Friday, January 07, 2005

Wha wha wha???

I come home for a couple hours before going out again, and this is what I have to put up with? My first reaction is that at least one of the following must be true:

1. The anonymous sources being oft-cited right now claiming the Dodgers are on the verge of signing Derek Lowe to a 4 year, $36 million contract are either have some extraordinary ulterior motives or are victims of a bizarre and expertly executed practical joke.

2. The Dodgers know a lot about Derek Lowe that I don't know and that Dave Wallace either didn't know or ignored until October.

What to make of these reports? My instincts say option #1. But let's do some optimism first.

For one, the starting pitching market in free agency looks pretty thin next season. If the Dodgers want to pick someone up, this might be the only time to do it.

Additionally, the Dodgers do have several injury risks on the staff. This is insurance, I suppose.

Lowe had an excellent season in 2002, maybe he'll repeat that? Thing was, that was built on a lot of luck and defense. His DIPS was only 3.66. Not so much. The past two seasons are probably much closer to his true talent level.

Lowe is an extreme groundball pitcher who had terrible luck last season. While I argued earlier this week that Dodger Stadium is a good destination for fly ball pitchers, that doesn't in itself reduce the value of ground ball pitchers. While I made an argument for an organizational philosophy that emphasized outfield defense and fly ball pitching, the counter-argument is that the ability to acquire starting pitching at any time is so limited that putting all one's eggs in such an organizational philosophy reduces one's options enough to nullify the advantage. This, of course, would be an example, as a limited free agent pitching market limits options.

It's not as if the Dodgers have actually implemented the strategy I advocated; they have (I think DodgerRoger used this phrase in the comments) the UZR all-star team in the infield. Based on past UZR data, I'll suggest roughly a +25, +10, +10, -8 infield for the Dodgers, which translates to a context neutral 37 runs saved above average over the course of a season. Meanwhile, their outfield should be around +25 by my estimation. Since Lowe is a groundball pitcher, he leverages the former much more than the latter. Going with some rough park and league adjustments and adjusting for Lowe's batted ball type ratios and the major difference between Dodger Stadium and Fenway, I translated Lowe's 2003 and 2004 performances to ERA's for the 2005 Dodgers. In making those translations, I didn't look at Lowe's defense-dependent numbers, with the exception of doubles (I could get into modeling that by batted ball types, but it's late).

What do the translations look like?

2003: 3.64
2004: 3.61

Looks pretty cool, huh? Keep in mind, of course, that a lot of the value added there doesn't belong to Lowe, per se. If I run those numbers for a completely average pitcher, the ERA looks like 3.90. Now, the average reliever has better numbers than the average starter, so let's say our baseline for the average starter would be a 4.05 ERA and that Lowe will be worth an extra .40 in ERA. Further, let's say Lowe can go 210 innings and the average starter only 160. If those extra 50 innings go to that average 3.90 pitcher, how much value does Lowe add? 10 runs. Yes, 10 runs.

Okay, maybe that's not fair. Maybe that average pitcher would put more strain on the bullpen, and his extra innings go to a replacement level pitcher who wows with a 5.00 ERA. Well, now Lowe's making us 16 runs versus this average cat. Judged by the standard that the completely average full-time player is worth $4 million, an extra $5 million for 16 wins looks pretty shoddy.

But maybe the Dodgers can't get that completely average guy. They have to allocate those 160 IP to Edwin Jackson or Joel Hanrahan or D.J. Houlton or a bewildered Elmer Dessens, and get a 5.00 ERA out of the deal, on top of that 5.00 strain on the bullpen. Now, Lowe saves a whopping 33 runs, or 3-3.5 wins.

Since they'd have to pay that replacement $320K, that's three and a half wins for the bargain price of $8.68 million! Score! And all it costs them is their first round pick in the amateur draft. Looks pretty rotten.

Then again, there is something to be said for it. It would help the long term development of Jackson/Hanrahan et al to stay in the minors, and it wouldn't start their service clock so quickly. So it could be argued, then, that if those guys will be good to go in 2006, signing Lowe or someone of his ilk means they'll get that prospects' 2008 season at the league minimum instead of at the arbitration price, and they could also hold onto the player for another year of his peak at the end for less than he'd make in free agency. Fair enough, I suppose; maybe worth $4 million.

Then again, that's not an issue with D.J. Houlton. If they don't use Houlton, he's lost anyway. Still, maybe there's not enough faith in D.J. to stop this.

On top of that, the hypothetical average pitcher isn't really available at this point. So getting Lowe for $9 million in 2005 maybe is worth 3.5 wins and saves another $4 million in the long run. Not bad; $1.4 million per marginal win. But there's still the matter of that first round pick; Logan White's got a pretty smoking record right now, so let's arbitrarily assign a $1 million price to losing that pick, since I don't have the data to quantify it otherwise. That's still only $1.7 million per marginal win. With the Dodgers' payroll advantage, that's not bad.


they still have to pay Lowe $27 million for three years in his thirties!!!

