New York Mets
What They Did: 77-85, 5th Place NL East.
Actual Runs: Scored 718 runs, Allowed 742.
Expected wins based on RS and RA: 78.6 (1.6 above actual)
Restated: Scored 715 runs, Allowed 741
Exp. wins based on restated RS and RA: 78.4 (1.4 above actual)
“Not bad for a quant but. . . C’mon pal, tell me something I don’t know. It’s my birthday. Surprise me. ”
Gordon Gekko, Wall Street
I get it. I get it. You can read a hundred 2012 previews a day on the web. You’re going to pick one to follow. Sure I told you the Tigers are going to struggle to win their division, the Rays are going to finish fourth, and the Royals are more likely bound for last place than second place – that’s solid work – but ultimately I’ve given you Boston, the Yankees, Detroit, the Angels, Texas, and Philadelphia as play-off teams so far. That’s the equivalent of giving you information, not getting you some. I can just hear you saying, “Tell me something I don’t know, Bud Fox.”
O.K, how about this: Not only are the Mets not going to finish in the NL East cellar, thanks to the expanded Wild Card round, I had them in the playoffs until an arbitrator repealed Milwaukee Brewer Ryan Braun’s 50-game suspension. As it is, I think the Mets will be in a dogfight for that last playoff spot. (With Mets fans waking up -- on the West Coast -- to the encouraging news surrounding the Madoff-related settlement, they can spend the day dreaming of better times ahead.)
This is so far out-of-consensus that I am certain I’ve already lost market share, in terms of readership, among the fans of the Atlanta Braves and Washington Nationals. Since the Mets are generally expected to win 72 games in 2012 and I’m projecting total wins to be more than ten games higher, I’ll go over the projection calculation in detail.
The Mets won 77 games last year, but I actually start them as a 78-win team thanks to very mild bad luck in converting their runs scored and runs allowed into wins (see box above.) The concept of a “replacement player” – the R in WAR – often comes under fire because some have trouble visualizing a standard replacement player. It’s true, for example, that 2011 mid-season centerfield call-ups Desmond Jennings and Brian Bogusevic for Tampa Bay and Houston respectively, were very different players with very different values. Jennings is a star in the making; Bogusevic is a Houston Astro. Tampa, therefore, had freely available talent (the typical definition of a replacement player) far superior to that of Houston’s. Critics of the WAR concept say there can be no “replacement player value” because it is not standardized.
If that logic troubles you when someone quotes Wins Above Replacement Player, or WAR, think instead of an entire team made up of nothing but players cast-off by other teams. A team, in other words, of replacement players. This would be a terrible squad; accounting for bench players not discarded by the 29 other MLB teams, this team would field, essentially, the 40th-best MLB player at each position. We can all agree, I believe, that the ineptitude of this team would be on par with the worst teams in MLB history. I set an all-replacement team at 45 wins and 117 losses. When you consider that only one team has won less than 50 games since (at least) 1970 – the 2003 Detroit Tigers at 43-119 – that seems about right. Using this baseline (and you could actually choose any win level, as you’ll see below) we don’t have to worry or argue about individual player WARs. We can simply calculate an entire team’s aggregate WAR.
The Mets won 78 games (using my restated wins, above) last year which means, therefore, collectively its players generated 33 WAR (78 minus 45). I break that WAR into three baskets as follows (note that the sum of the MLB averages, plus the baseline 45 wins equals 81 wins – an average MLB team with a .500 record):
2011 WAR Contribution MLB Average
Starting Pitchers 7.2 10.4
Relief Pitchers 3.1 6.7
All Hitters 22.6 18.9
Starting here and accounting for changes we can see where the 2012 Mets project to finish. It should come as no surprise that the Mets had such a below-average performance from its pitching staff last year. Opponents scored 742 runs on the Mets, fourth-most in the National League. The Mets’ bullpen posted the second worst runs against rate in the NL and given that it was just about impossible for any team to be worse than Houston in any aspect of play in 2011, a second-to-last performance is pretty bad. If you’re a Mets fan who suffered through the bullpen carnage last season, it might seem counterintuitive, or even ridiculous to assume significant improvement in 2012, but the fact is that the best predictor of any single team’s bullpen performance in year T, is the league average in Year T-1, regardless of how good or how bad that team’s bullpen was the year before. The reasons seem to be two-fold. Bullpen performance is, by definition, a small sample size, high-leverage endeavor and therefore highly volatile. Many relievers typically throw 40 innings per year, and often pitch with runners on base, making their year-to-year results highly variable, even if their underlying skill sets (strikeout, walk, and groundball rates) are stable. If a reliever with terrible results also has terrible skills, front office management can easily react in the off-season; relief pitchers are the easiest and cheapest members of a team to replace.
You can argue the approach (and you will undoubtedly cite the incredible consistency of 60+ inning-per-year Yankee bullpen anchor Mariano Rivera), but I have strong five-year data on my side. (Even the Yankees mean regress from year-to-year but not quite as much as the league average thanks to Rivera, a true exception to virtually any bullpen-centric rule.) As such, I add 3.6 wins to the Mets 2011 win total in starting my 2012 projection, since that’s how much New York’s bullpen finished beneath the league average in 2011.
