Friday, January 14, 2011

An Unbalanced Bench

Jim Thome will return to the Twins in 2011 on a 1-year, $3M (+ incentives) deal. He was a godsend for the Twins last season, hitting .283/.412/.627 with 25 HR, most coming after Justin Morneau was forced out of the lineup. His $1.5M base salary made Thome the free agent steal of 2010. Though he doesn't come quite so cheaply in 2011, this contract is still a bargain, especially if he can produce anything close to what he did last year.

Still, I'm lukewarm about this signing, for the following reasons:

Reason #1: Brett Favre

A sure-fire Hall of Famer, his numbers steadily declining over his late-30's, signs a 1-year deal with his former rivals and rattles off one of the most impressive statistical seasons of his career, helping his new team to an easy division title. Bring him back for one more year! Oops. The old man battled injuries all season, and when he made it onto the field, the results were usually pretty dreadful. The Vikes would have been better off with just about anybody else behind center.

But football is very different than baseball, and QB is a much more important position than part-time DH/bench bat. For a baseball comparison, look at Frank Thomas. He was limping through his late 30's, so the White Sox dropped him after 2005 (replacing him with Thome). Thomas signed a 1-year deal with Oakland and unleashed one last great season at age 38, hitting .270/.381/.545 with 39 bombs for the last A's team to make the playoffs. Toronto signed him up for 2 years that offseason. He wasn't nearly as potent in 2007 (.277/.377/.480 with 26 HR in 25 more PA), then was released halfway through the 2008 season after he could muster only a .723 OPS through 71 games.

Very few professional athletes can compete at a high level into their late thirties, let alone their forties. Even the very greatest are humbled by the relentless effects of aging. They may have one more great year in them as Favre, Thomas and Thome did. But they are almost never better at 40 than they were at 37-38. I doubt Thome will be any different. At age 37 his OPS was .865. At age 38 it was .847. The Twins would do well to expect his OPS to drop somewhere around .200 points from last season's lofty 1.039. If he's mostly able to avoid facing lefties, he might do a lot better than that, and I'd certainly take Thomas' 2007 line if Thome can match it. But I still wouldn't be shocked to see 2011 turn out to be Thome's least productive year since he was a rookie.

And don't forget, he doesn't play defense, he's much weaker against LHP, and he can't run. He's on the roster to hit against RHP, and that's it. 40-year old Jim Thome will be just as much of a luxury for the Twins as the 39-year old version. Which brings me to:

Reason #2: The Bench

Without Thome, the Twins still figured to have 4 left-handed batters in their everyday lineup: Denard Span, Joe Mauer, Justin Morneau, and Jason Kubel. Not surprisingly, the Tigers and White Sox are loading their bullpens with 2-3 LHP in order to match up with those hitters in the late innings. Here's how our guys have done against LHP over the last 3 seasons:

Span: .298/.380/.419 (+.060 OPS compared to RHP)
Mauer: .324/.388/.461 (-.116)
Morneau: .290/.343/.496 (-.119)
Kubel: .233/.312/.351 (-.218)
2010 AL vs. LHP: 256/.325/.393

Span would prefer to see a lefty on the mound. Southpaws reduce the M&M boys from godly to merely great. Still, all three are healthily above the league average against portsiders. No need to pinch-hit for any of them. But Kubel... ugh. LHPs transform him from a beast into a bee-yatch. It's crucial that the Twins have some way to protect Kubel from facing lefties, especially in late, game-changing situations. I would even say that it's the most important strategic position on their bench.

As of right now, it appears that the Twins' bench next season will consist of Thome, backup catcher Drew Butera, Jason Repko for the outfield, and Matt Tolbert as the utility infielder. As Christina Karl of Baseball Prospectus said (before Thome was signed), that group has the potential to be an "infamously bad" bench. They can all hit right-handed against LHP, but none of them do it well:

Repko: .211/.274/.281
Butera: .183/.203/.250
Tolbert: .271/.316/.393

The best the Twins can do for Kubel is to send up Tolbert, who is only slightly below league average and outslugs Kubel by a mere .042. It isn't really worth it to PH him and have Tolbert as the DH for the remainder of the game. If I were Gardy, I'd just take my chances with Kubel. Which plays right into Ozzie & Leyland's hands.

For his career, Thome has hit .238/.339/.424 vs. LHP. A fair amount better than average, better than Tolbert, way better than Kubel. But would Gardy pull Kubel back against a LHP and send Thome up instead? It looks to me like the Twins are still going to be vulnerable in that situation. Now, if the Twins want to protect themselves against late-inning lefties, it would require Gardy to do a couple of things for which there is no recent precedent under his tenure:

1. Make Cuddyer a Platoon Player

Consider Cuddyer's recent work vs. LHP in light of the list I gave above:


That's a mile better than Kubel. Obviously, you want Cuddy out there against lefties with Kubel on the bench. Against righties, it should be nearly as clear. Between 1B, LF, RF and DH, the Twins have 5 players for 4 positions. Thanks to Delmon Young's emergence in 2010, I expect him to be in LF just about every day. That leaves Morneau, Thome, Kubel and Cuddyer to fit into the last 3 slots. Here are their 3-year splits vs. RHP:

Morneau: .306/.407/.551
Thome: .266/.396/.543
Kubel: .289/.354/.527
Cuddyer: .261/.319/.423

Now, which one of those guys should be riding the pine against righties? It's not even close. And, with Cuddyer on the bench, not only is the heart of the order a RHP killing machine, but Cuddy is available to PH late in the game to protect Kubel from facing a tough lefty.

