nsacpi
Expects Yuge Games
average for left or average for all positions?
GDT 4/5 - Colon @ deGrom
- Thread starter Freshmaker
- Start date
average for left or average for all positions?
striker42
Well-known member
Ender in CF is more valuable than Ender in LF. That's not a real issue. The real issue is that defense isn't as big of a deal in LF as it is in CF. So the idea that an average defensive LFer gets a negative dWar is weird. If anything, I would think dWar would be limited in either direction just because of the lack of impact defense has in left compared to other positions.
Ender in LF would be just as valuable as Ender would be in CF. That's the whole point of the stat. To be able to compare across positions.
striker42
Well-known member
That makes zero sense.
nsacpi
Expects Yuge Games
The way I think about it is this. Imagine you have a team consisting of entirely average players for their positions. An average team is worth 80 wins. 40 of those from run creation (hitting and base running) and 40 of those from run prevention (pitching and defense).
The run prevention part breaks down roughly as follows: pitching part is worth about 30 wins and the defense part is worth about 10 wins.
You have to allocate those 10 defensive wins among the 9 positions. It has to add up to ten. If you force every position to be positive, you are restricting the range between the most valuable defensive positions (short and center) and the least valuable ones (left and first). To allow defensive WAR to truly reflect the range of value between short and left (and also have all the positions add up to 10 defensive wins) you have to allow the defensive value of an average player in left be below zero.
Agreed. Ender I far more valuable in CF bease he has the ability to make more plays.
How so? DRS (defensive runs saved), imo, is the best public defensive metric we currently have. It's what is used on bref for their version of WAR. DRS simply states how many runs you either saved or cost a team at a certain position when comparing your play to the rest of the league at that position.
In his limited career Ender has played much more Center than the corner spots but he does have some time there. Extrapolating his DRS across the 3 positions to 1000 innings gets him these numbers.
LF: +29
CF: +16
RF: +23
Now sample size issues do come into play here but it's obvious that Ender saves more runs in the corner outfield spots than he does in center. And it should be obvious as to why. It's not that his defensive ability has changed. He's just being compared to better overall defenders in center. If Ender is a 3 WAR CF he shouldn't become a 4+ WAR LF just by saving more runs there. That's where the positional adjustment comes in.
bravesfanMatt
Steve Harvey'd
LF.. which is his position.
I am saying you can't have it both ways.. You can't say X player is a horrible defender, yet he rates average to above average for his position. I personally don't care how bad Kemp would be in center. I only want to know how he does in LF.. If he is negative defender because he is in LF, then I think the stat is skewed, or more likely misrepresented. you see a lot of folks saying.. X player is a bad defender because he has -2 dwar.. but if the position is penalized, then he is not a bad defender for his position... Which is all I would want to know. I mean, why would I want to know how FF compares defensively to a 3rd baseman.
Then the people that say that aren't understanding the stat. Which does happen a lot. That's not the fault of the stat it's the fault of the person for either not understanding it or trying to mislead people.
The positional adjustment is mainly for total WAR calculations. Which is where you would want to know how FF compared to a 3B. If I'm looking at MVP voting for example and I'm deciding between FF and Kris Bryant for example. Say both were average for their position defensively. I would favor Bryant because 3rd is harder to play.
Again it depends what you are looking at and what you are wanting to know.
How many plays do you think a CF makes over a corner outfielder in a given year?
The only folks arguing against the defensive stats are the ones desperate to rate Kemp as a better overall player than he is.
The reason the numbers looks fishy now is because the season consists of 2 games. The overall value an average CFer provides a team on a day to day basis over the defensive value a LFer provides is being accounted for as a running total, even though there haven't been enough "good or bad" plays to justify it in the minds of folks who don't understand the stat.
If Kemp has played exactly average LF these 2 games, and Ender has played exactly average CF over the same 2 games, who is more valuable defensively? If they switched positions, Ender would play above average LF, and Kemp would play below average CF, and the numbers would generally be the same.
If you happen to think an average LFer is as valuable defensively to his team as an average CFer, well, there isn't really any point in discussing player valuation with you.
No argument I have seen has roots in anything other than ignorance. So the derpers will continue to derp along as usual. Derp derp!! Yay!! Kemp looks slimmer!! Derrrrrp!
The reason the numbers looks fishy now is because the season consists of 2 games. The overall value an average CFer provides a team on a day to day basis over the defensive value a LFer provides is being accounted for as a running total, even though there haven't been enough "good or bad" plays to justify it in the minds of folks who don't understand the stat.
If Kemp has played exactly average LF these 2 games, and Ender has played exactly average CF over the same 2 games, who is more valuable defensively? If they switched positions, Ender would play above average LF, and Kemp would play below average CF, and the numbers would generally be the same.
If you happen to think an average LFer is as valuable defensively to his team as an average CFer, well, there isn't really any point in discussing player valuation with you.
No argument I have seen has roots in anything other than ignorance. So the derpers will continue to derp along as usual. Derp derp!! Yay!! Kemp looks slimmer!! Derrrrrp!
Number of chances? I'd estimate approximately .5-1 more plays per game.
I think the point they are trying to make is that Kemp has been painted as a horrible LF'er because of his Dwar, when in fact he is actually average or above average amoung LF'ers.
All they are saying, is he should be rated by position instead of grouped with other positions.
Why would you rate a player with say a CF'er, when in fact he will NEVER play CF. LF'ers are usually average defenders, but above average offensive players. This is the NL's closet version of a DH. No one expects a LF'er to be stellar defensively, and if he is...he will be moved to right. It's really part of that position.
So, Dwar has WAY less value rating a LF'er , compared to other positions.
Kemp is bad compared to everyone. In fact, he is the worst defensive OFer not named Trumbo or Hanley Ramirez, neither of whom will be playing much (or any) OF this year...for a reason.
Again, someone that doesn't understand the stat.
And a DH gets an even larger defensive penalty than a LFer. So yet again, you don't know what you're talking about.
I'm honestly not sure what point they are making. But Kemp is not an average LF defensively. It is true that around -7 to -8 dWAR would make you an average corner outfielder for the position. The last two seasons he has been -24 and -21 on fangraphs.
bravesfanMatt
Steve Harvey'd
That is why I said misrepresented.. I wouldn't knock a stat for being a stat.. but I agree, people will use this incorrectly, either by ignorance or misguided agenda...
The point they are trying to make is that Kemp had a good game last night, he looks slimmer, and they will find any way they can to justify their opinion that having him on the roster is a positive.
In 2016 a CF had 0.61 more chances per game than a LF. .18 more chances than a RF. They do have more chances. It's why generally put your best defensive OF there and why they get a positional bonus for playing there.
When I look at defensive ability I like to look at just DRS or UZR. Simply the raw data when comparing a player to what other players are doing at their position. But you just have to know that positional adjustments are implemented into WAR and it can appear to skew things if you don't know what you are looking at.