Jenkins shouldn't get 25 meaningful (with meaningful being the key word) PAs versus LHP because he flat out can't hit them. But at the end of the year his overall tally will be more than that simply because there will be games where we are winning/losing 8-1 and there will be times where Werth might need a day off. We just have to hope that the games Werth misses are the ones where Jenkins' automatic out doesn't hurt us.

Another reason I like this platoon so much is because Werth isn't that bad against RHP so you don't automatically have to pull him as soon as a righty comes in from the pen in the 6th inning. That way you can save Jenkins for the opposing team's best RHP in the late innings.

There has been a couple of studies on BP that show that splits are pretty much equal for all players and it is based off of the player's true ability.

Anyway, how would you order the lineup against a lefty in a DH game? I'd start Jenkins and sit Dobbs (unless you start Coste over both of them). But if it came down to Dobbs or Jenkins, I guess you'd start Jenkins as Dobbs can play 3rd and OF.

Phight On! wrote:Jenkins shouldn't get 25 meaningful (with meaningful being the key word) PAs versus LHP because he flat out can't hit them.

but neither could milt thompson and yet he got around 65 pa's versus lefties in 1993 in what is generally considered to have been a nearly optimal use of a platoon.

phorever wrote:

Phight On! wrote:Jenkins shouldn't get 25 meaningful (with meaningful being the key word) PAs versus LHP because he flat out can't hit them.

but neither could milt thompson and yet he got around 65 pa's versus lefties in 1993 in what is generally considered to have been a nearly optimal use of a platoon.

But were more than 25 of the 65 "meaningful"? Milt often came in for his defense in 93 (given that the other option was Inky, not a difficult choice), so perhaps a few of his ABs were late in games with bigger leads?

Phight On! wrote:Jenkins shouldn't get 25 meaningful (with meaningful being the key word) PAs versus LHP because he flat out can't hit them. But at the end of the year his overall tally will be more than that simply because there will be games where we are winning/losing 8-1 and there will be times where Werth might need a day off. We just have to hope that the games Werth misses are the ones where Jenkins' automatic out doesn't hurt us.

The other team puts in a lefty in the eighth. The Phillies are due up with Jenkins-Victorino-Dobbs-Ruiz-pitcher. They've already used Wes Helms to pinch hit two innings ago. The bench is Werth, Coste, Bruntlett, and Snelling. You're down two runs. Do you pinch hit Werth for Jenkins or hold him for Dobbs or the pitcher? Bruntlett can't hit. Snelling is left handed. If you hit Werth here, then you hit Coste for Dobbs? Who hits for the pitcher with two on and one in? What happens if it goes extras?

Platooning is one thing but pinch hitting for Jenkins costs you Werth and the other team may bring in a righty later. I'd let Jenkins bat.

The 25 PAs were based on Dellucci. 27 in 2006 vs LHP and 23 in 2007 vs RHP.

Stu, I think the BP studies say that too frequently the sample size is too small to conclude that their splits are definitively different from normal splits, so they suggest using a weighted average of the average MLB splits and the individuals splits. Jenkins has played long enough that we can conclusively say his splits are abnormally bad.

There was ALSO a study about lefties having more deep plattoon splits than righties. I think that's a great piece of information to know, except I think I saw a study on Hardball Times that said a lot of the discrepancy between lefties and righties (in terms of overall numbers, and I believe in terms of splits as well) has to do with the fact that lefties typically do not play C, 2B, SS, 3B, and except for 3B, those are the positions where you typically expect worse hitting in general. Furthermore, either they showed (or I wondered and now have attributed it to them), that better hitters have more extreme platoon splits-- which makes perfect sense. Howard hits for an 1.000+ OPS against RHP and there's a lot further to fall. How bad could Brett Myers' batting splits be?

MattS wrote:How bad could Brett Myers' batting splits be?

Let's hope we never have to go beyond a small sample size!

