So far this offseason there has been an element of the Red Sox and the Yankees trading jabs. The Yankees added Curtis Granderson. The Sox made John Lackey insanely wealthy. The Yankees brought back Nick Johnson. The Red Sox signed Mike Cameron. The bombers traded for Javier Vazquez, and now the Red Sox have inked a deal with Adrian Beltre. Some of these moves were meant to replace departing players, and neither team may be finished, but now seems like a great time to approach the question: which team is currently better on paper?To do this I am going to run through the rosters of each team, as I believe they are currently constructed and see which lineup and rotation produced more wins in 2009. Baseball Prospectus and Fangraphs, respectively, will both be providing the WAR (or WARP1 for BP) stat, which is being used because it takes offense and defense into consideration. These two sites don't always agree on a player's win contributions, which is why I'm using both. The purpose for using the 2009 numbers, as opposed to a 2010 projection, is to be conservative in the estimate. Some players will play better in 2010; others will play worse. Hopefully for the players taken in aggregate their combined 2009 seasons will roughly match 2010.
On a final note, BP adjusts its WARP statistics from time to time, so the numbers may not align to what I've posted previously, but do reflect what is currently on the site.
C
Victor Martinez - WARP1: 2.7; WAR: 4.9
Jorge Posada - WARP1: 3.2; WAR: 4.0
For this position, I'm inclined to agree with Fangraphs. Martinez was probably a bit better than Jorge in 2009. The slight difference aside, these are two very similar offensively capable, defensively inept, switch-hitting catchers.
1B
Kevin Youkilis - WARP1: 4.7; WAR: 5.6
Mark Teixeira - WARP1: 4.4; WAR: 5.1
As much as it pains me to see that Youk was slightly better than Tex in '09, I'm not surprised when the former's higher OBP is taken into consideration. On the other hand, our first baseman is one of baseball's best and he doesn't look like a caveman.
2B
Dustin Pedroia - WARP1: 3.9; WAR: 5.2
Robinson Cano - WARP1: 5.1; WAR: 4.4
I'm always stunned when I see the high value BP places upon Cano, but 2009 was in many ways his best season as a pro. Pedroia, on the other hand, regressed considerably from 2008, regardless of which site you use to analyze him.
3B
Adrian Beltre - WARP1: 1.3; WAR: 2.4
Alex Rodriguez - WARP1: 5.0; WAR: 4.4
A-Rod would have been worth more wins had he not been hurt. It's tempting to project his win total upwards, but he has also played less than 140 games in each of the last two seasons. A similar argument can be made for Beltre, who missed much of 2009 with injuries as well.
SS
Marco Scutaro - WARP1: 5.6; WAR: 4.5
Derek Jeter - WARP1: 5.2; WAR: 7.4
Fangraphs lists Jeter as the better offensive and defensive player in 2009. I did not set out for this post to serve as a de-facto comparison of BP and Fangraphs, but in the back of my mind there is also an element of that. Right now Fangraphs is winning. There's no way Marco Scutaro was better than Derek in 2009.
LF
Jacoby Ellsbury - WARP1: 1.7; WAR: 1.9
Brett Gardner - WARP1: 0.6, pro-rated (0.3 actual); WAR: 2.1
BP penalizes a player for missing time. I've doubled Gardner's win total only because he played so little in '09 and is currently going to start in 2010. Either way, Fangraphs is a big supporter of that gritty little guy. For my part I think it depends upon how much you value defense overall. Ellsbury is a terrible defender while Gardner is excellent. Neither can hit the ball much farther than a little leaguer.
CF
Mike Cameron - WARP1: 4.1; WAR: 4.3
Curtis Granderson - WARP1: 2.7; WAR: 3.4
Granderson is one of the few players on either team I believe will see a big improvement in 2010. Regardless, I'm sticking with the 2009 results, which give Cameron the slight nod.
RF
J.D. Drew - WARP1: 3.6; WAR: 4.7
Nick Swisher - WARP1: 2.9; WAR: 3.5
BP feels more accurate here. Drew always misses a lot of games while Swisher is more durable. I'm skeptical of any system that suggests Drew is 1.2 wins better than Swisher; even if Fangraphs is projecting out to a full season, I would counter that it is unwise to do that if a player played more than 120 games.
DH
David Ortiz - WARP1: -0.1; WAR: 0.7
Nick Johnson - WARP1: 2.4; WAR: 2.4
I'm stunned that Nick Johnson is the only player the two sites agree on. I'm thrilled that David Ortiz still sucks.
So far, both BP and Fangraphs feel that the Yankees will put the better offense on the field next year, based on 2009 as a guide. The bombers come out ahead 31.5 wins to 27.5 according to BP and 36.7 wins to 34.2 according to Fangraphs.
Predicting either team's rotation is a bit tricky. Fortunately, I don't have to. At the moment each team has 6 potential starters and a confirmed closer. I'll limit my analysis to those arms. Also, BP doesn't provide its WARP1 stat for pitchers on the free portion of their site, so I won't link to it.
SP
Josh Beckett - WARP1: 4.2; WAR: 5.3
CC Sabathia - WARP1: 5.3; WAR: 6.0
SP
Jon Lester - WARP1: 5.8; WAR: 6.2
A.J. Burnett - WARP1: 3.2; WAR: 3.1
Lester's strong 2009 makes me wonder if Beckett will continue to be first in the Sox rotation in 2010. Either way, it's difficult to ignore the fact that the Red Sox's 1-2 punch was better than the Yankees' in 2009.
