World Famous StatSpeak Roundtable: June 16
June 16, 2008 9 Comments
The roundtable rolls on (thought of that one myself) this week to the doorstep of Dave Studeman, one of the fine folks over at The Hardball Times. You’ve probably read Dave’s column, Ten Things I Didn’t Know Last Week, and if you haven’t, you are a bad human being. Dave also runs the Baseball Graphs website. Today, Dave joins Eric and Pizza to discuss the long-suffering Cubs, WPA, and which GM should be spending his Monday morning updating his resume.
Question #1: Is WPA a useful tool for ranking players? In what instances? If not, can it be improved?
Dave Studeman: People who read my work at the Hardball Times know that I’m a fan of Win Probability Added. I think its inherent logic is deeply compelling. It doesn’t measure “true talent” or even “production.” But as a measure of value, it’s hard to beat. As a quantification of the ups and downs of a game, it’s the best. WPA is truly unique and endlessly fascinating.
However, there are a few things that could be done to improve WPA:
- Measure WPA against a replacement level instead of average. This is how you factor playing time into the measure, and I think it needs to be added to the discussion. Do it right, too: make the replacement level differ between starting pitchers and relievers.
- Find a way to add fielding to the mix. Eric proposed “visual WPA” a month or two ago, and I’d love to see something like this occur. In fact, there was a movement afoot to create such visual WPA a few years ago, but it never congealed into something useful.
- For those who consider WPA a “junk stat,” pursue ways to combine WPA and other win-based systems, like Win Shares, to overcome their objections. In particular, I hope to create a system in which the wins and losses of individual players adds up to the teams’ wins and losses — but that doesn’t have the “late inning emphasis” that WPA has.
Eric Seidman: I am a huge proponent of win probability added and the like, including WPA/LI, LI, and Clutch score. They each offer something different with regards to the win contribution of the player(s) in question and something I really enjoy doing is looking at the three side by side–a WPA slash line of sorts–so we can compare a) overall contribution, b) contribution in a context-neutral setting, and c) whether or not said player raises his game in crucial situations. For instance, Pat Burrell (at the time this answer was given) has a WPA slash of 4.22/2.80/1.24; he leads all of baseball with four and a quarter wins contributed, ranks third in contribution based solely on his performance and not the context in which that performance took place, and comes in fourth in terms of overall clutch score.
Put together, it seems that few, if any, have been more valuable to their team in terms of contributing wins than Burrell. Despite this, poll every somewhat sane fan on the planet, Phillies supporters or not, and Chase Utley is very likely to garner 95%+ of votes as far as who the Phillies MVP has been. Utley’s WPA slash is 2.06/2.53/-0.52; his performance has contributed less than half of Burrell’s but, while he has not been very clutch, his context-neutral win contribution is similar to Pat’s.
Utley has been a better player than Burrell this year, though, based on just about every other type of metric and, because of that, I would have to say that WPA is not a tremendous evaluative tool UNLESS we are strictly measuring win contribution on a constantly updating player projection, similar to how we should evaluate players on any statistic. I love using the stat to measure team-based win contributions as well, to see who has accounted for the highest percentages of their team’s victories. Someone with a 4.25 WPA on a team with 50 wins will have a lesser percentage than someone with a 4.25 WPA on a team with 35 wins.
Pizza Cutter: WPA has its uses as a rough measure of worth, but of course, WPA is affected by leverage, so WPA/LI makes more sense to me as a tool for ranking batters. This makes sure that players aren’t rewarded or punished for circumstances beyond their control (what’s the leverage at the given point that their turn to bat just happens to come around). Pitchers have more control over the leverage of a situation and are the only member of the team guaranteed to be involved in every at-bat in the inning, but even then, pitchers get everything that happens in an inning attributed to them, even the obvious fielding errors. Teasing apart who’s to blame (or credit) for the team in the field, whether the credit belongs to the fielder, pitcher, or simply the batter, is one of the great Sabermetric frontiers that people have only begun to venture into.
Question #2: Who will be the first GM fired this year?
Dave Studeman: I predict… Wayne Krivsky! That wasn’t hard.
