Let's talk about Adam Dunn destinations

[Eminor3rd]

What is the universe from which the weights are created? All of MLB history? The last 10 years? The last year? There's still some subjectivity there in deciding that part of it, but I agree that wOBA is better than most.

 

Typically the stats are adjusted every season. If I remember correctly from THE BOOK, the formula was created using historical data, but the probability weights can be altered to fit any sample size.

[2nd_city_saint787]

Agreed. Adam Dunn probably only has value to someone if the Sox pick up substantial money. A team with a really bad 1B or DH would, for example, probably be willing to pay an Adam Dunn $7m or so.

 

Mark Reynolds is a fair comparison, and he signed a 1 year 6 mill contract in the offseason...Dunn has been the more productive player both this and last year, so I think the 7-9m range is about right.

 

Once again, if the Sox plan to compete next year I'd just keep him, if not I'd chip in half his contract and trade him.

 

 

[Eminor3rd]

All stats are flawed by OPS is extremely flawed. I use it because it's simple, widely reported, and gives a base guideline. I also take into consideration the differences in value between two similar OPS's,

 

Taking just what guys do at the plate into consideration here - Adam Dunn has a .791 OPS, Jacoby Ellsbury .790. Logic says that Dunn should have created more runs in general - a microscopic measure, but more nonetheless - yet Ellsbury has been signficantly more valuable simply with a bat in his hands because Ellsbury has been on base about 40 more times than Dunn (don't have the exact number but it's good enough).

 

Right, Ellsbury is sitting on 113 wRC+ versus Dunn's 110, and the difference is in the run values of the types of events involved. If you jsut look at OPS, you'd have to argue about whether or not you think Dunn's power is more valuable than Ellsbury's on base ability, and you'd never agree because it would be subjective. But linear weights stats just tell you straight out, based on which of them produce, on average, more runs.

[iamshack]

Indeed. The Orioles are an example of a team that could get a ton of value from Adam Dunn if the Sox picked up some of that salary. The Rays could, too, but you;d have to assume that you;d need to pick up nearly all of the salary, and anyway they might still get nervous because of the whole Pat Burrell thing, lol.

Ok, so the question the Orioles should be asking themselves is "How much value would having Adam Dunn on our team for the remainder of this year provide us?". Well, that isn't a simple question. First, you say what is the production difference between what we have now and what Dunn should provide if he continues on his current pace. Then, you ask what is that production worth. Then you have to ask what is that production worth if it allows us to make the playoffs whereas we wouldn't have without him. Then what if his production allows us to win a playoff game, a playoff series, the ALCS, the WS, etc.

 

Then you ask yourself what is the worst case scenario.

 

These are very complex questions with no really easy answers.

 

Considering that Dunn has hit 38 home runs or more and knocked in 92 runs or more in 8 of the last 9 seasons, and the one outlier was not last year, and not representative of his production thus far this year, and that he is still only 33, I don't think his salary is really all that prohibitive.

[cabiness42]

Typically the stats are adjusted every season. If I remember correctly from THE BOOK, the formula was created using historical data, but the probability weights can be altered to fit any sample size.

 

OK, one formula I just looked up included reaching on error in the formula, but excluded intentional walks. I'm not getting that.

 

[Dick Allen]

Mark Reynolds is a fair comparison, and he signed a 1 year 6 mill contract in the offseason...Dunn has been the more productive player both this and last year, so I think the 7-9m range is about right.

 

Once again, if the Sox plan to compete next year I'd just keep him, if not I'd chip in half his contract and trade him.

That was getting Reynolds without getting anything back. I don't see where it does the Sox any good to pay Dunn $6-8 million or more to play for another team next year. Money shouldn't be an object with many large contracts off the books, and it's not like they have a couple of guys raring to go in the minors. They are better off keeping him.

[witesoxfan]

The Orioles can also maximize his value, much better than the Sox. The Sox count on him cleaning up and being the run producer, while the Orioles would hit him 6th or 7th and they'd be able to sit him against lefties (likely going with Pearce or Valencia or whatever righty bat they have on the bench).

 

There are other guys they can acquire too, but that matchup really makes a lot of sense for both teams.

[2nd_city_saint787]

That was getting Reynolds without getting anything back. I don't see where it does the Sox any good to pay Dunn $6-8 million or more to play for another team next year. Money shouldn't be an object with many large contracts off the books, and it's not like they have a couple of guys raring to go in the minors. They are better off keeping him.

 

 

If they don't plan on competing next season and can get a player who can contribute for the future if they chip in money then it makes a ton of sense. His value might actually be somewhat decent right now since he's been pretty hot for a good period of time.

 

 

Once again I personally think they should be competing next year, but for the sake of the thread I'm entertaining the idea of trading him.

[witesoxfan]

If they don't plan on competing next season and can get a player who can contribute for the future if they chip in money then it makes a ton of sense. His value might actually be somewhat decent right now since he's been pretty hot for a good period of time.

 

 

Once again I personally think they should be competing next year, but for the sake of the thread I'm entertaining the idea of trading him.

 

Competing or competitive? I think they can be a .500ish team next year, but like 90+ is a reach.

[Balta1701]

Right, Ellsbury is sitting on 113 wRC+ versus Dunn's 110, and the difference is in the run values of the types of events involved. If you jsut look at OPS, you'd have to argue about whether or not you think Dunn's power is more valuable than Ellsbury's on base ability, and you'd never agree because it would be subjective. But linear weights stats just tell you straight out, based on which of them produce, on average, more runs.

How big of a difference is 3 points in wRC+? Is that a small difference or a big difference? It seems very small.

[elrockinMT]

The rule of thumb is if Dunn were a free agent, would he get the contract he has now? With the answer being a pretty clear "no", you have to let him go.

