THE BOOK--Playing The Percentages In Baseball

Ian O’Connor, I do believe he is talking to you…

Sure, the results look the same, but:
a) He’s changed his approach, adding (variously) a changeup and a two-seam fastball.
b) Must sustain velocity for 100 pitches instead of 20.
c) Must face a lineup twice or three times rather than a few situational matchups.

It’s easy to sit back and say “Told you so”, but it’s not like this was a sure bet to work. If he couldn’t command that fourth pitch and couldn’t sustain velocity he was going to get lit up. It just so happens that he can command the pitch, and can sustain velocity, at least up to this point.

Well, The Book makes it quite clear that it’s much harder to start than relieve, to the tune of around 1.00 runs per game difference.

If a guy is a 2.50 RA per game reliever, he should be around 3.50 as a starter.  Some might be 3.00 and others might be 4.00.  And if he was the closer, then you might have a good argument for keeping him in the pen.

But as an 8th inning pitcher?  That hardly makes sense, then, or now.  The only alternative for Joba was what Johan went through and what Earl Weaver put his soon-to-be-great starters through: make them long men to get their arms in shape.

To purposelly want to keep a great young pitcher as an 8th inning pitcher is ludicrous.

Dan - Take scouting reports for what you will, but I had read about his talent to sustain velocity throughout his minor league starts. And take his minor league/winter? league starts for what you will, but he was dominant.

As for changing his approach, he had always been a starter, and had to change his approach to become a reliever. Having been a full-time starter only one year earlier, I think his forgetting how to start would have been more fluke than risk.

As for his changeup, well, he doesn’t throw it much. Fangraphs has it as 2% of his pitches thrown, up from 1% last year as a reliever. I don’t think it has had much of an effect on his numbers.

To be honest, I think the whole “Keep Chamberlain as a reliever” argument was because of two reasons:
People are risk averse.
People would prefer to see Chamberlain relieve (for non-Yankee/Chamberlain fans, the latent desire to see the Yankees do worse or get sub-optimal value).

Yes, people are risk-averse.

And Yankee fans are crazy.  They are so spoiled.  Every year, they have the same problem: “Who’s going to relieve in the 7th and 8th innings?”

Imagine, that is their biggest problem every year: the identity of their 16th and 18th best players.

Do you want Joba to be slotted in the role reserved for the 16th best player on the team?  Are people really that crazy?

It always amused me that anytime Joba being a starter was talked about Yankees fans would bring that (middle relief) up.  As if a team with designs on winning a championship didn’t need to improve on a rotation that prominently features Darrell Rasner and Sidney Ponson or something.

We, as sabermetricians, have no idea who is suited for starting or relief.  We leave that up to the teams and hope/assume they know what they are doing. What we do know, is how different they pitch, on average, when they start versus when they relieve, such that we can make an educated guess as to their value as a starter, an 8th inning reliever (say with an average LI of 1.5) or as a 9th inning reliever (with say, an average LI of 2.0), or any other role.

The people who thought it might not be a good idea to change his role didn’t know any more than we did as to whether he was “suited” best for starting or relieving, so their warnings or trepidations rang hollow. In fact, since he used to be a starter and since the team obviously felt that he WAS suited for starting, they should have assumed that he would be a successful starter to the tune of around 1 run worse than he was as a reliever (after regressing his relief performance of course - turning it into a true talent), and then they could have done the math:  60 IP of 2.50 ERA with a LI of 1.50 or 170 IP of 3.50 ERA with a LI of 1.0.

Of course, the “people” who worried about his conversion are the same people who think that Manny is 4-5 WAR (they don’t know what WAR is, but a superstar is 4-5 WAR by definition and they ALL think that Manny is a superstar), think that Griffey is still a pretty good player, think that Jeter is one of the best defensive SS and all-around players in baseball, that a player’s prime is in his late 20’s/early 30’s, that the Carlos Lee contract was a good one, since he is a “premier run producer,” etc., etc.

So let’s not get too carried away with criticizing them on this one, which is a tiny nuance compared to those other truisms and CW’s.

"that a player’s prime is in his late 20’s/early 30’s”

I know Bill James did stuff in the 80s that suggested prime was 26-28, or something like that.  Is that still the CW among the sabermetrically inclined?  It sure seems that primes are more prolonged these days, and one can offer plausible explanations without using words that begin with an “s” and end with a “teroids”....

ps--annecdotally, & as a Yankee fan, I’d say that it wasn’t so much fellow fans as national sports media meatheads who were all “o noes!” about lacking Joba in the 8th.  bullpen depth, for the first time in years, is actually a strength of the current Yankee team....

I know Bill James did stuff in the 80s that suggested prime was 26-28, or something like that.  Is that still the CW among the sabermetrically inclined?

There’s actually a lot of discussion on that topic over here:

http://www.insidethebook.com/ee/index.php/site/comments/another_aging_study/

But remember, all of that - from James’s work to MGL’s latest study in that thread - refers to a player’s hitting. The fielding aging curve peaks earlier.

Yes what Colin says.  There is some evidence or at least some analysis that suggests that peak hitting may be 29 or so.  As you can see if you read that thread, it is difficult to ascertain the proper aging curves.

As Colin also says, peak fielding (and base running) is probably in the early 20’s, so that we are pretty sure that peak overall production is less than 29, perhaps close to 27.

So Tango… how do you like your crow prepared?

Yeah, I know, no one could have really seen this one coming, but you’ve got to admit the timing was pretty spectacular, re: this thread.

Well nothing says that Joba wouldn’t have been injured in relief and/or that the starter that would have been starting in Joba’s place would have had an injury instead.

My impression is that the Yankees have two fundamental problems.

i) They have too litle quality pitching to go around. Leaving them exposed either in the bullpen or in the rotation however they alocate their personel.

ii) They are unfortunate to be in a tough division with the Red Sox and the Rays.

In my oppinion with an aging team and relativly bad chances of making the playoffs the Yankees made a BIG strategic misstake by not beeing sellers at the trade deadline.

And if that was deemed “politicly impossible” they needed to be much more aggrsive in aquiring pitching.

Anyway, if Joba’s injury is serious I predict that the Yanks are more or less toast…

Bjorn, if you look at the Yankees compared to other teams, they actually have one of the better pitching staffs in the majors, including rotation and bullpen.  Every team can always have more pitching depth, but every rotation would be screwed if they lost a 2.50 ERA starter.  (Maybe not Tampa Bay.)

Dan, obviously injuries happen to whoever, whenever.

In any case, it is beyond stupid to groom someone to be a setup man.  You can groom him to be your closer, or for the rotation.  Since they signed Mo long-term, and since Mo looks like he’s going to age like Gordie Howe or Nolan Ryan, it’s clear there was only one grooming track for him.

There’s no question that they should have made him a 2-inning reliever from the get-go this year, throwing 30 pitches for each game he came in, if not more.

Johan Santana should be the model here.  He was used as Earl Weaver would have done, and he made his debut in the starting rotation at age 23.  He probably should have been used not solely in mopup situations though.