Very interesting article from Candidate 6— Saber guy.
We continue our look at ways to improve the 2011 Mets with an examination of potential bullpen options. Once again, as a reminder, we’re looking for players that advanced statistics can identify as bargains. These may not necessarily be the best players available, but they are ones who might provide some additional bang for the buck.
I don’t need to tell any Mets fan how important the bullpen is to a team’s success. To demonstrate, how different would 2007 (where the bullpen played an oft-forgotten role in the team’s collapse) and 2008 (where the bullpen was front and center in the team’s demise) have been without the following moves and non-moves:
1) Mets trade Heath Bell and Royce Ring to Padres for Ben Johnson and Jon Adkins.
2) Mets trade Matt Lindstrom and Henry Owens to Marlins for Jason Vargas and Adam Bostick.
3) Mets trade Brian Bannister to Royals for Ambiorix Burgos.
4) Mets do not resign Chad Bradford and Darren Oliver.
5) Mets sign Scott Schoenweis to 3-year $10 million contract.
Interestingly, most of those moves were done out of the one sabermetrically-inclined tendency Omar showed as GM – the belief that relievers are largely fungible and prone to somewhat random fluctuations in their performance due to the limited number of innings they throw each season. In line with this philosophy, until 2009, outside of Billy Wagner, Omar did not throw big money at the pen, and was inclined to find new parts each off-season.
I certainly don’t believe a lot of money should be spent on relievers. But that’s only because I don’t think it’s that hard to identify quality pitchers who can make a valuable contribution on the cheap. Where Omar went astray was that he had those pitchers in his organization and let them go; replacing them with higher-priced, less-effective replacements and getting nothing in return.
In 2010, the Mets had a number of quality performances from their relievers. But, can we tell who is worth bringing back for 2011? Below, I look at each of the six main Mets relievers in 2010 and evaluate them according to four different statistics: their actual ERA, the number of K’s …, their BABIP and their WXRL. The last stat is not a radio station, instead, the good folks at Baseball Prospectus use WXRL to measure how a relief pitcher changes the outcome of the game, taking into account not just their performance, but the importance of the situations in which they’ve pitched. A higher number is better–Joakim Soria and Heath Bell (there’s that name again!) led their respective leagues with WXRL’s of 6.529 and 6.508 respectively.
| Pitcher | ERA | BABIP | SO/9 | WXRL |
| Takahashi | 3.61 | .309 | 8.4 | 2.89 |
| K-Rod | 2.20 | .298 | 10.6 | 2.86 |
| Feliciano | 3.30 | .351 | 8.12 | 1.64 |
| Dessens | 2.30 | .245 | 2.72 | 1.04 |
| Parnell | 2.83 | .381 | 8.5 | 0.23 |
| Acosta | 2.95 | .286 | 9.5 | 0.01 |
More good news, not only did the Mets get performance from their relievers, but their peripheral stats suggest that many of them should do even better next season. In particular, note that Bobby Parnell’s BABIP is extremely high, and we should expect him to have much better results next year. The only guy we shouldn’t ask back is Dessens, who quite simply is lucky.
However, things are not that simple because both Takahashi and Feliciano are free agents. Both are likely to seek two year deals, I would expect Takahashi to seek around $3 million a year and Feliciano to probably want $4 mil a season. I would resign Takahashi, but let Feliciano go and not run the risk that Feliciano would accept an arbitration offer. Feliciano’s numbers are good, and he’s been a loyal Scott Proctor to Jerry Manuel’s Joe Torre. Nonetheless, he’s still just a lefty specialist, and he allowed righties to hit .336/.436/.395 against him last year (lefties were at .211/.297/.276). In contrast, Takahashi, because of his ability to pitch multiple innings, and the fact that he has a good change up that can get righties out, is the more valuable of the two pitchers.
So, we have two spots to play with. Who should the Mets target? I would start with Chad Gaudin. He’s only 28 and struck out 7.3 batters per nine innings last year. And, thinking back to part one, his SIERA was only 4.32, more than a run lower than his 5.65 ERA. And, in a pinch, he can start for you as well. Maybe he gets a one-year deal worth $1 million with incentives.
