Time for a Truce Between Traditionalism and Sabermetrics
Part of the beauty of baseball is that it lends itself to so much analysis and quantification. From the time Henry Chadwick invented the box score in the mid 1800’s, fans have been able to formulate hardline opinions of who exactly is the best player in the game, whether it be at his particular position or overall. And they’ve had concrete facts to back it up.
Other sports have statistics, naturally. Touchdowns, rebounds, shots on goal, wickets, and so on. But as Billy Chapel (Kevin Costner, who else?) says in For Love of the Game, “We count everything.” That’s part of makes baseball so terribly interesting. There’s a consistent basic setup for every single play in baseball; the pitcher, the catcher, the batter, and the seven other men in the field. “You throw the ball, you hit the ball, you catch the ball,” to borrow a quote from another Kevin Coster baseball film (Bull Durham, for those of you keeping score at home).
Yes, that’s pretty much all baseball is. Except that’s completely not the case whatsoever. Baseball is both simultaneously wonderfully simple and mind-numbingly complicated. It’s possible to reach that level of intricacy without even delving into the current ideological war that will be covered below, that of traditionalism and advanced metrics. Before even reaching that level, there’s the matter of pitch sequencing, pitching management, positioning of the defense, baserunning plays, fielding, bunting and more. The manager isn’t just there to fill out the lineup card and make calls to the bullpen, and the new instant replay capacities have given him yet another toy to play with. Baseball is nearly just as good as football for those who like to armchair quarterback.
All of that is part of the traditional side of the game, and it’s a very necessary part of the game as a whole. In recent years, though, a total fundamental conflict has arisen in the game, and that’s what I’m going to try to address today. I don’t claim to have the right answer, only one that makes sense to me.
I’m talking, of course, about sabermetrics. Everyone who’s more than a strictly casual fan of the game has an instant reaction when they see that word, and to some it’s one of the four-letter variety. There are two extreme schools of thought when it comes to breaking down the game. For the sake of convenience, we’ll call them the “old school” and naturally, the “new school.” “Old school” generally implies backwards thinking and intrinsic wrongness, but let’s try to cast that aside for now. We’re going to try to be levelheaded here.
A number of people on both sides of this issue believe that their approach is the only real way to look at baseball. Those of the eye test and the Chadwick box score say that the game can tell you everything you need to know by simply watching it, and looking at a few cursory numbers. Those of the sabermetrics crowds say that you can’t possibly learn anything from just watching, and that trends and linear weights provide the road to enlightenment.
There are arguments to be had with both sides. The most accurate way of getting a true image of baseball, I say, is to supplement one with the other. This may be glaringly obvious to some, and I’ll admit I’m rather late to the party on this one. Oh well, better late than never.
For a while, I was staunchly of the new school. I still am, to rather large extent (I’m attending a conference about sabermetrics). That’s because information is just that, information. There’s not a single thing about sabermetrics that disproves them in any way, shape, or form. The formulas for each stat aren’t an arbitrary assumption of what may make sense. They’ve been developed by individuals with backgrounds in economics and mathematics. They’ve been tested and weighted to give largely accurate assessments of performance (or expected performance).
One of the main arguments against the proliferation of sabermetrics is that they detract from a relaxing enjoyment of what the game for what it is; a game. That’s a perfectly reasonable argument. As someone who is a huge proponent of advanced metrics, I don’t find myself thinking about them a whole lot when watching games. That’s especially true when I’m actually at the ballpark. My focus is on the game in front of me. The only non-traditional stat that really prods its way into my mind at the ballpark is BABIP, which as some of my readers may know by now stands for “Batting Average on Balls In Play.” If a defender makes a great play on a ball that would have normally gone for a hit, I like to say they’ve been “BABIP’d.” The BABIP gods giveth, and the BABIP gods taketh away (we occasionally call those instances “errors,” which I’ll address later). But that’s largely just part of my own personal oddities, anyway.
There’s also backlash against game broadcasts incorporating some advanced metrics into their shows. The traditional numbers have been just fine for more than a hundred years now; the argument goes. Why suddenly fix what isn’t broken? Well, that’s because the traditional numbers really don’t work all that well. The most glaring culprit is every traditionalist’s best friend, RBIs. Earlier claims of levelheadedness may go out the window here.
