Adam Dunn Trade rumor thread

[chetkincaid]

WAS is going to wait until near the deadline... offer Dunn a cheap extension... Dunn will likely turn that down... and WAS will trade Dunn to whomever has made the best offer so far.

 

I still think he's going to re sign with the Nats. He seems comfortable with them.

[Wedge]

Right now, CQ isn't worth a whole lot. His effectiveness as a hitter goes hand in hand with his health, and I really don't have any faith in him being able to stay healthy.

 

And to say CQ now is worth more than 2 months of Dunn and 2 draft picks is laughable. Yeah, after 08, but not now.

 

Adam Dunn, June-July 2009

13 HR, 36 RBI, .282 AVG, .384 OBP, .922 OPS.

 

Carlos Quentin, June-July 2009

14 HR, 40 RBI, .275 AVG, .389 OBP, 1.014 OPS.

[Balta1701]

Isn't Dunn's arbitration value around 17-18m? Who the f*** is going to offer him arbitration?

No. Dunn's side could ask for that much but they'd almost certainly lose.

[Balta1701]

Unless my counting was wrong, it looks to me that we could skip the 5th starter without pitching anyone on short rest enough times that a 5th starter needs to make 9 more starts. One of those would be the final game of the season, so I will subtract that one (if we still had not clinched the division going into the final game, someone would go on short rest or multiple starters would be available in the bullpen).

 

So let's say we need 8 starts from a 5th starter the rest of the year (assuming our others stay healthy). What is the difference between Hudson and Torres/Marquez in those 8 starts? How many of those 8 do we expect to win with Hudson vs. the others? I probably should have noted who the starts are against.

Well...first of all, you're assuming there's no negative effects on the guys we have from them pitching with zero extra rest. This season, Garcia's ERA is 6.03 on normal rest, 4.79 with 1 extra day, and 3.12 in games after 2 extra days of rest. Mark!'s ERA is 5.11 on normal rest, 3.67 on 1 extra day, and 2.48 after 2 extra days. Floyd is 4.48 on normal rest and 2.06 on extra rest (this could, of course, just be a function of when his extra rest games came relative to June/July, when he got hot). D1: the Pitching Danks is 3.59 on normal rest and 2.64 on extra rest.

 

So...with every other guy in our starting rotation, the extra day of rest has been worth 1+ runs to their ERA's.

 

Even ignoring that, I'd say that I think Hudson is likely to be close to a .500 pitcher the rest of the way, if he gets run support. I wouldn't be surprised if Torres/Marquez could win a game or two. So there's at least, to my eyes, a 2 win difference between Hudson and the other guys we have, compiled on top of the help of extra rest for the other starters.

 

Trying to use Mr. Offday as your 5th starter is a recipe for giving this division to the Twins.

[beck72]

Adam Dunn, June-July 2009

13 HR, 36 RBI, .282 AVG, .384 OBP, .922 OPS.

 

Carlos Quentin, June-July 2009

14 HR, 40 RBI, .275 AVG, .389 OBP, 1.014 OPS.

Well put, sir!

[BigSqwert]

Well put, sir!

Of course he used a small sample size since Carlos is seemingly injured half the time.

[Balta1701]

Of course he used a small sample size since Carlos is seemingly injured half the time.

And yet he still has more HR and RBI than Dunn over that period of time.

[BigSqwert]

And yet he still has more HR and RBI than Dunn over that small period of time.

Fixed

[ptatc]

Of course he used a small sample size since Carlos is seemingly injured half the time.

 

rant/

 

People need to stop using the "small sample size" defense. The last I looked nobody has done a power analysis to determine what the proper sample size needs to be to predict the outcome of a 162 game schedule. Until that's determined you have no idea what the adequate sample size need to be. People continue to use statistical analysis without regard for the reliability or validity of any of the processes. This is one of the problems with all of Bill James stats. He bases them off things he thinks makes sense, but has never done and will never do reliability studies on them. I've e-mailed him and asked him those questions and he replied once about 10 years ago. His respose was "they seem to make sense to me." He doesn't respond to any statistical analysis questions anymore.

 

I know I've stated this before but it bothers me when people use statistical analysis for their discussions when they really don't know how to use the terms.

 

/rant.

Edited July 28, 2010 by ptatc

[BigSqwert]

rant/

 

People need to stop using the "small sample size" defense. The last I looked nobody has done a power analysis to determine what the proper sample size needs to be to predict the outcome of a 162 game schedule. Until that's determined you have no idea what the adequate sample size need to be. People continue to use statistical analysis without regard for the reliability or validity of any of the processes. This is one of the problems with all of Bill James stats. He bases them off things he thinks makes sense, but has never done and will never do reliability studies on them. I've e-mailed him and asked him those questions and he replied once about 10 years ago. His respose was "they seem to make sense to me." He doesn't respond to any statistical analysis questions anymore.

