Look, I get it. A revolution that doesn’t happen in a hurry, that doesn’t come from its participants’ hearts in an uncontrollable rush, is not much of a revolution. It takes passion to create change, and passion’s nearly inseparable partner is impatience.
But I want to tap the brakes for a moment. Before I join the thronging crowds who are striving to overthrow pitcher wins and RBI, I want to know what happens right after we win. History suggests that deposing even a despicable despot is dangerous and ill-advised, unless and until a clear and credible successor can be identified. (Just look at what happened after Fay Vincent’s ouster.) Whatever their faults and failings, wins and RBI serve certain purposes, and we should pinpoint them and find an alternative that will do what they’re meant to do, fill their void, before tossing them out with the trash.
First, the win. What is the win really meant to convey? The conversation over this issue has devolved, as Internet conversations are wont to do, and has become about whatever intangibles might drive a win skill (i.e., pitching to the score), or whether a good outing is wasted if the team doesn’t win.
At its core, though, the win is meant to tell us something I think we should all want to know: In how many of this guy’s outings did he pitch well?
Because that’s an important thing to know about a pitcher, right? We would never and should never try to count up the good and bad games a position player has, because there are too many, and because that’s not the shape of the value a position player provides. A starting pitcher, though, gets 30-plus chances per season to have an outsized impact on the outcome of a game. We should try to identify guys who pitch pretty well consistently, and to weigh their value against the value of guys who alternate very good and very bad starts.
The win, of course, stinks at this. That’s okay. It’s an artifact of another time. It’s a sundial, and we’re all mad because the ancients who built it didn’t build a table-sized Citizen Eco•Drive instead. In the early 1900s, baseball was primitive. No one knew whether strikeout or walk rates mattered. No one worried about run support or ballpark factors, and certainly, no one spent time accounting for bullpen support. The win rule predates the term “bullpen” by decades.
That’s what the win was about, though. It was meant to measure single-game performance from the single person most in control of the game each day. So I propose pushing for wider acceptance, and a more prominent role, for Bill James’ Game Scores.
Game Scores start at 50, and the starter accrues points through a system of debits and credits that gives us a non-binary measurement system for their overall contributions. The points are not perfectly in line with the formulaic value of various outcomes; this is still an approximation. As a concept, though, it’s delightfully simple. It’s easy to read: anything under 50 is not very good. Anything under 35 is a very poor outing. On the other hand, anything north of 60 is a big step toward a win, and any start over 70 reflects dominance.
We can take the average of a pitcher’s Game Scores for the season. We can measure the average variance from one game to the next. We can identify consistent and inconsistent starters, instead of merely good and bad ones. Best of all, we’re focused in on the pitcher’s own performance, not crediting him (or not) for the efforts of his teammates.
It’s not perfect. It requires rewiring one’s brain and seeing what wins are really about, and that’s a tall order for many casual fans. It is, however, a nice step toward a fuller understanding of what starters do, and an accessible stat for those who disdain abstractions like FIP and WAR.
Just as we did with wins, as we seek to push out RBI, we must begin by answering the question: What is it that RBI does best?
What they do best, really, is measure fan affection for players. Because that’s what attaches fans to players, right? Popularity is rarely a function of personality in baseball, at least for players on good teams. It’s not usually about their hustle or their style or even their wide-angle value, in truth. It’s about positive moments. It’s about how often that player brings you to your feet when you go to the ballpark, and how often he has you pumping your fist or high-fiveing your son as you listen on the radio. Usually, those moments come when runs score.
That’s not what the stat is built to measure, of course. At least, that’s not what it measures directly. But that’s what the stat can really tell you. Joey Votto is miles better than Brandon Phillips, but is it reasonable to expect Reds fans to like Votto better, in the heat of another pennant race, when it’s Phillips whose hits are bringing them to their feet most often? I’m not sure it is. I’m not sure it makes one stupid or small-minded to get most excited when one’s team actually scores, and to most like the guy who drives the most of that scoring. I think that’s a perfectly acceptable mindset for a fan.
We can’t let this drive player evaluations, and we can’t simply paper over the nuances, the small things that lead to the big moments, because they’re harder to react to viscerally. We should, however, strive to retain a stat that rewards players who provide those moments that excite fans and turn games.
I’ll advocate Win Probability Added (WPA) here, but I must confess that I’m not thrilled with it.
For one thing, WPA is an abstraction. Based on the inning, score, base-out situation and (in some cases) other stuff, WPA can tell you by what percentage every plate appearance, every outcome, changes the likelihood of winning a game. It doesn’t achieve that with a crystal ball, though, or by examining all parallel universes to see how the game turned out there. It’s based on models, and while the models are impeccably accurate and I wholeheartedly trust them, models are hard to sell to many, if not most, fans.
For another thing, you can only see WPA after the fact. It would be very hard to tabulate and express WPA in real-time from the park. (It would and does translate fine to broadcasts, which is to the good.)
Worst of all, WPA comes out as decimals. A walk with a runner on first in the fourth inning of a tie game might be worth .031 wins, for instance. I think it’s hard for most fans to root for tenths and hundredths of wins added, even once they embrace the frameworks that make them aware of them. I love those marginal measurements. Baseball is great because it’s so much a game of thin margins and incremental advantages.
But discrete events and round numbers make for much nicer stories. Fans can enjoy them more readily. I’m all for precision, and certainly for helping people see the value in the lead-up to those standing-ovation moments, but there’s no question that WPA is a tough sell.
Those are my recommendations. But the conversation should continue. The Occupy movement died when those urging change realized they had no idea what their end-game was. It’s important to move on from the blunt instruments of Wins and RBI, but to leave them behind, we need to have a destination at the ready.Next post: The Playoff Races, and My Rule for When They’re Over
Previous post: The Rise and Fall of the Gentlemen’s Agreement on Waivers in Baseball, and the Modern August Trade Market