You are here
Home > MLB > Subjectivity in Baseball

Subjectivity in Baseball

It’s been a long time since I’ve written something, but better late than never, right? Maybe not. Anyway. Recently I received, and accepted an internship at Inside Edge which I am obviously over the moon about. It’s a company still based around data, however it’s data charted by humans, as opposed to the ever popular Statcast and PITCHf/x, which are charted by computers. I’ve been thinking about this for quite some time, and even hinted at it here. Baseball has become very stat-heavy in recent years, and yes, overall it’s been a very strong positive. However, it has created massive blind-spots for teams and baseball offices everywhere. Whereas advanced statistics used to be the market inefficiency, they may now be a different inefficiency in the same market.

Now, this isn’t to say statistics are overvalued. Statistics, algorithms, and code led Alan Turing to lead the allies to victory in World War II. These were all the same relative components behind Moneyball, and every championship team in the last decade. Teams caught on relatively quickly to this concept, and it shows. There are very few teams now, and very few minds in baseball who don’t think similarly. Or at least have similar values for players. You can even tell by the standings. The parity in baseball is unrivaled by any other professional sport. So now we’re at the point where teams have all this information, but different ways of valuing players. They have their own “WAR” so to speak, and projections, but they’re all likely quite similar, with differences in organizational values and weights. We even see this reflected in the market; players like Mark Trumbo, Jose Bautista, and even Edwin Encarnacion have all had trouble finding (or in Trumbo’s case, currently finding) a landing spot this offseason. Teams aren’t finding value in free agency, and as an organization, teams are generally trying to create as much value as possible, whether short-term or long-term. In a sense lately, every trade and free agency signing looks to be a bit of a stand-off. This is a sign of a nearly perfectly competitive market. Is it good for the sport? Absolutely. Is it good for the players? Sometimes. But what about the teams?

While teams keep hammering away at statistics trying to find the next market inefficiency, there seems to be something missing. Take for example, relief pitchers. Often described as starters who weren’t good enough to be starters, they were kind of just the 18-25 spots on the roster for many years. And then teams started to invest in “proven” closers, but didn’t find as much success as they thought they would. Relief pitchers are volatile, so why pay for a proven one? He’s just proven he hasn’t been volatile for those three or four years before his free agency/trade. Remember Brian Wilson? Chad Cordero? Jonathan Papelbon? All of a sudden, a bunch of middling teams were signing closers to record contracts because they thought the problem was the ninth inning. The other relievers were still referred to as more of an afterthought, by both traditional minds of baseball, and analytical minds of baseball. You think the analytical minds of baseball were the ones to discover the value of relievers? WAR is still incredibly broken in regards to reliever value. However, they were one of the first communities to suggest the change of the usage of relievers. Then the small-market Royals made it to the World Series on the backs of a generally poor starting rotation, but an exceptional bullpen. And then reliever prices went up, but didn’t quite explode. Everyone still pointed to speed and defense, combined with great contact as their rise to success. Which was certainly true, and was certainly prevalent. But then the Royals won the World Series, and reliever prices went up again, and usage began to change, especially in the 2016 playoffs. Now reliever prices have skyrocketed, and even at the deadline, the return for Andrew Miller was quite large. Aroldis Chapman, Kenley Jansen, and Mark Melancon made out like bandits this offseason.

I’d like to go back to Kansas City. The Royals, for a long time, were generally regarded as anti-analytics and carried a very bad reputation for a long time after the James Shields trade. But it had to take someone, who knew the game of baseball, to sit down for a second and think “there are no rules about when, and how you can use relievers”. This wasn’t, in any way, based on statistics. Was it based on analysis? Yes. But it was also based on common sense, and just knowing baseball. They were able to add value through speed, something sabermetricians have generally been against, and contact, which they’re not against but which they don’t really value. After 2014, the team projections for the Royals ranged anywhere from 79-84 wins. They won 95 games, and then eventually the World Series. The Orioles have beaten projections for years, and are still said to be “playing above their talent level”, which is just a lazy way of saying “we’re wrong”. And the Rays are consistently projected to be a .500 team, and the 2017 projections for the Angels also place them at about 84 wins. Does anyone really believe that?

Right now, teams are hard at work trying to find the next Andrew Miller based on the Indians’ success. But they won’t find another Andrew Miller. There is one Andrew Miller. Teams should look to change the usage of relievers, not necessarily changing the relievers themselves. Andrew Miller is a weapon because you can use him at any point in high-leverage situations. Which, once again, for some reason, both communities missed this for a long time, and viewed each inning as a static event. There wasn’t any difference between an 8-0 lead in the first, and a 4-4 game in the seventh. Thinking is starting to change regarding relievers and their usage, but this it’s not a secret, or an advantage anymore if everyone is doing it. Now it’s just expected.

