In which analysis is outsourced
I had been playing around with some different back-of-the-envelope projection methods for Yu Darvish, and then I found out that SG over at the Replacement Level Yankee Weblog had actually done all of that work already, and better. The key point to be made, with regard to Darvish, is just how much better he was in NPB than Matsuzaka. These were Matsuzaka’s ERA numbers in his last four seasons in NPB:
2.83, 2.90, 2.30, 2.13
Pretty damn good, right? And those were his age 22-25 seasons. There was every reason to be excited about the guy, it seemed, before we learned his fastball command was a big problem. However, Matsuzaka ain’t got nothing on Yu.
1.88, 1.73, 1.78, 1.49
And those are his age 21-24 seasons. We’re looking at a proportional difference around 75% of ERA - about the NPB equivalent of the difference between Jon Lester and King Felix. Just doing the very simplest conversion, based on nothing but Matsuzaka’s runs prevented and converting it based on Darvish’s runs prevented in NPB - Matsuzaka allowed a 4.25 ERA in the majors over 620 innings, good for 9.6 WAR. If Darvish produced in MLB at a rate equivalent to his superiority to Matsuzaka in NPB, he would project to an ERA around 3.00 and around 18 WAR in 620 innings.
Now, this is a pretty messy and ugly way of translating numbers, but luckily SG has gone and done the legwork for a number of better translations. He figured component translations for each major starting pitcher who has made the NPB -> MLB conversion in the last fifteen years, and then he used those as the basis for projecting Yu Darvish. This produces a pretty simple table - if Darvish translated like player X translated (Matsuzaka, say, or Kazuhisa Ishii or Colby Lewis), he would have the following MLE for his weighted average performance in NPB 2009-2011. The numbers are really, really good.
Player IP H R ER HR BB SO RA ERA FIP
Darvish, Yu 214 150 43 38 6 41 240 1.81 1.59 1.88
Darvish->Average 214 178 62 57 9 51 197 2.62 2.42 2.62
Darvish->Igawa, Kei 214 194 75 75 11 78 189 3.15 3.14 3.19
Darvish->Irabu, Hideki 214 214 72 75 11 37 168 3.03 3.18 2.83
Darvish->Ishii, Kazuhisa 214 173 60 51 7 61 172 2.52 2.14 2.88
Darvish->Kawakami, Kenshin 214 172 65 58 6 85 177 2.75 2.44 3.08
Darvish->Kuroda, Hiroki 214 155 52 45 6 52 247 2.21 1.88 1.97
Darvish->Matsuzaka, Daisuke 214 176 73 70 9 93 211 3.08 2.94 3.10
Darvish->Nomo, Hideo 214 169 50 46 11 29 210 2.11 1.95 2.34
Darvish->Yoshii, Masato 214 187 62 56 12 45 211 2.62 2.37 2.59
Darvish->Lewis, Colby 214 156 59 54 11 96 209 2.49 2.28 3.26
Darvish->May, Darrell 214 187 62 51 8 44 141 2.61 2.16 2.95
Darvish->Worst 214 254 108 115 19 119 116 4.56 4.84 4.93
The “worst” line there, the only one where Darvish doesn’t project as an All-Star, is Darvish’s MLE if he gets the worst of all possible worlds - adding walks like Matsuzaka, adding runs like Kei Igawa, and losing Ks like Darrell May. All of the other lines look like a pitcher than any club would be happy to drop 6/100 on. The numbers are so good that they’re making me question whether I was wrong to be so convinced that Darvish couldn’t break Matsuzaka’s record for monetary outlay for an NPB pitcher. By the numbers, Darvish is easily the second best pitcher on the market, and there would be a non-crazy case for him over CC.
Reader Comments and Retorts
Go to end of page
Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.
1. DarrenRegardless of what happens you can expect every article about Yu to include the names Daisuke Matsuzaka and Hideki Irabu and not in a positive way. Maybe I'm wrong but I still think the posting fee is going to be $35-40 million with the contract being roughly 1:1 bringing the total value to $70-80 million.
Plus, any time we can ignore Darren, that's just fun for all of us!
Here's a youtube video of Darvish's mechanics, which look off to me for reasons I can't pinpoint. He starts slow, with weird glove position, but he gets moving after that and does the little "step over" thing that CBW liked.
The thing I learned, googling Darvish, is that a number of high-payroll clubs without quite the clout of the Sox or Yankees have been linked to Darvish - in particular the Rangers, Blue Jays, and Nationals. None of those clubs relish getting into it with the Yankees in a traditional bidding war for Sabathia or Wilson, and with one big move they could cut the Yankees out of the Darvish stakes. (Just like the Red Sox and Mets did with Matsuzaka.) I would not be at all surprised to see one of them drop $50M or more on the posting fee.
His hand position (hooking) isn't great after he breaks from the glove. He brings the ball almost straight up during his scap load, too - doesn't really "reach back" like you see most pitchers do. That's not necessarily a red flag in and of itself, but combined with the hooking action it does make things look a little awkward. I'd want to watch more, but the first thing I'd worry about is elbow issues there.