The aforementioned Henry Chadwick created RBIs, and some other popular numbers like batting average and errors. Chadwick certainly had good intentions when he set himself about trying to make sense of the game of baseball. Part of that was for his own benefit, as he was an Englishman who was much more familiar with cricket than baseball. Approaching the game from the perspective of an outsider, he decided that batting average and RBIs were the best measure of offense. It’s easy to see where he was coming from. Batting average is the measure of how often in his at bats a hitter does his job and records a hit. RBIs are how many runs the hitter has knocked in, and scoring runs is the main object of the game. Henry Chadwick had done his job, no?
Wrong. Henry Chadwick had only muddied our understanding of the game. Unless an at bat results in a solo home run, a hitter cannot record RBIs without there being men on base in front of him. That means if Miguel Cabrera hits a grand slam, he is credited 4 RBIs (the three men on base, and himself, score). However, Cabrera can do the same exact thing with his bat while the bases are empty. Why should Cabrera get more credit for the grand slam? He didn’t put the runners there. His RBI total is largely more reflective of the lineup around him than his own ability. Cabrera is of course a ridiculously good hitter, but if you put him in the middle of, say, the pre-All Star break Padres lineup, he wouldn’t have nearly as many RBIs. He also would have fewer home runs in all likelihood. That’s partially because Petco Park is a cavern where home run hitters go to die. Laser shots into the seats become routine fly balls on the warning track. It’s also because of the less threatening lineup, and Cabrera could be more easily pitched around.
See how RBIs are so situational? How can they possibly be counted on as a reliable measure of a hitter’s acumen? They simply can’t be, that’s how. Batting average is also very screwy. There’s two reasons for that. First and foremost, not all batting averages are created equal. You can win a batting title by .330. It’s a very impressive number in today’s game. You can also hit .330 by hitting nothing but singles. A great example is last year model year of Braves third baseman Chris Johnson. For a good part of the season, Johnson was the leader of the pack in the hunt for the NL batting title. He also did it by hitting largely singles. On the whole, Johnson managed to be not terribly valuable (2.7 fWAR) while still contending for a batting title.
The fact of the matter is, Miguel Cabrera gets as much credit in his batting average for the grand slam he hit earlier as he does for a little bloop single over the second baseman’s head. If we take away years of bias from ancient baseball dogma, in what world does that make sense if this statistic is to be the go-to measure of a hitter’s performance?
The other problem with batting average is that it doesn’t draw from the hitter’s total performance. See, batting average is measured from at bats, not plate appearances. What exactly does that mean? Well, walks (intentional, unintentional, and hit by pitch) don’t count as an at bat in the old Chadwick box score. This is because Chadwick originally counted a base on balls as an error accredited to the pitcher. That eventually was changed, but walks were never converted to at bats. We’re all familiar with reading that (for efficiency’s sake, we’ll continue to use Cabrera as our human guinea pig) Miggy went “2-4 with a walk.” But he didn’t have four plate appearances, he had five, and in that fifth plate appearance he got on base. He deserves credit for that along with his hits. While a single may be more immediately useful than a walk (you can’t give a runner a chance to go first-to-third with a walk), there’s still one more runner on base for the next batter to potentially knock in.
In a nutshell, batting average is incredibly deceptive. One of my favorite names to drop in my writing is Adam Dunn, and the Big Donkey’s going to get another mention here because frankly, he’s fun to talk about and is the poster boy for a certain type of not-so-valuable hitter. Adam Dunn is your stereotypical all-or-nothing kind of guy. He swings big and misses big, and when he finally does make contact it’s got a pretty good chance of going into the seats. The side effect of that is that Dunn will usually put up a batting average in the low 200’s, and strikeout totals sometimes just below that range as well. Chris Carter of the Astros (until his recent ridiculous hot streak, that is) and Mark Reynolds of the Brewers are also guys with this approach.
Dunn, however, has a secret weapon. What puts him above guys like Carter is the fact that he has a very prolific talent for walking. This year Dunn is only hitting .228 (actually a pretty good mark for him), but he owns a .351 OBP. There’s a couple reasons for this. First is that pitchers know that if they give Dunn something high in the zone, he could make them regret it in a hurry. So, they pitch on the edges and down in the zone, which means that some of those pitches are going to miss or get called balls (see below, via BrooksBaseball).
Dunn typically also bats in the 4 or 5 spot in the Chicago lineup, and that gives pitchers less pause about pitching around him because, frankly, after a certain spot the White Sox lineup isn’t too scary. Jose Abreu has certainly made the lineup deeper by a country mile, but Dunn still usually doesn’t have a ton of protection outside of Dayan Viciedo or Alexei Ramirez, as Avisail Garcia was hurt until recently. No harm in putting Dunn on base, then.