 

I know I've stated this before but it bothers me when people use statistical analysis for their discussions when they really don't know how to use the terms.

 

/rant.

I guess I don't understand the point of cherry picking a random 2 month stretch from last year to prove a point. Especially since the comparison shows almost equal results from the 2 players.

[ChiSox_Sonix]

I guess I don't understand the point of cherry picking a random 2 month stretch from last year to prove a point. Especially since the comparison shows almost equal results from the 2 players.

 

It's actually this year, he just typed the wrong year.

[clyons]

Looks like a three-way deal is possibly in the works.

 

Hudson and fillers to Arizona.

Dunn to the White Sox

Jackson to Washington

 

If we have to give up a starter to get Dunn, and AZ is willing to give up Jackson for Hudson and fillers, we might as well sweeten the pot for the D'backs, keep Jackson for ourselves, and take LaRoche too.

 

He's not the bat Dunn is, but that would improve our offense without creating a hole in the rotation.

 

 

[BigSqwert]

It's actually this year, he just typed the wrong year.

And that's why he used a 2 month stretch for comparison because Dunn actually plays full seasons whereas Carlos doesn't.

[ptatc]

I guess I don't understand the point of cherry picking a random 2 month stretch from last year to prove a point. Especially since the comparison shows almost equal results from the 2 players.

 

But there is nothing that proves that 2 months worth of numbers isn't predictive of how a 6 months worth of numbers will turn out. An adequate sample size is only large enough to predict what the numbers will be for another given larger group of numbers. 2 months is 1/3 of 6 months. That could very well be a large enough size to predict the end results within a given standard error of measure. Determining the power analysis is always the first step in running predictive analysis that what you know the number of data needed for a reliable and valid outcome.

[chw42]

rant/

 

People need to stop using the "small sample size" defense. The last I looked nobody has done a power analysis to determine what the proper sample size needs to be to predict the outcome of a 162 game schedule. Until that's determined you have no idea what the adequate sample size need to be. People continue to use statistical analysis without regard for the reliability or validity of any of the processes. This is one of the problems with all of Bill James stats. He bases them off things he thinks makes sense, but has never done and will never do reliability studies on them. I've e-mailed him and asked him those questions and he replied once about 10 years ago. His respose was "they seem to make sense to me." He doesn't respond to any statistical analysis questions anymore.

 

I know I've stated this before but it bothers me when people use statistical analysis for their discussions when they really don't know how to use the terms.

 

/rant.

 

Judging anybody based on a two month sample size isn't a good idea. We've seen numerous times where players go on incredible 2-month power streaks (Chris Shelton, Jonny Gomes) and end up playing like absolute crap for the remainder of the season. As unlikely as it seems that a nobody could run into 20 some odd home runs, it is possible, but somewhat improbable. But still possible. I just read yesterday that the NFC had won 14 straight coin tosses in the SuperBowl, the odds of that happening are 1 in 16384. Seems incredibly improbable, but it's still happening.

 

http://www.fangraphs.com/blogs/index.php/a...tingrandomness/

[ptatc]

Judging anybody based on a two month sample size isn't a good idea. We've seen numerous times where players go on incredible 2-month power streaks (Chris Shelton, Jonny Gomes) and end up playing like absolute crap for the remainder of the season. As unlikely as it seems that a nobody could run into 20 some odd home runs, it is possible, but somewhat improbable. But still possible. I just read yesterday that the NFC had won 14 straight coin tosses in the SuperBowl, the odds of that happening are 1 in 16384. Seems incredibly improbable, but it's still happening.

 

http://www.fangraphs.com/blogs/index.php/a...tingrandomness/

 

But you and this article are doing the same thing that Big Sqwert admonished Balta for. You are picking events to justify an idea you have about the sample. It's guessing.

 

I'm not saying that 2 months of numbers is an adequate sample size or not. I haven't done the power analysis either. But you using the term sample size wrong. You can say that he is bias in using certain data to make his point, as you and the article did above. But before you start using the statistical terminology in relation to statistics, not events, you should have the data to back it up. There is always statistical error, like the example you pointed out. This is another reason why people shouldn't live and die by Bill James' made up stats during the discussions on this site.

Edited July 28, 2010 by ptatc

[chw42]

But you and this article are doing the same thing that Big Sqwert admonished Balta for. You are picking events to justify an idea you have about the sample. It's guessing.

 

I'm not saying that 2 months of numbers is an adequate sample size or not. I haven't done the power analysis either. But you using the term sample size wrong. You can say that he is bias in using certain data to make his point, as you and the article did above. But before you start using the statistical terminology in relation to statistics, not events, you should have the data to back it up. There is always statistical error, like the example you pointed out. This is another reason why people shouldn't live and die by Bill James' made up stats during the discussions on this site.