The advantages of a Cuddyer/Kubel platoon are obvious, but I don't expect it will happen. Cuddyer has played at least 144 games in each of the last 5 seasons in which he didn't have broken bones. Also, a Cuddy/Kubel platoon would cost $15.75M, of which 2/3 would be paid to the guy getting 1/3 of the PA. Such a sunk cost shouldn't have any bearing on how the lineup is constructed, but that doesn't mean it won't. Cuddyer is going to get a lot of PAs against RHP, at the expense of a righty masher. Unless...

2. Make Cuddyer the Super-Utility Guy

Last year, Cuddyer made starts at RF, CF, 1B, 2B and 3B. He's a defensive liability at all of them. But even against RHP, he's a much better hitter than Tolbert or Repko. The Twins could plan on having Cuddyer start in RF against all left-handed starters (about 1/3 of the season), then spell all four IF and the other OF a couple of times a month. When Young sits, Kubel would probably start in LF. When Casilla gets a day off, Nishioka could play SS. When Span is out, Nishioka leads off. The thought of Cuddyer in CF is harrowing, but so is the thought of Repko hitting - he can come in as a defensive replacement late in the games.

That would give Cuddyer about 55 GS as a RF vs. LHP, a dozen each as a RF, CF, 1B and 3B and 2 dozen as a 2B vs. RHP. That adds up to 127 starts - nearly 4/5 games. That, plus frequent PH opportunities, would hopefully make him and Gardy feel like they were getting everything they could out of him. And it would mean that the Twins could do without Tolbert altogether, opening up the last bench spot for someone who can DH against lefties.

But it's hard enough to imagine Gardy making due without Nick Punto, let alone Tolbert. And I admit it would be useful to have a good IF glove available to sub for Cuddyer late in his starts at 2B and 3B. So, assuming Tolbert still makes the team, the only way I can see that the Twins can avoid the LOOGY trap with Kubel is...

3. Have a 5-Man Bench

That used to be a pretty ordinary thing. The Twins could still sign one of the remaining RH hit-only FA left on the market. Here are a few who could top the league average (and Tolbert), based on their work over the last 3 seasons:

Troy Glaus: .225/.352/.407
Manny Ramirez: .288/.396/.489
Vladimir Guerrero: .297/.351/.481
Marcus Thames: .265/.333/.500

Of those, I'm guessing Glaus and Thames would be the cheapest. My preference would be for Glaus, since he has a little more credibility as a corner IF than Cuddyer. He's broken down in the NL the last couple of seasons, but if he only took the field a couple of times a month, I bet he would hold up pretty well. His 3-year splits vs. LHP don't look so hot, but for his career they're much better (.273/.396/.542). Also, if the opponents counter with a RHP, Glaus is far from hopeless there, too (.248/.344/.471). But any of those guys would do the job of giving opposing managers something to think about when they try to match up with Kubel.

However, a 5-man bench means a 6-man bullpen, and that, for some reason, almost never happens anymore. It shouldn't be a hardship if the Twins can get more out of their rotation this season (which should be their objective in any event, given the uncertainty surrounding most of the projected members of the 'pen). The 2010 'pen threw 457 IP, 31.5% of the Twins' total. Split 6 ways, that comes out to a little over 76 IP per guy, which is a much higher workload than the typical reliever accumulates these days. 70 IP would be a bit more reasonable, though still a number that would stretch most of the Twins' bullpen arms to their limits:
  • Matt Capps has thrown over 54.1 IP just once in the last 3 seasons;
  • Jose Mijares' career high as a major leaguer is 61.2 IP;
  • Joe Nathan, as a Twin, has averaged right around 70 IP, but he's coming off TJS;
  • Pat Neshek reached 70 IP in 2007, but he's also trying to come back from TJS
It isn't clear that the Twins have 6 relievers who could handle 420 IP. And that amount would still leave 37 extra IP for the starters to pick up in 2011.

Could they handle that? Carl Pavano, should he return, shouldn't be expected to do any more than the 221 IP he gave the Twins last year. Francisco Liriano ought to be able to improve substantially on last season's 191.2 IP - a full 33 GS and better luck on balls in play should get him easily to 210 IP. If those two could pull that off, then the other 3 starters still have to average 200 IP. Nick Blackburn managed that between 2008-2009, but was so dreadful at times last year that his combined workload between the Twins and Rochester in 2010 was just 182.2 IP. Scott Baker has reached 200 IP just once in his career. Kevin Slowey and Brian Duensing have never done it.

Much as I'd like to see the starters pushed harder in 2011, I doubt that the Twins will feel they have the luxury of carrying only 6 relievers for any part of the season.

Thome is still a productive hitter, and a great guy, and his chase of the 600 HR plateau will undoubtedly be great for business this summer. But unless the Twins surprise me by enacting one of the scenarios I enumerated above, his signing means that they'll be susceptible to LHP late in games. And that could cost them more games against their AL Central rivals than whatever is left of Thome's bat can replace.

1 comment:

Ben said...

What you say about splits makes sense on paper, but I think you have to consider the need to keep Thome healthy by limiting his playing time. I think they should use the same approach they had last year before Morneau got hurt. It sure would be nice if Kubel or Thome were right handed though. If Pavano is signed, I wonder if Kubel and a starter are actively shopped.