WAR Thread

[Angelsjunky]

A thread for all things WAR. Nerd!

Anyhow, it was actually Clayton Kershaw who inspired this, not Trout. If you haven't noticed, Kershaw is off to a great start, with a 1.89 ERA through 6 starts, including 1.2 fWAR and a 9.71 K to 1.18 BB rate in 38 IP. Kershaw may be a bit under-appreciated these days, as he hasn't been among the very best pitchers since his peak in 2011-16, after which he dropped a level from best pitcher in the world to merely top 10 or so. But even then, since 2017 he's had a 2.68 ERA and averaged 4.9 WAR per 200 IP...not quite his 2011-16 level (2.06 ERA, 6.8 fWAR per 200 IP) but still among the better pitchers. What I find particularly impressive about "latter day Kershaw" is that he seems to overpower batters less, while still knowing how to keep the runs down...meaning, he's gotten smarter.

Meaning, Kershaw is under-appreciated if you over-emphasize WAR. He's not dominating in the same way he used to on pure stuff and peripherals, but he's still putting up stellar ERAs.

As far as career fWAR is concerned, with 74.5 he's now tied with Warren Spahn in just about exactly half the innings at 27th all-time. He's almost certainly going to pass Robin Roberts and Kevin Brown to finish the year at 25th, behind Justin Verlander. If he has another decent year next year, he'll be in the top 20, though will be hard-pressed to reach the top 10 (95.7 fWAR).

OK, Trout update. He's been pretty good this year, but clearly not peak level. He's at 1.5 fWAR through 29 games, good for 9th in the majors. But he hasn't really gotten hot with the bat yet, so expect a spike at some point and with a chance at finishing 1st for the first time since 2019.

Trout passed Joe DiMaggio earlier in the year and is at 83.6 for his career, 34th all-time. He's going to pass a bunch of guys this year:

25. Cap Anson 91.2

26. Al Kaline 88.9

27. Albert Pujols 88.8

28. Wade Boggs 88.3

29. Roger Connor 86.2

30. George Brett 84.6

31. George Davis 84.6

32. Chipper Jones 84.6

33. Adrian Beltre 83.8

34. Mike Trout 83.6

35. Joe DiMaggio 82.6

 

As you can see, Trout needs 5.3 WAR to pass Albert Pujols, which he should do--barring injury--sometime in the second half. He has a chance at becoming the 26th hitter in major league history to reach 90 WAR by year's end.

[jsnpritchett]

Side note, but the gap in fWAR and bWAR for Spahn is one of the largest you'll find: he's at exactly 100 bWAR vs. the 75.5 fWAR you cited for him.  In his case, it sure seems like bWAR would more accurately value his career.

[Angelsjunky]

9 minutes ago, jsnpritchett said:

Side note, but the gap in fWAR and bWAR for Spahn is one of the largest you'll find: he's at exactly 100 bWAR vs. the 75.5 fWAR you cited for him.  In his case, it sure seems like bWAR would more accurately value his career.

I honestly wish the warnerds would get together and standardize. I know that bWAR has some wonky numbers (e.g. Aaron Nola at 9.7 in 2018), but I also feel somewhat in resonance with Bill James' view that such numbers should better reflect what a player actually does (e.g. prevent runs) and not what they "should" have done (e.g. FIP-based WAR).

BTW, Fangraphs' RA9-WAR is even higher for Spahn, at 109.2. That puts him at 9th all-time. Kershaw is at 83.9, or 29th.

[happybat4]

Zack Greinke and Kershaw have similar Bwar. Kershaw is at 77.8 BWAR and Greinke is at 76.9. Lots of people seem to think Greinke isn't a lock for the hall of fame. They forget how good he was.

[Inside Pitch]

2 hours ago, jsnpritchett said:

Side note, but the gap in fWAR and bWAR for Spahn is one of the largest you'll find: he's at exactly 100 bWAR vs. the 75.5 fWAR you cited for him.  In his case, it sure seems like bWAR would more accurately value his career.

The biggest discrepancy is in how the two systems gauge their pitcher WAR. Fangraphs uses FIP, BBRef uses RA9 -- if you look at fWAR's RA9 WAR they actually rank Spahn more highly than bWAR while still accounting for FIP.

Spawn's RA9 fWAR is 109.2.  He was legitimately great.

https://www.fangraphs.com/leaders.aspx?pos=all&stats=pit&lg=all&qual=y&type=6&season=2023&month=0&season1=1871&ind=0&team=0&rost=0&age=0&filter=&players=0&startdate=1871-01-01&enddate=2023-12-31&sort=3,d

Many people believe fWAR's RA9WAR is the best version of pitcher WAR -- It certainly paints a better picture of Jered Weaver who always got dinged by his FIP data.