What’s interesting here is that some of the most disappointing teams this year are led by some of the most respected GM’s in the game. Cleveland/Mark Shapiro. Detroit/Dave Dombrowski. San Diego/Kevin Towers. Milwaukee/Doug Melvin. Maybe you want to blame Omar Minaya for not developing young talent for the Mets (I know the media does). New York has not been aggressive in the draft, and they’re paying for it now. But it seems as though Willie Randolph will be scapegoated before Minaya is. That leaves Bill Bavasi and the Seattle Mariners. Well, they’ve stuck with him so far, despite some questionable judgments and seasons in the past, right? Why change horses now?
Eric Seidman: I’d like to say Bill Bavasi (to make Dave Cameron happy) or Brian Sabean (to make San Francisco happy), but I have a sneaking suspicion JP Ricciardi’s tenure in Toronto is close to ending. Granted, it’s pretty tough to compete in the AL East but they have arguably the best pitching rotation in the game and are still struggling. I wrote back here a few weeks ago how poorly his trades have worked out, via Win Shares, and how they’ve cost the Blue Jays plenty of wins over the years. The fact is that he has been there for a while and no real improvement in the team’s W-L has taken place. If the Mets fall too far out of the race in the coming weeks/month I could also see Minaya getting the axe; though it could just as easily be Willie Randolph to go first. Sticking to one answer, I’ll go Ricciardi, but I ultimately believe it will be whatever team out of NYM or Tor falters quicker and moreso than the other.
Pizza Cutter: If Hank Steinbrenner wants to make a statement, especially given that the Yankees are still floundering in third place (when did they get out of last place?… this happens every year), perhaps Brian Cashman might just be the best target? However, Hank needs Brian Cashman as a punching bag. So, those at Fire Bill Bavasi will probably get their wish, and deservedly so. On top of the bad trades (but then we’ve all made those) and the strange free agent signings that show little clue as to the basics of understanding baseball statistics (hi there, Mr. Beltre), there’s the possibility that Bavasi and the Mariners will pull off a 100 loss season with a $100 million payroll. Previous roundtable guest Dave Cameron from U.S.S. Mariner points out that the Mariners are already in the mood for a good purging, and perhaps it’s best this time to start at the top.
Question #3: The Cubs… is the curse finally over?
Dave Studeman: You know, I find this curse thing with the Cubs a little silly. I’ve lived in Chicago for 25 years, and I never heard of the Billy Goat Curse until a couple of years ago, when the Chicago media felt the Cubs needed something like the Curse of the Bambino. It’s just a trumped-up PR stunt.
If you want a real curse, go back to the very first year of organized league baseball and the curse of Mrs. O’Leary’s cow. I’m talking about 1871, the first year of operation of the National Association of Base Ball Players, the precursor to the National League. Chicago had a team in that inaugural year, and their season was thrown for a loop when, on 10/17/1871, with the Chicagoans clinging to a slight lead, the city burned down. Including their ballpark, their uniforms and their equipment. They had to play the remainder of their games on the road and they lost every single one. The A’s took the pennant.
It’s never been easy to win it all in Chicago.
Eric Seidman: This is tough because anything can happen in the playoffs and a “curse” signifies other-worldly forces preventing something from occurring regardless of how high the chances of said instance occurring may be. The Cubs seem to be the best team in the National League but it is still just 69 games in; not to say they could miss the playoffs by 15 games or anything along those lines but I don’t think we’ll really know how good this team is until at least the end of July, if not later. Kerry Wood hasn’t even landed on the DL yet for gods sake! If I was a betting man I would still put money on the Red Sox but it wouldn’t surprise me in the least if the curse ends this year; of course, it also wouldn’t surprise me if they get swept in the divisional series.
Pizza Cutter: Well, for another couple of weeks, the Cubbies will be my neighbors, so I suppose the politically expedient answer is yes. With that said, the Cubs are right now the best team in the National League. They managed to find some good complementary pieces in Fukudome and the surprising Geovanny Soto. Then the Cubs managed to find a lot of hitters who fall into the “Yeah, I guess he is pretty good” category. Now, the dark side: Marmol and Dempster have dangerously low and unsustainable BABIP’s, and you have to wonder how long a team can survive using Jim Edmonds in center field, but… I feel like it’s 2003 again. This Cubs team is good. It’s not “so amazing that I’m already printing up championship t-shirts” good. The Cubs will make the playoffs, but then there’s a good chance that they will be bumped out simply because eight teams make it and only one wins. Which means that Chicago will find another villain in the ballad of the curse.