Laexile wrote:

Phight On! wrote:Jenkins shouldn't get 25 meaningful (with meaningful being the key word) PAs versus LHP because he flat out can't hit them. But at the end of the year his overall tally will be more than that simply because there will be games where we are winning/losing 8-1 and there will be times where Werth might need a day off. We just have to hope that the games Werth misses are the ones where Jenkins' automatic out doesn't hurt us.

The other team puts in a lefty in the eighth. The Phillies are due up with Jenkins-Victorino-Dobbs-Ruiz-pitcher. They've already used Wes Helms to pinch hit two innings ago. The bench is Werth, Coste, Bruntlett, and Snelling. You're down two runs. Do you pinch hit Werth for Jenkins or hold him for Dobbs or the pitcher? Bruntlett can't hit. Snelling is left handed. If you hit Werth here, then you hit Coste for Dobbs? Who hits for the pitcher with two on and one in? What happens if it goes extras?

Platooning is one thing but pinch hitting for Jenkins costs you Werth and the other team may bring in a righty later. I'd let Jenkins bat.

Was the line-up Rollins, Utley, Howard, Burrell here? Vs a righty starter? You think Charlie's going to bat Victorino 6th? Helms pinch hit in the 6th against the 1st lefty reliever? Who has 2 lefty relievers?

Either way, that's 1AB. That situation will happen, Phight was saying it shouldn't happen more than 25 times a season.

MattS wrote:The 25 PAs were based on Dellucci. 27 in 2006 vs LHP and 23 in 2007 vs RHP.

Stu, I think the BP studies say that too frequently the sample size is too small to conclude that their splits are definitively different from normal splits, so they suggest using a weighted average of the average MLB splits and the individuals splits. Jenkins has played long enough that we can conclusively say his splits are abnormally bad.

Not exactly.

http://www.baseballprospectus.com/unfil ... p?s=phenom

http://proxy.espn.go.com/mlb/columns/st ... id=1728907

MattS wrote:The 25 PAs were based on Dellucci. 27 in 2006 vs LHP and 23 in 2007 vs RHP.

but in those seasons he had 301 and 199 total plate appearances. that means ph and spot starter. do we want that from jenkins?

you should have pointed out dellucci's 2005. he managed 518 plate appearances with just 38 against lefties!!!!!!! wow.

was there a dearth of lefty pitching in the al west that year? looks like it. one lefty starter each on the angles, a's and m's. for comparison, in the nl east in 2007, each of the 5 teams had two lefty rotation spots. i don't know offhand what the current nleast rotation projections are for 2008.

mcare89 wrote:

BigEd76 wrote:

1 wrote:the phillies should sign TO

HUH?!?

They were complaining that the Phils sign these average players all the time but never go after the one player that will make them World Series contenders...

Who exactly is that one player?

Is Babe Ruth available or something?

that's the problem. they have a panel of idiots criticizing the lack of moves, but they had no suggestions for what they should be doing

Disco Stu wrote:

MattS wrote:The 25 PAs were based on Dellucci. 27 in 2006 vs LHP and 23 in 2007 vs RHP.

Stu, I think the BP studies say that too frequently the sample size is too small to conclude that their splits are definitively different from normal splits, so they suggest using a weighted average of the average MLB splits and the individuals splits. Jenkins has played long enough that we can conclusively say his splits are abnormally bad.

Not exactly.

http://www.baseballprospectus.com/unfil ... p?s=phenom

http://proxy.espn.go.com/mlb/columns/st ... id=1728907

I don't really see how this refutes what I was saying. He says you generally need to weight the overall lefty split numbers a certain amount and his another amount, and it depends on the number of at-bats before we can get a good enough sample size. Silver's point was that he needed to feed in a whole lot of historical information before he could conclude a batter necessarily had an abnormally bad split. Jenkins has 1066 career AB against lefties with an 162 point OPS differential. I don't know exactly the normal split, but that seems high enough to warrant suspicion that he'll be relatively worse against lefties than other lefties, I think.

The Karros thing was more about small sample size, which is a problem with ascertaining righties true abilities against LHP.