2 hours ago, Angelsjunky said:

but I also feel somewhat in resonance with Bill James' view that such numbers should better reflect what a player actually does (e.g. prevent runs) and not what they "should" have done (e.g. FIP-based WAR).

BTW, Fangraphs' RA9-WAR is even higher for Spahn, at 109.2. That puts him at 9th all-time. Kershaw is at 83.9, or 29th.

The entire point of using FIP is to try to normalize defense.  It's not aiming to tell you what a player SHOULD have done, it's an attempt to grade a player with all things being equal so that you're getting a better measure of what HE did based on the things HE had control over which is why it's a good choice to use it in a model attempting to weigh players evenly.

Bill James entire belief system was based on finding the league average and using that as the starting point.

 

[AngelsLakersFan]

2 hours ago, Inside Pitch said:

The biggest discrepancy is in how the two systems gauge their pitcher WAR. Fangraphs uses FIP, BBRef uses RA9 -- if you look at fWAR's RA9 WAR they actually rank Spahn more highly than bWAR while still accounting for FIP.

Spawn's RA9 fWAR is 109.2.  He was legitimately great.

https://www.fangraphs.com/leaders.aspx?pos=all&stats=pit&lg=all&qual=y&type=6&season=2023&month=0&season1=1871&ind=0&team=0&rost=0&age=0&filter=&players=0&startdate=1871-01-01&enddate=2023-12-31&sort=3,d

Many people believe fWAR's RA9WAR is the best version of pitcher WAR -- It certainly paints a better picture of Jered Weaver who always got dinged by his FIP data.

The entire point of using FIP is to try to normalize defense.  It's not aiming to tell you what a player SHOULD have done, it's an attempt to grade a player with all things being equal so that you're getting a better measure of what HE did based on the things HE had control over which is why it's a good choice to use it in a model attempting to weigh players evenly.

Bill James entire belief system was based on finding the league average and using that as the starting point.

 

I suspect the folks over at bb-ref don't even know how their version of war is calculated.

[Angelsjunky]

14 hours ago, Inside Pitch said:

 

The entire point of using FIP is to try to normalize defense.  It's not aiming to tell you what a player SHOULD have done, it's an attempt to grade a player with all things being equal so that you're getting a better measure of what HE did based on the things HE had control over which is why it's a good choice to use it in a model attempting to weigh players evenly.

Bill James entire belief system was based on finding the league average and using that as the starting point.

 

Yeah, FIP has its uses, I just dislike it as a totalizing stat of a pitcher's ability. You cited Jered Weaver as an example. There are pitchers who, over time, out-perform their FIP. When it is consistent enough, to me it speaks of an ability to pitch to context, and perhaps even "pitching IQ." Tom Glavine is another example. You also have the opposite: Guys like Javier Vazquez with great peripherals, but highish ERAs.

But yeah, I like and use RA9-WAR. I just wish that Fangraphs included it on their main stat charts next to WAR. They also have a 50/50 option in some places.

 

[ten ocho recon scout]

Like clockwork, Kershaw got lit up by Tatis and the Padres last night 

AJ, I'm glad you finally used your powers for good instead of evil.

[itsKnoppUitsme]

19 hours ago, happybat4 said:

Zack Greinke and Kershaw have similar Bwar. Kershaw is at 77.8 BWAR and Greinke is at 76.9. Lots of people seem to think Greinke isn't a lock for the hall of fame. They forget how good he was.

I only remember that as an Angel, he was a complete disappointment!

[Angelsjunky]

33 minutes ago, ten ocho recon scout said:

Like clockwork, Kershaw got lit up by Tatis and the Padres last night 

AJ, I'm glad you finally used your powers for good instead of evil.

I'm all over the place, I guess. I do like Kershaw, though.

[Angelsjunky]

19 hours ago, happybat4 said:

Zack Greinke and Kershaw have similar Bwar. Kershaw is at 77.8 BWAR and Greinke is at 76.9. Lots of people seem to think Greinke isn't a lock for the hall of fame. They forget how good he was.

Its worth noting that Greinke has pitched 660 more innings. But yeah, I see Greinke as a lock. He's sort of in the category of guys like Glavine and Mussina, imo. Though unlike those too, he's actually had a couple years in which he was probably the best pitcher in baseball. All three were so consistent over a long period of time.

[Inside Pitch]

5 hours ago, Angelsjunky said:

Yeah, FIP has its uses, I just dislike it as a totalizing stat of a pitcher's ability.

I don't think anyone should be using it that way, not sure I know any site that does.   I love FIP as a tool to unlock other stats or rather give them greater context.  Personally, I think it's best used when looking at team stats or league wide, it's one of those "the bigger the sample the clearer the picture is "stats.  When looking at individual players I always cross reference FIP with BaBip and the /9 rates.

