Has Billy Beane lost his touch?

[Lex Luthor]

One year ago the Oakland A's had the best record in baseball at 59-36. Then on July 31st the A's traded Yoenis Cespedes for John Lester. The A's record before the trade 66-41. The A's record after the trade was 22-33.

In the offseason Billy Beane traded Josh Donaldson to the Toronto Blue Jays for third baseman Brett Lawrie, right-hander Kendall Graveman, left-hander Sean Nolin and minor league shortstop Franklin Barreto. Donaldson and Cespedes were the two best players on the A's roster. Now they're both gone and the team went from first to worst. Josh Donaldson is having an MVP season. Cespedes barely missed making the All Star team and is having a solid season. The A's have the worst record in the American League at 41-50.

It looks to me like Billy Beane has destroyed this team. Has Billy Beane totally ****ed up, or do Lawrie, Graveman, Nolin and Barreto have so much potential that the Donaldson trade could eventually work out in favor of the A's?

[Hootie]

I want DWAR to tell me who these best SHORTSTOPS are, the best FIRST BASEMAN are, the best CATCHERS, the best LEFT FIELDERS ... I don't want them to just tell me who the best defenders are and then litter the whole ****ing thing with SS, C and CF's. No ****ing shit.

['Hamas' Jenkins]

Originally Posted by Prison Bitch

Yes, there is a positional adjustment. In total, defense is 1/3 of positional player values. DH are given a "penalty" to begin as are 1B. Nonetheless it's not as important as offense and even SS who are terrible defensively can overcome it with a good bat.

No, it's not. You aren't even applying positional value correctly.

Positional value is based on the value of the defender at that position to an aggregate "average" defender. 1B are considered 12.5 runs worse than an average MLB defender, shortstops, 7.5 runs better.

Similarly, MLB 1B are given a positional adjustment which treats them as better than an average hitter. Thus, an average MLB hitter (100wRC) is still a net negative as a 1B.

Importance is relative and depends entirely on the player. A catcher is far more important defensively than offensively, as are shortstops, but to a lesser degree. Other positions, like LF or 1B are less important defensively and generally derive more value from their bat based upon the total number of chances they receive in a game.

['Hamas' Jenkins]

Originally Posted by Saul Good

To me, this smacks of someone essentially deciding what they want the outcome to be and building a formulathat will come to the predetermined conclusion. I can't put much stock in something done this way.

It's useful when comparing players within a single position. It's worthless when comparing them across positions. It's really not even meant to say how good a player is...just how much value they provided in their role. Some roles simply matter less, so the opportunities to provide value aren't equal.

It's not. Each position receives an average number of defensive chances per game. It's been found over tens of thousands of baseball games. Each outcome has a run value assigned to it, even down to the strike.

For example, imagine a pitch thrown on the outside corner on 0-0. If a catcher frames that pitch in a manner that it is called a strike, that is worth 0.08 runs to his team. If he frames a 3-1 pitch, that's worth .251 runs.

Overall, every successful frame is worth, on average, .14 runs.

From this and other metrics you can ascertain how many runs a catcher can save, and since we know how many runs equates to a win, you can assign a value.

Similarly, the ability of a CF to get to a ball in the gap or not is worth a certain number of runs on average. They also have a certain number of chances per game. However, other CFs, like Cain, have greater range, and their additional range allows them to save more runs, which adds to their value.

The number of actual runs saved by the fielder compared to the league average of runs saved by players making average plays with average range (or average frames) allows one to determine both the value of the player to those at his fellow position, but also to an aggregate average defender because past adjustments have been factored by looking at the play of players who have switched positions (within the same year) and by using the play of dozens of them, statisticians are able to ascertain a fair value for the relative difficulty of one position to another.

Is it perfect? No, and no one would argue that it is. But it's still an excellent metric.