The Mets’ starting pitchers weren’t quite as bad as the relievers in 2011, but they did finish the season ranked 12th, out of 16 teams, in the NL in terms of runs allowed. That should improve this year simply based on the return of Johan Santana to the rotation. I don’t have Santana approaching his previous Cy Young Award-level of performance but when an entire staff is below league-average, replacing the worst of the lot – in this case Chris Capuano whose 4.39 career ERA (4.55 with the Mets in 2011) is now the problem of the Los Angeles Dodgers – with an above-average pitcher, that alone is worth a couple of wins. Due to a presumed 23 starts from Santana and my model’s strong out-of-consensus appreciation for Jon Niese, (he’s an extreme ground ball pitcher who strikes out more than the league average and walks batters at well-below league average rates. He’s a breakout candidate in 2012.) I’m projecting the starters provide the Mets with two more wins this year despite a certain regression from R.A. Dickey and the 3.28 ERA he posted a year ago.
Equaling the production of the hitters from last year will be a problem for New York solely due to the departure of perennial All-Star Jose Reyes, who by most measures turned in the best year of his career in 2011. His league-leading .337 batting average, .384 on-base percentage, and .493 slugging percentage – all career highs – will not be replaced, at least not by the bat of new shortstop Ruben Tejada. However, it’s possible that a merely healthy line-up from the rest of the starters could offset a lot of the effect of Reyes’ departure. (Note: I’m aware Ike Davis has contracted desert fever, and David Wright may have a more-serious-than-reported oblique injury, but for now the projections are not adjusted for Spring Training ailments but will be when the final projections are updated on the eve of the regular season. This could also significantly affect the NL East competition. See Utley, Chase and the Phillies.)
The Mets had to play a second baseman, Daniel Murphy, at first base most of last year and while Murphy is an above-average hitting second baseman, at first he cedes a lot of production to his counterparts across the league. Getting regular first baseman Ike Davis back in 2012 for more than 36 games, allows Murphy to slide back to second base. This means Davis essentially replaces the combination of Justin Turner and Ruben Tejada at second base. But it gets better for Mets’ fans. Since Tejada is moving to shortstop, effectively Davis, a lifetime .271/357/.460 hitter is replacing Reyes from the 2011 line-up. While Davis’ lifetime on-base and slugging numbers are a bit south of Reyes’ performance last year, suddenly the drop-off is manageable. Factor in that 5-time All-Star David Wright only played 100 games at third base last year, and there is the possibility of added production from that position as well. While I think the infield may actually out produce last year’s unit at the plate, even with the loss of Reyes, there will be a drop-off from the outfield bats. The Mets’ 2011 leader in slugging percentage, Carlos Beltran will not be around to provide the nearly 100 games of production he did last year, before being shipping to San Francisco for future pitching star, Zack Wheeler. In just the 98 games Beltran did play in, he managed to lead the Mets in home runs with 15 (seriously.) While I have the Mets hitters overall producing 32 less runs on offense (the equivalent of just over 3 less wins) in 2012, I have them making up one of those lost wins with better defense. (One final note of minutiae: You'll see the final projection has the Mets scoring 15 less runs, not 32. That's due to an expected small increase in the overall MLB run environment in 2012.)
The Mets have changed the configuration of Citi Field and the effect on both the hitters and the pitching staff is unknown at this point. However, the Mets staff induces groundballs at an above average rate so perhaps the shorter fences will aid its hitters more than it will hurt the pitchers. I haven’t tried to account for that new factor in 2012 but adding 3.6 wins from the bullpen, 2.7 from the starters, 1.4 wins from better defense, while subtracting 3.1 wins from the offense give the Mets a net gain of 4.6 wins. Add that to the team’s restated wins above of 78.4 and I get 83 wins. Good enough for 3rd place in the NL East and a down-to-the wire battle for the last wild card spot.
This projection is heavily dependent on average health from the projected starters. In the case of Davis and Wright that means between 125 and 135 starts apiece. If instead, either one spends a month on the disabled risk, the Mets have very little chance of finishing above .500. And, two weeks from now, that may just be what the updated projection reveals. But for now, Blue Horseshoe likes the Flushing Nine for a trade.
Oddsmakers’ expectations: It takes a little bit of digging to realize that Ike Davis for Jose Reyes is the effective change in the bats the Mets’ infielders will bring to the plate in 2012 compared to 2011. That is, without a doubt, not a 5 game drop-off but I believe Reyes’ departure is why the Mets over/under for total wins, set no higher than 72, is five games below last year’s actual wins. Improved defense from both a healthy Wright at third base and anyone in right field with knees better than Beltran’s in 2011, plus better results from the pitching staff largely due to the hidden improvement from the a mean-regressing bullpen adds up to a great deal of value on the over. With the caveat that this is a preliminary projection until all pre-season injuries are known, right now the Mets, at a market of 72 wins, represent the over prediction where my conviction is strongest.
83-79 – Second in NL East
703 Runs Scored 688 Runs Allowed
Mop Up Duty:
Joe Peta is the author of Trading Bases, the Newsletter, a companion piece to Trading Bases, A Story about Wall Street, Gambling, and Baseball* (*) Not necessarily in that order, a Dutton Books/Penguin (U.S.A.) February, 2013 release.
If you have been forwarded this issue and would like to be placed on the mailing list, please send an e-mail to email@example.com
All newsletter archives are located at http://tradingbases.squarespace.com
You can follow me on Twitter here: @MagicRatSF
I will be updating progress on the path to publication on Facebook as well where I can be found here: http://www.facebook.com/#!/profile.php?id=1761681056
If you want to be taken off the e-mail list, please let me know at firstname.lastname@example.org