 

 

He still has better offensive numbers than anyone on the team currently (not including average)

[witesoxfan]

He still has better offensive numbers than anyone on the team currently (not including average)

 

This is irrelevant because the team is not going to win anything of significance this year and will be selling pieces off. Wins and losses no longer matter.

Edited July 17, 2013 by witesoxfan

[Eminor3rd]

How big of a difference is 3 points in wRC+? Is that a small difference or a big difference? It seems very small.

 

3% of league average.

[Eminor3rd]

OK, one formula I just looked up included reaching on error in the formula, but excluded intentional walks. I'm not getting that.

 

I'm not aware of any formula that includes reaching on error. For what it's worth, I usually reference FanGraphs because they use Tango's formulas:

 

wOBA = (0.691×uBB + 0.722×HBP + 0.884×1B + 1.257×2B + 1.593×3B + 2.058×HR) / (AB + BB – IBB + SF + HBP)

 

Also, here are all of the historical component run values by year: http://www.fangraphs.com/guts.aspx?type=cn

[Balta1701]

3% of league average.

This number just doesn't mean anything to me. I don't know what "league average run production" is off the top of my head and I don't know how it scales.

[witesoxfan]

This number just doesn't mean anything to me. I don't know what "league average run production" is off the top of my head and I don't know how it scales.

 

The numbers will vary from year to year, which is when using wOBA (always measured as a straight number) will be nicer.

 

For comparison's sake, an average to slightly above average hitter was putting up .800 OPS's in the early 2000s. Now a days, an above average to good hitter is putting up .800 OPS's. In the raw sense, they are both an .800 OPS, but using OPs+, you see that an .800 OPS in 2000 is a 105 OPS+ while an .800 OPS is, say, 120, depending upon home parks. For instance, Jacoby Ellsbury is at .790 with an OPS+ of 113 while Kendrys Morales is at .800 with an OPS+ of 127. Morales is 14% higher because of the park he plays in.

[Jake]

wRC+ varies year by year. For instance, Dunn's wRC+ would look worse in the steroid era because the league in general was hitting so much more.

 

3% isn't a lot, but it's also not nothing. We're talking a few runs - which may not sound like much but ultimately could add up to a game of value.

[Eminor3rd]

This number just doesn't mean anything to me. I don't know what "league average run production" is off the top of my head and I don't know how it scales.

 

Ok, another way of looking at it is wRAA, which is essentially wRC without the scaling. It's measured in runs

 

Dunn = 7.1

Ells = 10.5

 

So the sum of Dunn's offensive contributions would create, on average, roughly 7 more runs than a league average player. Ellsbury is a touch over 10. To put THAT in context, historically, teams tend to get one win for every 10 runs they score ABOVE the amount of runs they give up. So, ignoring defense in this case, Ellsbury's contributions are worth about one additional win COMPARED to average.

 

Does that make sense?

Edited July 17, 2013 by Eminor3rd

[Jerksticks]

Plus the Orioles have to figure he will be useless in the late innings of playoffs. He's an automatic out to the majority of LOOGYs and good pitchers in general.

[TaylorStSox]

Plus the Orioles have to figure he will be useless in the late innings of playoffs. He's an automatic out to the majority of LOOGYs and good pitchers in general.

But if you have to go to your pen and use your LOOGY to get Dunn out, it's almost a positive contribution.

[witesoxfan]

But if you have to go to your pen and use your LOOGY to get Dunn out, it's almost a positive contribution.

 

Then you bring in said LOOGY, pinch hit with Steven Pearce, and boom goes the dynamite.

[chw42]

But at least for now, we're mentally calibrated on OPS. I know what a .900 OPS hitter is doing, I know what an .800 OPS hitter is doing, I know what a .700 hitter is doing, I know what a .600 OPS hitter was doing. When I quote those numbers, everyone knows roughly what that hitter is doing without me defining the standard. If you post RC+, great, but you have to calibrate me in the same post, I'm just not set on it yet.

 

Not exactly. A guy with a .900 OPS might have a .330 OBP but a .570 SLG. Or he might be .400/.500. There's a big difference between those two hitters.

 

I have to deal with OPS because that's all most websites can show me. If I'm more interested in the player, I go on FanGraphs.

 

I've also gotten to the point where I can glance at OBP and SLG and sort of guess the wOBA. I'm usually like 5-10 points off though.

Edited July 17, 2013 by chw42

[chw42]

Typically the stats are adjusted every season. If I remember correctly from THE BOOK, the formula was created using historical data, but the probability weights can be altered to fit any sample size.

 

They alter the weights every year.

 

But when your sample size is a entire league for the entire year, it's probably alright to adjust it every year.

[Jake]

Not exactly. A guy with a .900 OPS might have a .330 OBP but a .570 SLG. Or he might be .400/.500. There's a big difference between those two hitters.

 

I have to deal with OPS because that's all most websites can show me. So it's just a slight glance for me. If I'm more interested in the player, I go on FanGraphs.

 

Definitely. Most of my web searches are "x player fangraphs"

[Balta1701]

Ok, another way of looking at it is wRAA, which is essentially wRC without the scaling. It's measured in runs

 

Dunn = 7.1

Ells = 10.5

 

So the sum of Dunn's offensive contributions would create, on average, roughly 7 more runs than a league average player. Ellsbury is a touch over 10. To put THAT in context, historically, teams tend to get one win for every 10 runs they score ABOVE the amount of runs they give up. So, ignoring defense in this case, Ellsbury's contributions are worth about one additional win COMPARED to average.

 

Does that make sense?

And so Ellsbury's offense has been worth 1/3 of a win more than Dunn's offense so far this season?