 

I'm sure somebody has tried to do a power analysis. Guys like Tom Tango and Mitchel Lichtman have been doing sabermetric research for decades. They have a whole book dedicated to things like bunts, clutch hitting, lineup order, hot and cold streaks, etc. If you dig deep enough, you might find something about power. I haven't come across it myself, but it might be out there.

 

Bill James is just some guy who is looked at as the guy who makes up all these stats nobody has an idea about, but he's not really one of the current innovators of sabermetrics. Asking him about something 10 years ago probably isn't going to give you the whole story.

[harfman77]

The other problem you have with Dunn is finding a position for him. He has said many times that he does not want to DH so it doesnt make sense to go trade for a guy who doesnt want to be here. He is not as good defensively at any position as the players we have now. The other question is if the Sox make the trade will they offer arbitration or will they be afraid of getting stuck with a ~$13M contract for next season that they dont have budget room for?

[Balta1701]

The other problem you have with Dunn is finding a position for him. He has said many times that he does not want to DH so it doesnt make sense to go trade for a guy who doesnt want to be here. He is not as good defensively at any position as the players we have now. The other question is if the Sox make the trade will they offer arbitration or will they be afraid of getting stuck with a ~$13M contract for next season that they dont have budget room for?

Unless he gets hurt, there's no reason to think he isn't eminently tradeable the moment he signs his arbitration deal, if he decides to go that route (there is also zero chance he'd accept arbitration unless he gets majorly hurt before the end of this year).

[Kyyle23]

why the hell do the Nationals want Edwin Jackson so bad? The dude has had one decent year and has been garbage the rest of the time..

 

I think it is a classic case of people judging his stuff over his results. The guy still can get it up there, and he threw a (imperfect) no hitter this year. There is always a team out there that thinks they can catch lightning in a bottle

[Balta1701]

I think it is a classic case of people judging his stuff over his results. The guy still can get it up there, and he threw a (imperfect) no hitter this year. There is always a team out there that thinks they can catch lightning in a bottle

There are also cases where teams think a guy would perform particularly well in their ballpark or working with their pitching coach, or that he'd be a particularly good match for the other guys on their staff.

[ptatc]

I'm sure somebody has tried to do a power analysis. Guys like Tom Tango and Mitchel Lichtman have been doing sabermetric research for decades. They have a whole book dedicated to things like bunts, clutch hitting, lineup order, hot and cold streaks, etc. If you dig deep enough, you might find something about power. I haven't come across it myself, but it might be out there.

 

Bill James is just some guy who is looked at as the guy who makes up all these stats nobody has an idea about, but he's not really one of the current innovators of sabermetrics. Asking him about something 10 years ago probably isn't going to give you the whole story.

 

Most of the research by Tango and Lichtman are descriptive and not predictive. They've done mostly retrospective studies. I was using Bill James as an example. Even the newest sabermetrics are based on the concept of"how can we determine (insert example) well this makes sense so let's try these factors." Things like WAR are very abstract. Who is the replacement? Is it the average player? Is it AL? Isit NL? We all know the leagues are different so how can anyone player be generalized to everyone. I'm not saying they aren't useful, I've always looked at numbers in baseball as useful. They are helpful for comparison. But when you are discussing stats and the use of them you really need to look at the reliability and validity and how you apply the terms.

 

Back to the original point most people on this site use the concept and term of sample size wrong and it is not a good argument when discussing the predictive value of a players performance.

Edited July 28, 2010 by ptatc

[Kyyle23]

There are also cases where teams think a guy would perform particularly well in their ballpark or working with their pitching coach, or that he'd be a particularly good match for the other guys on their staff.

 

yeah I agree with that. Washington doesnt have a really homer friendly park, and Edwin Jackson is prone to the longball. That staff would bring the heat, thats for sure(provided Strasburg isnt really hurt)

[harfman77]

Is Dunn even a Type A free agent in the AL? According to the lastest Elias rankings from MLB Trade Rumors the lowest type A in the AL has a ranking of 77.297 and Dunns ranking is 76.667? Just asking cause I think that is a factor as well.

[caulfield12]

Look at our entire offseason.

 

It was predicated on small(er), in some cases, season-ending, sample sizes for Pierre, Kotsay, Thome and Jermaine Dye (not to mention Vladimir Guerrero declining as well).

 

Of course, there are many other factors (the money Pods wanted, two year contract, Ozzie's personal bias towards Pierre) involved, but it probably wasn't the best way to go about filling the DH spot, that's for sure.

 

Andruw Jones has done exactly as predicted, following his statline from Texas very closely. Almost as predicatable as Alexei Ramirez getting off to slow starts every season.