['Hamas' Jenkins]

Originally Posted by Hootie 2.0

I want DWAR to tell me who these best SHORTSTOPS are, the best FIRST BASEMAN are, the best CATCHERS, the best LEFT FIELDERS ... I don't want them to just tell me who the best defenders are and then litter the whole ****ing thing with SS, C and CF's. No ****ing shit.

Then use DRS.

[Saul Good]

Originally Posted by Hootie 2.0

I want DWAR to tell me who these best SHORTSTOPS are, the best FIRST BASEMAN are, the best CATCHERS, the best LEFT FIELDERS ... I don't want them to just tell me who the best defenders are and then litter the whole ****ing thing with SS, C and CF's. No ****ing shit.

Sure...but that's not what DWAR does. That doesn't make it a bad stat. It just means it's the wrong tool for the job you want done. (Or maybe its fine as a tool, but you have to look at the number in the context of other SSs instead of just looking at the number itself.)

[Saul Good]

Originally Posted by 'Hamas' Jenkins

It's not. Each position receives an average number of defensive chances per game. It's been found over tens of thousands of baseball games. Each outcome has a run value assigned to it, even down to the strike.

For example, imagine a pitch thrown on the outside corner on 0-0. If a catcher frames that pitch in a manner that it is called a strike, that is worth 0.08 runs to his team. If he frames a 3-1 pitch, that's worth .251 runs.

Overall, every successful frame is worth, on average, .14 runs.

From this and other metrics you can ascertain how many runs a catcher can save, and since we know how many runs equates to a win, you can assign a value.

Similarly, the ability of a CF to get to a ball in the gap or not is worth a certain number of runs on average. They also have a certain number of chances per game. However, other CFs, like Cain, have greater range, and their additional range allows them to save more runs, which adds to their value.

The number of actual runs saved by the fielder compared to the league average of runs saved by players making average plays with average range (or average frames) allows one to determine both the value of the player to those at his fellow position, but also to an aggregate average defender because past adjustments have been factored by looking at the play of players who have switched positions (within the same year) and by using the play of dozens of them, statisticians are able to ascertain a fair value for the relative difficulty of one position to another.

Is it perfect? No, and no one would argue that it is. But it's still an excellent metric.

I kind of butchered my post. It's a good tool for determining value based on how a player is used and how they perform in that role. It's not a good tool for determining how good a player is (if that makes any sense). It's like having pay grades at a company. The janitor can have a degree in Economics from Duke, but he's still only worth $25k a year if all he's doing is emptying trash cans and scrubbing floors...even if he's great at it and is capable of much more. Make him the CFO, and that same guy becomes much more valuable to the company. Same guy, same skills...just applied differently.

[Prison Bitch]

Originally Posted by 'Hamas' Jenkins

No, it's not. You aren't even applying positional value correctly.

Positional value is based on the value of the defender at that position to an aggregate "average" defender. 1B are considered 12.5 runs worse than an average MLB defender, shortstops, 7.5 runs better.

Similarly, MLB 1B are given a positional adjustment which treats them as better than an average hitter. Thus, an average MLB hitter (100wRC) is still a net negative as a 1B.

Importance is relative and depends entirely on the player. A catcher is far more important defensively than offensively, as are shortstops, but to a lesser degree. Other positions, like LF or 1B are less important defensively and generally derive more value from their bat based upon the total number of chances they receive in a game.

On the bottom of each FG player profile is a value dashboard, showing offense + defense + base running + positional adj. As I stated above, DH/1B get a positional penalty.


I do not agree that a 1B gets any penalty on his wRC+ data (offense value). It's only adj for park and league. There is only the positional penalty at the end of the equation. If that's what you're saying I would agree.

[Prison Bitch]

Originally Posted by Hootie 2.0

dude. I get that. I do. I really, really, really do.

I just can't believe DWAR is going to try and calculate one of the worst defensive players at his position and then try and paint him as a positive player because he plays a tough position

that's just ****ing stupid

"positive player".


