) to Victor Reyes in the number 255 spot (30 teams with 8 position players, and 15 DHs are our “starters”), the range was from Trout’s 8.9 wins to Castro’s 0.6. From spot 256 (Rio Ruiz) to spot 390 (30 teams with 13 position players means that spot 390 is the very end of the bench, where we find Adam Haseley), we range from 0.6 to -0.2 wins. For the bubble players, we can range downward to the end where Stevie Wilkerson posted a -1.6 “win” season in 2019. It’s probably not fair to assume that anyone would fall on those hard of times, but we see plenty of players in the -0.7 range. There’s some variation in those buckets and those are only the players who stuck around long enough to get to 100 PA. 

We can do the same thing with pitching and among pitchers in 2019 who had at least 20 innings of work. In spot 390 on our WARP sort, we see Dario Agrazal with -0.2 WARP. If 30 teams have 13 pitchers each, then that’s roughly where the line of marginality begins. And it ranges all the way down to… well, I’m sure Glenn Sparkman (-4.2 WARP!) doesn’t want to relive 2019, but we can range down into -1 WARP territory easily.

If the marginal player who would get a roster spot were just chosen at random from these bins, which is effectively how WAR models do it, then we’d expect something around the median from within those bins. But the Angels aren’t actually selecting from a random pool of replacements. They’re selecting from within their own organization. Ohtani’s dual nature gave them several different options to choose from in what they would do with that “extra” roster spot. The players who would be the marginal beneficiaries are known to them, and one of them is probably a better option than the others, maybe the one who is a little better than that median. Another team might go in another direction based on what they had on hand, but they’re not going to pick the median option. They’re going to pick the best available one, whatever that is.

Or they might do what the Angels did where the actual value of the “extra” roster spot isn’t in who fills it, but in the fact that it allows them to give their starters longer breaks, or escape a bullpen day, or roster a platoon bat that keeps a key player out of certain situations. Sometimes what doesn’t happen is as important as what does happen. The value in Ohtani’s so completely filling two entire roles on a Major League roster comes from the range of options that are then available. The more options you open up, the more likely that you have something that you now can do but previously couldn’t that is actually an upgrade. Ohtani essentially provides his team with a wild card, to be deployed as demanded by the needs of the moment. Because WAR is team-agnostic, it doesn’t have the ability to incorporate that value.

And we can start to see how that value can be a few tenths of a win. That’s not huge, but it’s real value. The WAR model is based on what is becoming an obsolete idea. Teams are moving away from the idea of a defined starter at each fielding position to pair with dedicated backups, and even away from the three-times-through-the-order starter. As flexibility within a roster becomes the way that teams are put together, it’s a good idea to remember that flexibility has value, but that WAR isn’t set up to capture it.

If this is Shohei’s superlative week, let’s applaud one of the greatest feats of positional flexibility out there. Shohei Ohtani might not be the player who breaks WAR, but now that there’s an honest two-way player in MLB, we need to have an honest conversation about whether our metrics are up to the task of valuing them.

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
morillos
7/21
Given just the basics of what you say here (all totally reasonable), Ohtani's WAR value is kinda more like a WA(Rx2) value, where, R, as you point out, is something a bit over 0. Yeah, and that's not really math, and real math would be beyond my ability to construct, but I think you're right: Ohtani as WAR(Pitching) + WAR(Hitting) is undervaluing him.
Tony Mollica
7/21
How many All Star two way players are going to give us the problem of having WAR not adequately measure their value? I'd argue just Ohtani.
dodger300
7/23
I'd argue that WAR does not adequately measure the value of any major leaguer.

It's a clue, one of many, but that's all it is.