The other thing largely accepted is the valuation of defense. I linked my older article up near the top, but it still said the same thing I’ll be saying now; WAR is based on relativity. Meaning, if you put a good center fielder in right field, his value explodes because of his skill relative to other outfielders. Is it valuable? Yes, defense is valuable. Is it three full wins valuable? I’m going to say not likely. I wrote about Jason Heyward and how I thought he was past his prime due to his debut age, and how his WAR figure was artificially inflated due to his defensive numbers. Sure enough, the entirety of last year happened, and the contract now looks bloated. Not to say WAR isn’t useful; it’s certainly the best stat we have to estimate value, and is generally a very useful stat. But is Jason Heyward really a 6+ WAR player? Remember, even when he was coming up, he was a well-rounded prospect and regarded as the best in the game. Looking back at some years old FanGraphs chats a few months ago, I remember one writer saying he had 40+ home run potential. If the WAR values of the player are equal, but you get to choose whether you take 40+ home runs or right field defense, I’d probably take the 40+ home runs. He redistributed his value, but I certainly don’t think he’s the same superstar player sabermetricians made him out to be. It’s not to say Jason Heyward isn’t a great player, he is (or was), and it’s not to say defense isn’t valuable (it is), but we also forget defensive stats are still very, very floaty. Adeiny Hechavarria is another example of this; a terrific defender by the eye test and scouts, for some reason, UZR and DRS do not like him. While, on the other side, Jhonny Peralta was strangely loved by UZR and DRS for a few years. It’s just not very reliable. Even this offseason, for example. Mark Trumbo is actually a good defensive first baseman, but it’s not reflected in his WAR calculation. Yes, I understand the penalty associated with the position of first base in WAR calculations, and he’s not worth $20 million a year or whatever ridiculous price he wanted, but for $10-12 million a year, I don’t think he’d be a complete disaster.

Baseball is a $10 billion industry, with a lot of money (clearly), to throw around. The point of money in baseball is to, undoubtedly, attract talent, like any industry. But who’s to say the talent has to be limited to players? Theo Epstein’s contract was a bit of a surprise, but why? After three World Series winning teams, each staunchly different from the last, he’s a proven commodity. Or somebody like Ray Searage on the Pirates. Given the surplus value he’s created for the Pirates, he’s likely worth a lot more than he actually is. It’s a $10 billion industry with talent bursting at the seams of the market, but nobody is noticing. Is there really talent out there? Yes. It’s just that the teams smart enough to invest in scouting and development have accrued a lot of that talent. Like the Cardinals or Giants, for example. They develop both at the major league level, and minor league level.  They’re able to extrapolate and create value from nothing. Who was Aledmys Diaz before last season? Who was Jedd Gyorko? Why’d they pay so much for Jhonny Peralta? Because they have something other teams don’t. The pitchers they always choose are terrific athletes with multi-sport backgrounds, another completely subjective component of their process. The Yankees also have terrific development and coaching. This team looked like potential hot garbage after the trade deadline, and still squeaked out 84 wins. And then in the same division, we see the horrid test tube concoction of the Rays, who had a bevvy of interesting players that projections liked, but many of them had strange behavioral problems and there were enormous clubhouse issues. Of course this is going to weigh on a team. I’d like to stop pretending like there’s no such thing as a “bad teammate” or that players at this level are unemotional robots. Zack Greinke almost quit baseball due to anxiety. Yangervis Solarte’s wife died, Jean Segura’s child died. The mental makeup of a player is enormous. It’s one of the things scouts consider heavily about prospects, and is completely subjective. The Chicago Cubs for example, all seemed like such a great group of guys that enjoyed playing the game. Same with the Indians. They just looked like they were having fun. And obviously winning is fun, but this was for the entire season. Having a team full of good guys, and a manager that can reach them, keep them in check, and keep them engaged for 162 games has value. Just because we can’t assign it a figure doesn’t mean the value isn’t there. I’m still a big supporter of Buck Showalter. Terry Francona and Joe Maddon seem to be terrific at reaching their players. Even guys like Bud Black and Dusty Baker often get good reputations from players. The idea that this is all “static noise”, is wrong. In the meantime, while we can’t put a definitive number on that, teams will blindly be investing in better coaching and development, but probably seeing noticeable results. Essentially, what it comes down to is I’m not sure there’s a definite advantage to be had in stats right now. I think a lot of the new advantages are going to be realized in in-game strategy, scouting, and development. It’s a lot cheaper, and creates a lot of the surplus value in players. Instead of trying to buy value in free agency, which seems like a backwards concept, pay less to develop the talent you have in-house. That’s actual value; you don’t pay for value, because at that point it’s not value. Mental health should be focused on, or mechanics in pitchers to prevent injury, or to create spin. Managers who connect with players, and coaches who weren’t just great hitters/pitchers or organizational favorites, but real minds of the game should be hired. Especially the ones that can combine the right stats and the right eye for baseball. As someone who is a Management Information Systems major with a Data Analysis minor, and who got into baseball because of stats, and continues to follow it because of stats; keep it simple, stupid.