The other factor is that despite Dunn’s tendency to put up huge strikeout numbers, he’s very good about laying off pitches he knows he can’t make good contact on. And for Dunn, he knows he wouldn’t be much of a producer if he really tried to hit for average. He gets his share of singles, but there’s a reason Dunn is 35th on the all-time home runs list. He does what he’s good at, and that’s launching dingers and taking walks. In fact, Dunn has a career 15.9% walk rate, despite his 28.5% K rate. Dunn’s very much a three true outcomes (walk, strikeout, home run) kind of guy, which means that’s not very valuable. However, his on-base skills make for a fun little case study in looking at why batting average misses a lot of stuff. For a version of a player that derives part of his actually substantial value from mid-level average and high on-base (plus some Gold Glove defense), meet Jason Heyward.
Either way, what Henry Chadwick failed to realize, and thus the greater baseball world for many years as well, is that plate discipline and taking walks is a skill. Part of is directly related to pitchers wanting to stay away from batters like Dunn, but those batters are stayed away from for a reason. However, pitch recognition is a huge part of a batter’s arsenal. There’s endless talk about how little time a batter has to decide whether to swing or take, and that’s all true. It’s a verifiable skill to be good at picking pitches, and forcing pitchers to take risks with their offerings. Nobody in their right mind pounds the middle of the strike zone against Miguel Cabrera. You may be able to get away with that against the Brendan Ryan’s of the world, but not the scary middle-of-the-order names like Cabrera. You have to be crafty and chip away at the outside of the zone.
What separates men like Cabrera from, say, Adam Jones, is discipline and a good eye. Everyone knows how good Jones is. He’s the AL MVP this season if it wasn’t for the existence of Mike Trout and Felix Hernandez. But the one thing Jones doesn’t do is walk. For his career, Jones has a ridiculously small 4.4% walk rate. What’s even more surprising is that besides currently enjoying what’s clearly his best season yet, Jones is sporting the lowest walk rate of his career at 2.9%.
Plate discipline and picking pitches doesn’t sound much like sabermetrics, does it? Not on the surface, no. But sabermetrics looks to quantify stuff like that, and attempt to put all of one aspect of the game (like hitting) into one tidy little box. Ideally, we’d be assessing hitters with a catch-all stat that incorporates any time a guy reaches base, and gives more credit for a double than a single, and so on. This number still wouldn’t be perfect, but it would be progress. Well thank goodness someone came up with wOBA (weighted On Base Average), then, because that’s exactly what I’ve been talking about.
Fangraphs does a better job of explaining how wOBA is a “gateway” sabermetric stat than I ever possibly could, so give that link a read when you have the time. But basically, wOBA is what you’d come up with if were a statistician and you’d never heard of Chadwick’s numbers or seen a triple slash line (which I’ll get to in a second) if you wanted to measure a hitter’s production in one entry. It assigns value to walks, singles, doubles, triples and home runs, and tosses them into one number. It’s like on-base percentage on sterioids.
It’s numbers like wOBA that I think present an idea of where we should be as baseball analysts and fans looking to assess their favorite players. I can sympathize with fans that want to follow the game without taking a stats class and learning what a bunch of acronyms stand for. That’s a perfectly reasonable request, and it’s unrealistic to ask all fans to be sabermetric gurus. However, the depiction in popular media of sabermetrics as unnecessary rocket science is nothing if not ridiculous and willfully ignorant. No other sport but baseball fosters such a violent reaction to change and additional knowledge. Part of that in all likelihood is the fact that baseball has been around, seemingly, for eons. Sports like basketball and football don’t have legendary figures such as Ty Cobb and Babe Ruth from so long ago to point to and say, “They did great things without the luxury of your sabermetrics, so why can’t we? We can know everything there is to know about the game just by watching it!”