You need to define what you mean. Semien is 25th fWAR player at his position. Relative to league avg, he's most certainly negative. What he is "positive" relative to is a replacement player. Who would surely be worse.

[Hootie]

as much as I hate fangraphs, I guess I'll have to start using that website when I want to know how good of a season a defender is having because sorting by DWAR on ESPN is stupid as ****

I do, however, as someone who has watched almost every Royals game, have to wonder how in the **** Hosmer is the 13th best defensive 1st baseman in baseball?

Dude is crazy good over there.

I guess the only thing I can come up with is ... he's played too many innings, and for some ****ing idiotic reason every 1st baseman always has a negative DWAR because the people who created this metric are ****ing idiots

[Demonpenz]

DWAR figures in- .4 runs for every douche haircut and -.2 every time hosmers eyebrows look like wasp wings.

[duncan_idaho]

I get the positional adjustments and such.

What I don't get is what is going on with the infield metrics. How Eric Hosmer can be viewed as the 13th-best guy at 1B by Fangraphs just doesn't make sense to me.

I can buy a few guys (Adrian Gonzalez, for example) being ranked above Hosmer, and I know it's too early for the 2015 sample to produce a reliable defensive projection... but this is something that has been consistent throughout his career.

And it still doesn't track for me.

[Hootie]

Originally Posted by duncan_idaho

I get the positional adjustments and such.

What I don't get is what is going on with the infield metrics. How Eric Hosmer can be viewed as the 13th-best guy at 1B by Fangraphs just doesn't make sense to me.

I can buy a few guys (Adrian Gonzalez, for example) being ranked above Hosmer, and I know it's too early for the 2015 sample to produce a reliable defensive projection... but this is something that has been consistent throughout his career.

And it still doesn't track for me.

he must get negative marks for every ****ing ridiculous scoop he has at 1st base that makes guys like Esky and Moose look so good

I mean, good ****. I want to know who is playing better 1st base than Eric Hosmer in baseball and start watching those games so I can figure it out

[Saul Good]

Stats, particularly defensive stats, will always remain a data point in my eyes as opposed to the definitive metric in terms of evaluating players. Certain things will always be too subjective.

['Hamas' Jenkins]

Originally Posted by Saul Good

I kind of butchered my post. It's a good tool for determining value based on how a player is used and how they perform in that role. It's not a good tool for determining how good a player is (if that makes any sense). It's like having pay grades at a company. The janitor can have a degree in Economics from Duke, but he's still only worth $25k a year if all he's doing is emptying trash cans and scrubbing floors...even if he's great at it and is capable of much more. Make him the CFO, and that same guy becomes much more valuable to the company. Same guy, same skills...just applied differently.

All of the roles already have values assigned to them by the positional adjustment, so it's built in to their WAR calculation. The positions in baseball are more alike than they are different, so I don't see how the janitor/CFO comp really makes much sense.

The only thing that ultimately matters in baseball is run scoring and run prevention. If you can find a metric that properly evaluates a player's ability to do both, as both offensive and defensive metrics do, then you can clearly evaluate him and his value.

The positional adjustments aid in that evaluation. Given the cost of a win, and the value of replacement level, you are able to ascertain:

1) How many runs that player saves on defense
2) How many runs they aid on offense

Compared to what league average players do in both situations. The Def metric is just an additional tool to help someone analyze the fielding ability of a player at one position compared to another.

['Hamas' Jenkins]

Originally Posted by duncan_idaho

I get the positional adjustments and such.

What I don't get is what is going on with the infield metrics. How Eric Hosmer can be viewed as the 13th-best guy at 1B by Fangraphs just doesn't make sense to me.

I can buy a few guys (Adrian Gonzalez, for example) being ranked above Hosmer, and I know it's too early for the 2015 sample to produce a reliable defensive projection... but this is something that has been consistent throughout his career.

And it still doesn't track for me.

Defensive ratings require three years for an adequate sample size.