What I like about this signing:

1. Jenkins will be valuable in 2008. He'll put up a higher OPS against righties than either Victorino or Werth. Players like Lohse and Silva could easily have a repeat of 2006 and be worthless.

2. His price is reasonable.

3. The signing is only two years, so he doesn't handicap them long term.

If the Phillies didn't spend the money on Jenkins or Cameron they probably would have either not spent it or spent it on someone who is shaky like Lohse. I like spending it on Jenkins better.

Gillick sounded like he expected them to add no more than a reliever the rest of the off-season.. So this is the payroll or a touch more.

Laexile wrote:What I like about this signing:

1. Jenkins will be valuable in 2008. He'll put up a higher OPS against righties than either Victorino or Werth. Players like Lohse and Silva could easily have a repeat of 2006 and be worthless.

2. His price is reasonable.

3. The signing is only two years, so he doesn't handicap them long term.

If the Phillies didn't spend the money on Jenkins or Cameron they probably would have either not spent it or spent it on someone who is shaky like Lohse. I like spending it on Jenkins better.

Gillick sounded like he expected them to add no more than a reliever the rest of the off-season.. So this is the payroll or a touch more.

I wasn't big on this deal in the rumor stage but now that it has happened I am happy as pie. Agree with above points, plus:

4. Reportedly a great clubhouse guy, and should help fill void left by Rowand (Not sure if I really buy into this but what the heck - it makes me feel better about the team's chemistry)

Warszawa wrote:should help fill void left by Rowand (Not sure if I really buy into this but what the heck - it makes me feel better about the team's chemistry)

We were -1 on our goatee quota when Rowand left.

bleh wrote:

Warszawa wrote:should help fill void left by Rowand (Not sure if I really buy into this but what the heck - it makes me feel better about the team's chemistry)

We were -1 on our goatee quota when Rowand left.

imagine if Utley grew a goatee. OMGz

It's funny because I don't see too many goatees in my day to day anymore.

Disco Stu wrote:

MattS wrote:The 25 PAs were based on Dellucci. 27 in 2006 vs LHP and 23 in 2007 vs RHP.

Stu, I think the BP studies say that too frequently the sample size is too small to conclude that their splits are definitively different from normal splits, so they suggest using a weighted average of the average MLB splits and the individuals splits. Jenkins has played long enough that we can conclusively say his splits are abnormally bad.

Not exactly.

http://www.baseballprospectus.com/unfil ... p?s=phenom

http://proxy.espn.go.com/mlb/columns/st ... id=1728907

This is a bad application of statistics. Significance is itself a statistic. Common ones include chi square, t-test, and p-tests.

This is why confusing samples and populations is a fundamental error that throws a lot of the multi-variate analyses into question.

Sampling is the process of selecting units (e.g., people, organizations) from a population of interest so that by studying the sample we may fairly generalize our results back to the population from which they were chosen.

http://www.socialresearchmethods.net/kb/sampling.php

A random sample is a portion of a population randomly selected. (So obviously selecting one season, regardless of how many plate appearances we're talking about would be inappropriate--it's not random.)

And when we check what Karros did in 2004, he was terrible. But much more terrible against righties than lefties.

http://www.baseball-reference.com/pi/bs ... &year=2004

philliesr98 wrote:

bleh wrote:

Warszawa wrote:should help fill void left by Rowand (Not sure if I really buy into this but what the heck - it makes me feel better about the team's chemistry)

We were -1 on our goatee quota when Rowand left.

imagine if Utley grew a goatee. OMGz

Or if Howard got that disease that turns your skin white!

Disco Stu wrote:

philliesr98 wrote:

bleh wrote:

Warszawa wrote:should help fill void left by Rowand (Not sure if I really buy into this but what the heck - it makes me feel better about the team's chemistry)

We were -1 on our goatee quota when Rowand left.

imagine if Utley grew a goatee. OMGz

Or if Howard got that disease that turns your skin white!

My skin already is white!