I didn't see this anywhere else, so I figured I would mention it. Baseball Prospectus has just introduced a new hitting metric called Deserved Runs Created plus (DRC+). Its supposed to be more descriptive and have better predictive power than wRC+ or OPS+. It is like those in that it has 100 equal to league average, but I have found several players where the difference between the it and the old ones is more than 10 points. Its also nice because on each player's page it not only has their DRC+ but also the standard deviation amount so you can tell how confident they are in the number and what kind of range of possibilities seem plausible. They have also folded this number into their WAR calculations, so we can probably expect their numbers to be different than the other two. 

 

https://www.baseballprospectus.com/drc-deserved-runs-created/

So by this metric the 7th best offensive force in baseball history (currently tied with Mike Trout) is Frank Thomas? 

30 minutes ago, Balta1701 said:

So by this metric the 7th best offensive force in baseball history (currently tied with Mike Trout) is Frank Thomas? 

Yeah, like I said, its better than the other ones haha

I also found it interesting how big the separation is between Ted Williams and everyone else. 

Edited December 4, 20187 yr by GenericUserName

https://legacy.baseballprospectus.com/card/104176/eloy-jimenez

The stat also seems to work for most minor league stat lines. Eloy posted a 150 DRC+ at AAA last year. 

I haven’t seen anything close to a decent summary of the differences yet, so I’m not sure how I feel about it yet. It sort of seems like they’re just using drastically different linear weight constants?

 

17 minutes ago, Eminor3rd said:

I haven’t seen anything close to a decent summary of the differences yet, so I’m not sure how I feel about it yet. It sort of seems like they’re just using drastically different linear weight constants?

 

https://www.baseballprospectus.com/news/article/45445/comparing-drc-ops-and-wrc/

2 hours ago, GenericUserName said:

I didn't see this anywhere else, so I figured I would mention it. Baseball Prospectus has just introduced a new hitting metric called Deserved Runs Created plus (DRC+). Its supposed to be more descriptive and have better predictive power than wRC+ or OPS+. It is like those in that it has 100 equal to league average, but I have found several players where the difference between the it and the old ones is more than 10 points. Its also nice because on each player's page it not only has their DRC+ but also the standard deviation amount so you can tell how confident they are in the number and what kind of range of possibilities seem plausible. They have also folded this number into their WAR calculations, so we can probably expect their numbers to be different than the other two. 

 

https://www.baseballprospectus.com/drc-deserved-runs-created/

Not a fan of this one. It's getting away from actual hitting results and adding in the subjective "expected" hitting results. I can't seem to find the formula used. Anyone find it?

2 hours ago, GenericUserName said:

https://www.baseballprospectus.com/news/article/45445/comparing-drc-ops-and-wrc/

Thanks, this article clarified it greatly. 

Essentially this is like the hitter equivalent to xFIP. It’s like wRC+, except it normalizes BABIP in the same way that xFIP is like FIP except that it normalizes homerun rate. It also appears to treat park effects differently, though it wasn’t clear to me exactly how. 

My quick is that this is definitely useful, but very much not a replacement for wRC+. This metric would have been really useful when we were arguing about Avisai Garcia’s one good year. Instead of looking at his wRC+ and just noticing that his BABIP was sky high, we could have looked to DRC+ as an estimate for what he “should have” been  given neutral “luck.”

I take issue with using it as a WAR component though. While BABIP is certainly heavily luck-dependent, it is not SOLELY luck dependent, so this metric is going to underrate guys that consistently run high BABIPs (like speedy slap hitters and line drive machines) and overrate TTO monsters (like Gallo or Trumbo who do nothing hit hit homers and this have low BABIPs). 

Also makes sense why it like Frank Thomas so much. 

Edited December 4, 20187 yr by Eminor3rd

4 minutes ago, Eminor3rd said:

My quick is that this is definitely useful, but very much not a replacement for wRC+. This metric would have been really useful when we were arguing about Avisai Garcia’s one good year. Instead of looking at his wRC+ and just noticing that his BABIP was sky high, we could have looked to DRC+ as an estimate for what he “should have” been  given neutral “luck.”

Avi's DRC+ last year was 118. For fun I mapped that directly to a 118 wRC+, and last year the only player with that score was Shin-Soo Choo with a line of

.264

.377

.434

They make a pretty compelling case for why DRC+ outperforms wRC+ at both predicting future results *and* comporting better with the past results. So far I've yet to see a compelling counterargument to DRC+ other than that you'll never be able to calculate it in your head. Much less sensitive to extreme events, changes in the run environment (like the changing ball), and has clever ways to adjust for quality of opponent to avoid over-weighting strange results.

I also really appreciate the fact they're publishing measures of uncertainty for DRC+, so you can see how much noise is built in. There's no easily calculated or principled way to do this for wRC+ or traditional statistics, which makes it hard to know how much to read into the number for each player.

13 hours ago, Eminor3rd said:

I haven’t seen anything close to a decent summary of the differences yet, so I’m not sure how I feel about it yet. It sort of seems like they’re just using drastically different linear weight constants?

 

Essentially they are regressing BIP luck based on some batted ball data. Basically a bit like FIP, if someone for example has a super low babip or a huge amount of doubles they will regress it to a more normal rate to make it more predictive.

10 hours ago, Eminor3rd said:

I take issue with using it as a WAR component though. While BABIP is certainly heavily luck-dependent, it is not SOLELY luck dependent, so this metric is going to underrate guys that consistently run high BABIPs (like speedy slap hitters and line drive machines) and overrate TTO monsters (like Gallo or Trumbo who do nothing hit hit homers and this have low BABIPs). 

 

This is exactly what I was going to post. While luck is a major factor in BABIP, oftentimes, better hitters just have better BABIPs (consistently, over multi-year periods). I think a more effective method would be to look at quality of contact, hard hit percentage, line drive percentage, relative to soft hit percentage, pop up percentage, etc. coming up with a function and finding a way to factor that in. Doing that might help normalize some of the problems you outlined such as overrating low BABIP guys like Gallo

I think it does look at some metrics (hard hit, gb/fb...) to find a deserved babip.

Narvaez was the highest ranked at 109, Abreu at 106 and Palka at 99.

Kevan Smith 97

Avi 91, Davidson 92

Castillo 88

Moncada/Rondon/Delmonico 87, Anderson 85

David Bote 79, Yolmer 78

Tilson 75, Leury 72

LaMarre 67, Engel 63

2 hours ago, caulfield12 said:

Narvaez was the highest ranked at 109, Abreu at 106 and Palka at 99.

Kevan Smith 97

Avi 91, Davidson 92

Castillo 88

Moncada/Rondon/Delmonico 87, Anderson 85

David Bote 79, Yolmer 78

Tilson 75, Leury 72

LaMarre 67, Engel 63

Gotta sneak that Cub in there ?

1 hour ago, Quinarvy said:

Gotta sneak that Cub in there ?

I was actually surprised Baez was so low...he still would have topped the Sox, but he was hyped like the second coming of sliced bread and a surefire MVP winner in July and August.  I get it, his defense and base-running are big factors, but he wasn’t anywhere close to the MLB leaders for this specific metric.

Edited December 5, 20187 yr by caulfield12

The upshot is by this metric, almost every single White Sox batter was overrated by popular alternatives. I wonder if this might have something to do with strength of schedule.