It’s completely true that Cobb and Ruth were remarkable players. The old numbers and the new support that. However, if we’re going to appreciate the game, why not do so correctly? The traditional triple slash (batting average/on-base percentage/slugging percentage) is a decent enough assessment, and one I’m actually rather fond of. It’s simple, concise, and above all familiar. Also, comparing average to on-base is a fun exercise that tells you how good a hitter is about taking his walks. For instance, in a pretty small sample size through his first few games, new Cubs phenom Javier Baez’s batting average was identical to his on-base because he hadn’t taken a single walk. That’s normally not good whatsoever, but again, small sample size. Baez was hitting .267, had an identical on-base, and a slugging percentage over .600 because he immediately started clobbering home runs as soon as he came up. It’s certainly possible to be valuable in that way, by hitting the snot out of the ball all the time without walking a whole lot. Just ask Adam Jones. But not as valuable as someone who also walks.
We didn’t know that not too long ago. Heck, even in recent years Dusty Baker discouraged his players from walking, claiming that in doing so they “clogged up the bases.” It’s part of why Joey Votto’s transformation into an OBP monster has drawn all kinds of ire. This sentiment is, of course, absurd. By placing himself on base much more often, in addition to hitting for average, Votto not only continues to knock in runs but gives the guys behind him (Brandon Phillips and Jay Bruce, for instance) opportunities to knock in more runs.
And yes, you’re correct in that that sounds like RBI talk. It does, because as I said earlier, the ultimate goal of the game is to score more runs than the other team. There’s no problem with that whatsoever, it’s using RBIs as a measure of a player’s worth that’s the problem. It’s using batting average as a measure of talent that’s the problem. It’s using errors, the only completely arbitrary statistic in sports, as a measure of fielding ability that’s the problem. Only in baseball can an official scorer decide whether or not a fielder “should have” made a play, or whether the ball was placed just wide enough that the hitter should get credit for a hit. Consider this: if a player cleanly fields every ball hit at him, he theoretically should be charged no errors. However, that fails to account for every ball that passes by his reach. It’s possible to an awful defensive player and still look like you’re amazing, because you don’t make errors. Vintage Derek Jeter is the classic example of this.
Jeter’s defensive woes throughout his career are no secret, frequently being mentioned everywhere from SportsCenter to the most obscure of sabermetric blogs. Mainstream sports media often also makes much talk of how batting averages are deceptive, and how RBIs are not an entirely accurate metric. In fact, almost none of what I’ve said is new information. Yet these same broadcasters and writers cannot stand even the mention of sabermetrics, and refuse to approach the game a different way.
The dogmatic traditionalism of baseball completely defies reason. In no other field (except for perhaps politics) is the influx of new information looked down upon. In baseball, it’s often not only looked down upon, but also derided. The media aren’t the only culprits either, as many front offices share the same sentiment and traditionalism. I don’t think I need to elaborate on where that’s gotten the Diamondbacks and Phillies, the Rockies and Padres. I don’t need to talk about the magic it’s worked for the Athletics and Pirates, the future that seems attainable for the Astros and Cubs, the ride the Rays went on and the ride the Boston farm system and courageous recent trades are about to produce.
It’s not all about numbers though, and I think that’s the thing that detractors need to realize, and sabermetric purists need to come to grip with. There’s still a place for scouting, there’s still a slot for “makeup” (personality and work eithic) on the scouting report. Even the Astros’ famous “Ground Control” sabermetric division still confers with traditional scouts, and their testimonies together help shape GM Jeff Lunhow’s decisions about the future of the rebuilding club.
And scary acronyms like wOBA aren’t just created for kicks, and neither are things like wRC+, xFIP, SIERA, WAR, RAA9, UZR, and so on. They’re all designed to look at what’s actually happening in a ballgame, not who’s displaying the most grit. Don’t want to deal with a million different numbers? Perfectly fine. And the big media outlets realize that, which is why things are streamlined and they’ve stuck with the numbers your parents taught you to read. But it’s time to start assessing the game correctly, and with all due apologies to Henry Chadwick, it’s time to put his system to bed. The Information Age is here, and while that doesn’t mean baseball has to stop being fun, we should actually know what the heck is actually going on out there on the diamond. Sabermetrics don’t seek to completely overhaul the game and make it something it’s not. They just try to seek out what we’ve been missing that could aid in out understanding and playing of baseball, and quantify the fundamentals that are so integral to success. Sabermetrics are, at their very core, a strange incarnation of the old school. The sooner we stop treating them like the boogey man, the sooner we can possibly make baseball itself even better.
For more on sports injuries, check out our friends at Sports Injury Alert.
Thank you for reading. Please take a moment to follow me on twitter @StelliniTweets, and our department, @LastWordOnMLB. Support LWOS by following us on Twitter – @LastWordOnSport – and “liking” our Facebook page.