Next, I’d target Justin Miller, a 33-year old right-hander who pitched 24 innings for the Dodgers in 2010. In limited work, Miller had a 4.44 ERA. But, he struck out 11.1 per nine, his BABIP was an abnormally high .321 and his SIERA was a dominant 2.75. My guess would be he could be had for a Spring Training invite.
Finally, for a lefty-specialist to replace Feliciano, I would invite Taylor Tankersley to Spring Training. The now ex-Marlin is still only 28 and has held lefties to a .223/.313/.372 line, and was even better in limited work at the big league level in 2010. For his career, he’s struck out 8.8 per nine, and has shown a propensity for the long ball, which would be helped by moving to Citi Field.
All this is to say that there is a lot of inexpensive talent out there. If we assume that Feliciano is going to cost $4 mil next year, if we take that money and find a minimum salary talent like Tankersley, there’s almost enough money left over to sign an Orlando Hudson, or better yet, sign some draft picks to above-slot bonuses.
8 comments
Prismo
10/26/2010-1:49pm at 1:49 pm (UTC -4)
Nice piece, though I would contend that 24 innings is far too small a sample size to pull any meaningful metrics from.
TRS86
10/26/2010-1:53pm at 1:53 pm (UTC -4)
While true, it’s those kinds of guys that if you can get to come in for a minor league invite can really shape your pen.
Ceetar
10/26/2010-2:20pm at 2:20 pm (UTC -4)
No, I don’t think Acosta or Dessens should be back, and while Parnell’s sample size was small, I’ve liked what I’ve seen from him in general.
I differ on Feliciano though. I’d offer him the arbitration. I don’t think 3-4 million is something that’s going to keep the Mets from doing everything they need to do, and I kinda like a little of the “known quantity” in Feliciano. If he does walk, you get the extra sandwich pick or whatever it is you get in compensation. I’d rather more picks than just throwing more money at a couple of “better” picks.
stickguy
10/26/2010-2:24pm at 2:24 pm (UTC -4)
I don’t mind Acosta back. He showed enough to be a viable candidate. I agree though that Elmer should go, and offer the arb to Pedro, hoping that he gets an offer elsewhere.
He pitched a ton of games, but as noted, how much do you want to spend on an OOGY?
TRS86
10/26/2010-2:28pm at 2:28 pm (UTC -4)
I have no problem with Elmer at least getting a ST invite. Guy has actually produced the last 2 years. He is not a go to guy but for sure someone who can eat innings and do it well enough to keep us in games. In some ways he has been the RH Oliver for us.
stickguy
10/26/2010-2:22pm at 2:22 pm (UTC -4)
I am all for looking for the cheap options for the pen, instead of chasing guys on the back side for big $$.
and Prismo, a problem with many RP options (once you get past closers and locked up primary set up guys ) is that they usually don’t pitch that many innings, so you end up with small sample sizes.
So again, you have to combine the stats side (sure, that Miller guy had nice #s) with the scouts that actually saw him pitch and have an idea about what they think he will be able to do.
Toughest thing about any stat is always figuring out to translate it to the future. Excellent snapshot of the past though.
GMs keep their jobs by identifying who WILL be good, not by telling you who WAS good.
Ceetar
10/26/2010-2:42pm at 2:42 pm (UTC -4)
It’s sentences like this that make advanced statistics hurt peoples brains:
Ceetar
10/26/2010-2:45pm at 2:45 pm (UTC -4)
the formula for SIERA: (since I didn’t know, I imagine most of us didnt
SIERA = 6.262 – 18.055*(SO/PA) + 11.292*(BB/PA) – 1.721*((GB-FB-PU)/PA) +10.169*((SO/PA)^2) – 7.069*(((GB-FB-PU)/PA)^2) + 9.561*(SO/PA)*((GB-FB-PU)/PA) – 4.027*(BB/PA)*((GB-FB-PU)/PA)
It’s basically a ERA scaled stat that tries to equate to ERA but trying to eliminate luck (so things like lots of fly balls in Yankee Stadium being HRs being a luck thing. or having a crappy 2B that lets a lot ground balls shoot through) Built off of fielding independant pitching stats,etc.