SP
John Lackey - WARP1: 3.2; WAR: 3.9
Andy Pettitte - WARP1: 2.4; WAR: 3.3
I was a strong advocate of John Lackey's until I saw the size of the Sox's offer. At $17 million per season I'm glad he's not the Yankees' problem.
SP
Daisuke Matsuzaka - WARP1: 0.4; WAR: 0.5
Javier Vazquez - WARP1: 7.0; WAR: 6.6
Vazquez projects to be a 5-win pitcher next year, meaning he will not repeat his 2009 performance. Still, even if he's only a 4-win pitcher he'd be a top 3 guy on either team. Am I still smiling about this trade? You bet I am. Dice-K, on the other hand, remains a disappointment with a full no-trade clause.
SP/RP
Clay Buchholz - WARP1: 2.1; WAR: 1.2
Joba Chamberlain - WARP1: 1.0; WAR: 1.5
SP/RP
Tim Wakefield - WARP1: 2.0; WAR: 1.9
Phil Hughes - WARP1: 2.6; WAR: 2.2
CL
Jonathan Papelbon - WARP1: 4.6; WAR: 1.9
Mariano Rivera - WARP1: 6.5; WAR: 2.0
Fangraphs is surprisingly cruel to closers in terms of their value. Here I'm inclined to support BP. If you don't believe me, just ask a Phillies fan about Brad Lidge.
The Yankees come out ahead in terms of their pitching as well, according to each site. BP favors the Yankees 28 to 22.3 while Fangraphs favors the pinstripes 24.7 to 20.9.
Researching this post has led me to be that much more skeptical of the various wins statistics on the web, only because the two sites disagree so often, many times not even agreeing on which player is better. That said, in aggregate, each method draws the same conclusion: The Yankees' core looks stronger than the Red Sox's core. Baseball Prospectus puts the final tally at 59.5 to 49.8, while Fangraphs comes to 61.4 to 55.1. The actual 8-win difference between the two 2009 teams is right between these two estimates, adding credibility to the final tallies.
So far in the offseason, advantage Yankees.
I think you have to expect a healthy Matsuzaka would put up a lot better numbers than are represented here as well. He had a WAR of 3.3 in 2008/WARP1 of 4.9, which puts him a bit more on par with Vasquez.
ReplyDeleteSay Matsuzaka is a 2.5 win pitcher, and as you say, Vaszquez is a 4 win pitcher, you've got a swing of nearly 5 games.
I guess we'll wait and see.
A replacement level team projects to win 49 games, so taken in total the Yankees, if their starters all repeat their 2009 performances, win between 98 and 101 games. Of course some people will probably regress, but others may improve, and the bench and bullpen isn't accounted for. So I'd say 100 wins seems about right.
ReplyDeleteIts true that Matsuzaka's total is artificially low due to injuries, but that holds true for other players as well, including Alex Rodriguez and Adrian Beltre. I decided to avoid projecting any injured players for an entire season only because I felt it would need to be applied to too many players to keep the post accurate across both teams and because there are bound to be unforeseen injuries this season.
ReplyDeleteOn the whole I'm happy with the outcome of the projections because they are inline with what I've seen in other places. So far I've been reading that the Yankees project to be about a 100 win team. I was concerned this method would have resulted in something materially different from that.
Sorry - advantage Bosox unfortunately.
ReplyDeleteThose statistics in most cases are either statistically insignificant or don' take into account injuries, intangibles etc.
ReplyDeleteI think Twasp just won the prize for my favorite comment in Yankeeist history.
ReplyDeleteThanks Larry!
ReplyDeleteMatsuzaka was great in 07 (200+ innings in 32 starts) but has disapointed since. in 08 he made 29 starts but only piched 167 IP. And last year, well, he was better than Wang, I'll give him that! its easy to say he was hurt in 09 and this year he'll recover to 07 or even 08 form, but there is no way to know that. I thought Wang would be back to form in 09 after injuries in 08, and look how that turned out!
ReplyDeleteOh and BP has definitions for WARPs on their site, but can't find one for WAR on FG.
unless someone steps up, the biggest advantage the Red Sox have over the Yankees is their 1-2 punch - Jon Lester is way better than AJ Burnett
ReplyDeleteTex over Youk. cano over pedroia. Jeter over scutaro. Arude over beltre. Swisher over jdrew. Grandy over ellsbury. Johnson over papi. Posada over Martinez. Damon or Gardner/? Platoon over Cameron. No contest Yankees .
ReplyDeleteCloser: Sandman over crazy ass papelbon.
Middle relief: Hughes/Marte/Robertson over those Japanese guys in Boston bullpen.
Starting pitching : cc over Beckett , vasquez over Bucholz, petite over lackey.
If it wasn't for Randy Johnson oops I mean aj Burnett , it would be a clean sweep.
How'd you like that comment , Larry. Second best on yankeeist ever?
ReplyDeleteI looked all over to find out how Fangraphs defines its WAR statistic, but was not successful. I believe, however, they are projecting the player to a full 162 games. This is purely inference, given the numbers, though.
ReplyDeleteFrankly, I've come away leery of either statistic. Wins based value metrics are enticing because they attempt to capture a player in a single number, but they lose value if we can't agree on a single method. Why does BP recalculate so often? As far as I can tell its to make me look bad. Why do the two sites disagree so severely? I wish I knew the answers.