I have no idea what entity is responsible for this rumor, whether it's true or not. But that sure doesn't seem characteristic, now does it? $27 million for a pitcher whose performance the past two seasons is only a little bit above average? Keep in mind, for him to add a lot of value from leveraging the Dodgers' infield, they have to continue to have an excellent infield defense, and the three main defensive pieces are under the Dodgers' control for 1, 2, and 3 years, respectively. Maybe Joel Guzman is close and will be converted to a superb defensive third baseman. And defense is typically undervalued by other teams. Still, that's a big investment that requires continued leveraging, and it's tough for me to see how this is a worthwhile long term investment.

Maybe there's something about aging patterns of groundball pitchers I don't know about, or maybe DePo is, in a shocking turn, giving in to the conventional wisdom. Or maybe there's enough marginal revenue to be gained at playoff time that a slight increase in the team's odds in October is worthwhile.


But I'm really, really creeped out.

[Edit: One last point: at least Lowe won't be affected by the loss of foul ground since he gets no infield pop-ups anyway; infield flies in 1% of PA last season.]

A few more comments on lowe. Fenway lead the league in giving up doubles, DS hurts doubles. This is a strength of lowe's already - not giving up doubles so he only get's stronger. He'll probably give up a lot of singles and get a lot of double plays. If he could lower his walk totals he could be very good.
Lowe had something like 28 unearned runs last year, that should drop a lot.
I don't think there is any way around it - we overpaid. I think Lowe will thrive and there is a chance he will be worth the contract, but we definitely overpaid.
Well, I thought I saw on the Baseball Primer thread that one of Lowe's problems WAS giving up doubles. But going from prime double territory Fenway to Chavez should majorly inhibit those. Also, Tom, like you I think the 4/36 may be a Boras inspired story. In any case, the Lowe signing will be contingent on a restructured Green trade which you've convinced me is a major winner. Like you and others, I can't see DePo dishing out 9/yr 3 and 4 years from now for Lowe to block the emergence of Billingsley, Tiffany, Miller, et al. And losing a first rounder is no fun either.
lloyd, you are absolutely correct. My brain got muddled, I think. Lowe was 23 in the AL in giving up doubles. Boston had 4 of the top 31 pitchers in the doubles catagory.
Tom, be sure to check out MGL's post #77 here:

He seems to show that groundball pitchers have a huge advantage in Dodger Stadium, while flyball pitchers are close to league average. The reduction in foul ground should only increase the relative advantage for gb pitchers.
Yeah, I saw mgl's post, and I talked with him about it a little bit earlier. There were a lot of sample issues with what he did, so I'm not certain. For example, the Dodgers have had some pretty poor outfield defenses in recent years and some pretty good infield defenses, and there's no control for that in what mgl did.

In any event, though, I already applied the park factors in my translations, so unless the way that it reduces doubles is unique to GB type pitchers, 3.5-3.7 looks about right.

Given that I've been thinking about this a ton this past week, there is an explanation that would explain to me why the doubles factor would apply more to GB pitchers. Robert Dudek's great article in the THT annual (buy it!) about hang time shows some evidence that GB pitchers, when they give up fly balls, tend to give up fly balls with lower hang times. I can see how maybe the fact that FB pitchers give up balls hit with a lot of hang time means that their doubles aren't affected by the park because the issue is that they stay high and are just hit somewhere that the fielder can't get to. That means that the batter has time to get to second base before it's fielded. GB pitchers, OTOH, give up balls that are sharply hit, and so if the advantage in DS is that the arc of the fence gets those balls back in quickly, then the doubles advantage would apply to them and not the FB pitchers.

That's just an idea, though.

Keep in mind, my post earlier this week was a very cursory look at the evidence to see if it supported a more general theory-- the sample size was so small that it should only be looked at as a starting point. I may have argued the point a little too strongly for what I was going for.
Keep up the good work Tom. It is something that definitely is worth looking at.
"GB pitchers, when they give up fly balls, tend to give up fly balls with lower hang times."

This makes sense to me. A long fly in the gap is a double anywhere, but a liner is a different story. The outfielders in Dodger Stadium can space themselves so they can prevent a variety of extra base hits, without sacrificing anything. Anyway, I appreciate you looking into this. I find it fascinating, since I get the impression that Depo is really building his team around the ballpark.
One more thing, Tom, I was wondering what you think of reduction of foul ground in the stadium. I was thinking of it as a negative since it would make it more of an average park, thereby eliminating an advantage the Dodgers could exploit. Now I'm starting to think the opposite. If indeed it already is a haven for gb pitchers and near average for fb pitchers (and I know this isn't proven), then the changes should make it even more exceptional, since they will mostly hurt the fb guys. Lowe in particular should hardly be affected by the change. I'm starting to wonder if the changes might have been heartily endorsed by Depo. Especially if the aditional revenue gives him a higher payroll.
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?