Other than that, I like his arm action at release. Good extension and good follow-through.
The most recent reports suggest he's still planning on asking to be posted, thought it may not be a done deal.
I don't disagree with that, but I think it is going to be the reality. I just think given the Matsuzaka hype teams are going to tread cautiously, perhaps to their detriment.
To your point though, it just takes one and I think given the circumstances being similar to 2006 (missed the playoffs, dire need of pitching) the Red Sox could well be the one again.
Well, one data point is that he has hired an American agent (Arn Tellem).
The issues about workload are obviously the most troubling. Given the way kids get worked in NPB, is it ever worth it to commit six years to a young NPB pitcher? As I said in the initial post, if Darvish had translated exactly as Matsuzaka did, he'd have produced a little under 20 WAR for the Red Sox, which at a cost of $100M is not a terribly good deal. It's fine, and no one would really complain, but that's with Darvish putting up very good numbers early in the contract. An injury in the first couple years of a contract kills its value, even if the pitcher is quite good. Darvish has to be reasonably durable to be a good buy.I do want to emphasize, though, that the actual pitching performances of Irabu and Matsuzaka are positive indicators for Darvish. If his numbers translate in a way similar to the way that Irabu's and Matsuzaka's numbers translated, he'll be an ace. He was just that much better than any of these guys.
The Sox cannot cannot cannot make decisions like that right now (ever really). If they think Yu is going to be a star and can handle the scrutiny they should pursue him vigorously. If he pitches well the fans will be fine with him. The only thing the Sox should be asking is if he has the ability to pitch at a high level in the Major Leagues. The single best thing this team can do from a PR standpoint is win 95 games in 2012. If they do that with a Japanese pitcher, Bucky Dent's nephew and Harry Frazee's granddaughter no one is going to care. If they go 82-80 there is no amount of good cheer they will be able to produce.
Except that the Sox look like they really need to add a pitcher. The current climate of "Red Sox Nation" and attendant scrutiny are going to apply to anyone they bring in. The FA market for pitching is pretty thin (C.C. isn't leaving the Yankees even if he opts out, and Wilson could easily cost twice the total commitment to Darvish, with many suitors) and the risk is going to be there anyway. At least the Sox have the experience in transitioning a similar player.
You mean where he doesn't leave NPB and wait until he reaches free agency to transition to MLB?
You have to translate his numbers, scout his stuff and command, and figure out his injury risk. Those are the jobs for Sox, and for any other team that might bid on Darvish.
Specifically, I'm not sure if it's true or not statistically, but I've heard some talk about there being a decline in offense in NPB that's been equal to or greater than the decline in MLB. Is this true? If so, could that be skewing the numbers in Darvish's favor relative to the stats posted by pitchers like, e.g., Irabu and Matsuzaka five or 10 years ago, or do the translations already reflect any such changes?
Also, I've heard the official NPB ball, which reportedly is smaller than the one used in MLB, has changed a time or two in recent years. If true, have the offensive numbers tracked up or down with the change in the ball (as happened in Cuba), or has any offensive decline (or spike) seemingly been unrelated?
People have been grossly overestimating what Wilson will get.
People have been grossly overestimating what Wilson will get.
I don't know about that. I put him at around 100-120m, and I had to convince people here a few weeks ago that he'd get more than Lackey/Burnett. But yeah, Darvish is getting more than half of that (including posting).
No, I just wasn't really paying attention and thought I had made a revision I hadn't when I posted (not that that doesn't mean people aren't over or underestimating what people will get).
I'd actually guess that Darvish will wind up with something comprable to DiceK's deal, with a bit on the top for inflation (say $110M). But I think Wilson is easily looking at $20M yr/ 6 yrs. I can see the Yankees offering $18m for 5-6 and Texas going over that to keep their ace.
For Boston the concern is his salary, IMO. The FO seems concerned with keeping under the CBT threshold, which the posting fee doesn't count towards. But someone made a point in the other thread that a smaller market team could go all in on Darvish since they won't be able to land a top FA pitcher, and then negotiate exclusively. The Nats seem like a good candidate for this maneuver.
Wilson has the advantage of being a known quantity both in terms of temperament and ability. Yu is going to suffer from Matsuzaka fatigue I think. If Wilson doesn't get 5 years I'll be shocked.
7/161 - CC Sabathia
6/138 - Johan Santana
5/120 - Cliff Lee
7/116 - Barry Zito
6/91 - Carlos Zambrano
5/85 - Jered Weaver
5/82 - AJ Burnett
5/82 - John Lackey
Somewhere in between Lee and Lackey seems like a reasonable guess for Wilson. I agree that Wilson will certainly not cost double Darvish, but I think guessing $20M per is pretty reasonable.
You must be Registered and Logged In to post comments.
<< Back to main