Fun with WAR

When I was researching Harrison Bader for my article this past weekend, I found myself on a pretty wild tangent, figuring out exactly how WAR was calculated by Fangraphs. What does that have to do with Harrison Bader’s defense, you might ask? That’s an excellent question! Not much! Moving on. Finding the formula hidden beneath the single number that comes up took a little digging, but not so much because it’s hidden. It’s just very fiddly. You have to account for a lot of different things in a player’s line, and then they aren’t done in WAR units or anything, they’re in their own random units. In any case, I got familiar enough with it to feel comfortable thinking about the different ways that a player can be average, above average, or replacement level.

The defining feature of the latter-day Cardinals has been a whole mess of absolutely average players. Depending on who you listen to, that’s either a bad or good way to build a team, and either frustrating to watch or fun to see the kids get good. I have one kind of specific issue with it, though. It’s boring! The kind of guys who are ‘average players’ are below-average to watch. Jon Jay? Bored already. Stephen Piscotty? Touching story, all the best to him, can’t get that excited about his game. Jedd Gyorko? Next! Lance Lynn? Well, he gave pretty good interviews, but you get the idea. Now, obviously one way to solve this problem would be to go way more stars-and-scrubs. Bryce Harper and Manny Machado are interesting. They do interesting things. Mike Trout is interesting despite being maybe the blandest person of all time because he baseballs interestingly. That’s a little beyond my purview, though. Instead, I’m going to mess around with WAR math and create some interesting average players. Does this have useful predictive or analytic value for this year’s Cardinals? I mean… no. Not even close. I just think the team would be a lot more fun this way. Here are some players who, by the purest technical definition, would be average players for the Cardinals while being incredibly more interesting to think and write about than some of the current guys.

Jose Martinez, Turned Up To Eleven

Jose Martinez is quite bad at defense and pretty good at offense. That’s neat. Players like that have existed before, though. Let’s make him really, really bad at defense, basically as bad as anyone has been, and then turn up the offense enough to make him average. The worst first baseman I can find on record (with reasonable innings minimums) is 2009 Adam Dunn. How awful was he? Well, he converted 65% of balls hit his way into outs. That’s worse than all but six first basemen to field 500 innings of the position in the last 20 years. He also had a worse fielding percentage than all but nine first basemen in the same sample- eight errors in 504 innings of first base is pretty unthinkable. He improbably piled on some throwing errors, which is impressive considering that he wasn’t really getting to that many balls in the first place. This was part of a season in which Dunn put up the worst defensive performance on record. Let me just basically say that this guy was one of the worst defenders ever. Okay, we’re going to make Jose Martinez’s true talent level that of 2009 Adam Dunn. A league-average hitter who fielded like 2009 Adam Dunn would be a -2 WAR player, challenging for the worst this century (2002 Neifi Perez with -2.9, if you’re curious). That gets us to the baserunning. Bad defenders tend to be bad baserunners, and Jose has been no exception for his career. In this game of extremes, that’s going to cost him- after I gave him roughly the tenth-worst baserunning season of the century, he’s down to -3 WAR.

That’s pretty clearly far from an average player overall. Let’s start adding offense until we get to average. His career .306/.374/.487 slash line barely makes a dent- he’s still a -1 WAR player. Let’s turn it up a notch, and make him a match for 2018 Matt Carpenter at the plate. Now his .277/.389/.591 line makes him the fifth-best hitter in all of baseball. He’s now an MVP candidate on offense, and worth a whopping 1 WAR. Somehow even hitting better than anyone else in the National League isn’t enough to make Extreme Jose average. No, to get him to league average, he’d have to run a 168 wRC+, which would look (after adjusting for park factors) something like Joey Votto’s 2016, .324/.446/.550 with a 17.7% strikeout rate vs. 16% walk rate. That iteration of Votto was worth more than five wins despite being a bad defender and baserunner. This Jose Martinez is worth 1.8 wins. He’d be the most maddening player on the field for both teams. You think this year’s Martinez is frustrating to watch on defense? This version is like that, only missing an extra ball every series or so. Over the course of the year, he’ll turn 36 MORE groundouts into singles than current Jose, already the most aggravating defender in recent memory. He’ll make up for it at the plate, though, walking and slugging his way to the best offensive season by a Cardinal since 2009 Albert Pujols on his way to a dead-average season. I’d watch that player- sometimes with hatred, sometimes with adulation. Just a quick side note- this hypothetical butcher form of Martinez would improve by over two wins if the Cardinals just put him out of his misery and traded him to a team that could DH him full-time. Yikes!

Harrison Bader, But Hitting Lefty

I hear you, past version of me. Harrison Bader’s defense is likely to regress. Well, honestly, I don’t care. For our purposes here, it’s staying. He’s going to be the best centerfielder in the league, starting today, effective forever. He’s conveniently already playing like that, so we don’t even have to adjust his numbers. Just imagine the current version of Harrison Bader playing centerfield for 150 games a year with reckless abandon. He’s also going to be pinch-running and is just going to be overall a great baserunner. How good, you ask? Well, 2008 Willy Taveras good. Taveras batted a putrid .251/.308/.296 and still somehow stole 68 bases while only getting caught seven times. It will go down as one of the greatest baserunning seasons of all time- or it would, if not for that slash line being so bad that he only played in 130 games and ended up with a negative WAR. Okay, so Bader is the best defensive centerfielder and best baserunner of our generation. What’s the catch? Well, he’s batting lefty. Why? Who knows. Maybe the crazy hypotheticals Effectively Wild listeners are always proposing have finally started to affect the real world.

Either way, let’s start destroying value. The first place I went was Cardinals-era Peter Bourjos. In part-time play in 2015, Bourjos hit .200/.290/.333, an absolutely abysmal batting line that would nonetheless make ultra-Bader a 3.5-win player (think 2017 Kyle Seager or Alex Bregman). No, we’ll have to do at least a little bit worse to get to an average player with such unrealistically good secondary skills. What if he hit like fellow Cardinal Dexter Fowler’s soul-crushing 2018? The victim of either a witch’s hex, a really bad run of luck, or Father Time, Dex is hitting a miserly .180/.278/.298 this year. That should tank Bader, and tank him good. I’m almost afraid to say this, but Dexter Bader (Harrison Fowler? Dexison Badler?) would still be worth 2.7 wins, and every Cardinals fan would riot about giving at-bats to this offensive black hole who would have been the 69th-best position player in the majors last year by WAR. If you’ve looked at Dexter Fowler, there’s only one place to go next- Chris Davis. Chris Davis is hitting an unthinkable .158/.242/.299 on the year, OPS’ing comfortably less than Matt Carpenter’s slugging percentage in a hitter’s park. After a little adjustment for park effects, Bader could put that line up for the Cardinals and be… exactly average. So, there you have it. Chris Davis could be an average player this year! All he has to do is become the best baserunner and centerfielder in the majors.

Peculiarly Platooned Kolten Wong

We’re veering away from the realm of possible players here, but I can’t help it- I’m having too much fun. How would Kolten Wong do if his platoon splits became more stringent in a truly ridiculous way? What I propose here is that Wong is arbitrarily limited to 300 PA a year. He can pick who they’re against- it’s going to be bad righties, basically. The downside is, someone else needs to cover the other 300 PA that would get him to a full 600-PA season. Who’s covering those at-bats? Well, a completely replacement-level player, that’s who. Patrick Wisdom? Someone like that. Someone who’s worth actually 0 WAR in those games. With clever platooning, the Cardinals can still get Wong something like 100 games worth of innings in the field (he’s going to be a defensive replacement a lot), so he’s at least got that going for him. In the interest of not repeating the Bader counterfactual, we’ll impose at least a little reality on Wong here- he can defend a little better than his career rate due to better rest, but he’s not going to be as otherworldly as he’s been this year. Let’s call it 10 UZR/150. He’s also an okay baserunner, let’s give him 2 runs a year worth of baserunning. With that all out of the way, I plug it into the formula and… Wong needs to be a 97 wRC+ hitter to rack up his league average 1.85 wins of total value. That only requires a modest line- say, .253/.328/.392. That doesn’t sound like much, but that plus Wong’s defense and baserunning would get him up to average in only 300 PA. Hey, wait a second! That’s just Kolten Wong’s career line against righties! It’s almost like Wong was a plant in this scenario to show that we have a player who is average in a really interesting way. The nerve of some writers!

Greg Holland AND Bud Norris

You didn’t think we were getting through this without a Greg Holland hypothetical, did you? Imagine this frankly nightmarish pitcher. For the first eighty innings he pitches, this guy is literally 2018 Greg Holland. I’m going to use RA9-WAR here because that’s going to get us to a more ridiculous number, and I didn’t come here to not be ridiculous. So that I can get him a reasonable amount of innings, he’ll be a starter, and he’ll go six innings every game. Want a neat fact about Greg Holland? Well, he had a 7.24 ERA with the Cardinals this year- and 21% of the runs that scored against him were unearned. His RA/9 was above ten! After those eighty innings, the pitcher is going to become 2018 Bud Norris. Then, he can pitch as often as he likes until he reaches league average. How many innings of Bud Norris would you need to combine with eighty innings of Greg Holland to result in a league-average pitcher? It’s an experiment I’d dread seeing in real life.

It’s also an experiment I’m happy to conduct in writing. Greg Holland’s WAR over 80 innings would be unthinkably bad- minus three wins, to be precise. There’s a hidden amazing part of pitcher WAR- good pitchers cause there to be meaningfully fewer runs per game, so each run they save is worth more. The opposite is true for bad pitchers- what I’m saying is, Greg Holland was actually SO BAD that the runs he allowed hurt the Cardinals less, and he was still awful enough he’d be worth -3 wins in eighty innings. Okay, so, nightmare time over. It’s time for a Bud. Norris has allowed a wonderful 3.23 runs per nine innings, comfortably better than league average. As even better news, he causes fewer runs to be scored in his games, so his run prevention is worth more than Holland’s… I don’t know, lack of run prevention? Even so, it takes Norris 110 innings just to undo Holland’s damage. To get to the amount of WAR a league average pitcher would accrue in 180 total innings, the Holland/Norris hybrid would have to throw 254 innings. Of course, now they’re behind the amount of WAR an average pitcher would accrue in 254 innings. This is a cruel game- it takes the hybrid quite a while to catch up. The breakeven is at 297 innings pitched- 80 by Mr. Holland, and 217 by Dr. Norris. That leads to an interesting question- would you roster this player? He’s throwing 300 innings in six inning chunks, so he’s basically throwing every third day. You’d also have to live through 80 innings of watching Greg Holland pitch. In exchange, though, the rest of the year you’re getting six innings of a closer every third day. If nothing else, it makes you think.

Conclusion

Well, I feel comfortable in saying that no one learned anything too useful here today. Still, it’s fun to imagine some of these exaggerated players. I’d love to see a team with both lefty Bader and Darkest Universe Jose. I’d be a little more tolerant of watching Greg Holland pitch if I knew he was about to transform into Voltron. I’d certainly be more willing to play a platoon with Wong if… okay, look, again, that’s just actual Kolten Wong. I didn’t alter his stats pretty much at all. In any case, enjoy the nonsense. The Cards are tied 1-1 with the Nats as I speak. Here’s hoping we see both Norris and Holland in THIS game.