The problem with FIP, like most stats is that people don't always use it correctly.

5 hours ago, Angelsjunky said:

You cited Jered Weaver as an example. There are pitchers who, over time, out-perform their FIP. When it is consistent enough, to me it speaks of an ability to pitch to context, and perhaps even "pitching IQ." Tom Glavine is another example.

This is where batted ball data and statcast information in particular has made it a lot easier to determine guys we are inducing soft contact .vs guys who are just getting lucky, which BaBip helps identify.  Jaime Barria is very much starting to look like a FIP outlier almost entirely due to what appears to be a real ability to induce soft contact.   He's also seen his average FB velocity rise 2-3 MPH since first coming up.

5 hours ago, Angelsjunky said:

You also have the opposite: Guys like Javier Vazquez with great peripherals, but highish ERAs.

Not sure I follow, Vasquez's FIP tended to be lower than his ERA for most of his career -- his big issue was that he would have seasons where he gave up HRs. This is yet another area where statcast data would have been super helpful.  xFIP attempts to normalize HR rates and I'd see people use it to try to argue it was a better measure of his ability but I've always believed that to be a much weaker stat.  There is no question GB and FB pitchers exist.  Plus the ballpark data always tended to jive with his HR issues.

Vasquez is a nice name to bring up, he was IMO massively underrated -- he had his blips along the way, but he had a solid decade where he was essentially a 5 fWAR pitcher.

[Angelsjunky]

31 minutes ago, Inside Pitch said:

I don't think anyone should be using it that way, not sure I know any site that does.   I love FIP as a tool to unlock other stats or rather give them greater context.  Personally, I think it's best used when looking at team stats or league wide, it's one of those "the bigger the sample the clearer the picture is "stats.  When looking at individual players I always cross reference FIP with BaBip and the /9 rates.

The problem with FIP, like most stats is that people don't always use it correctly.

Well that's just it - and is true of so many things, no? FIP is central to fWAR, which I think is used as a totalizing stat, which is a misuse, as you say.

EDIT: Meaning, sites don't use it incorrectly, but people - media, fans, etc - do, because they generally don't know what it represents and WAR has become so ubiquitous as a catchall stat.

Anyhow, it would be fun to do a search for the biggest outliers - in bWAR and fWAR, fWAR and RA9-WAR, etc. Ryan Doumit comes to mind, as far as bWAR vs fWAR, and Aaron Nola.

31 minutes ago, Inside Pitch said:

This is where batted ball data and statcast information in particular has made it a lot easier to determine guys we are inducing soft contact .vs guys who are just getting lucky, which BaBip helps identify.  Jaime Barria is very much starting to look like a FIP outlier almost entirely due to what appears to be a real ability to induce soft contact.   He's also seen his average FB velocity rise 2-3 MPH since first coming up.

Barria is sneaky valuable. He's the type of guy that we'll miss when he's gone, sort of like the pitching version of Orlando Palmeiro.

31 minutes ago, Inside Pitch said:

Not sure I follow, Vasquez's FIP tended to be lower than his ERA for most of his career -- his big issue was that he would have seasons where he gave up HRs. This is yet another area where statcast data would have been super helpful.  xFIP attempts to normalize HR rates and I'd see people use it to try to argue it was a better measure of his ability but I've always believed that to be a much weaker stat.  There is no question GB and FB pitchers exist.  Plus the ballpark data always tended to jive with his HR issues.

Vasquez is a nice name to bring up, he was IMO massively underrated -- he had his blips along the way, but he had a solid decade where he was essentially a 5 fWAR pitcher.

Yeah. I remember him well as someone I would go after in my peripherals-heavy fantasy baseball league towards the end of my run (about 15 years ago now...jeez), because his K/BB rates were so good. Anyhow, I used him as an example of a guy whose stuff was better than his ERA implied - and FIP represented this. Meaning, an example of a guy whose FIP was better than his ERA, in contrast to Weaver types.

Edited May 6, 2023 by Angelsjunky

[Inside Pitch]

2 hours ago, Angelsjunky said:

Its worth noting that Greinke has pitched 660 more innings. But yeah, I see Greinke as a lock. He's sort of in the category of guys like Glavine and Mussina, imo. Though unlike those too, he's actually had a couple years in which he was probably the best pitcher in baseball. All three were so consistent over a long period of time.

Mussina was a significantly better pitcher than either Glavine or Greinke, but yeah, he was never that truly dominate force that the others were on occasion.  

Dude was just consistently very good.   You look at the all time leaderboard for pitcher WAR and the only guy to compile more WAR while having pitched the same or less innings is Pedro.  Like Schilling, the dude was just great at limited HRs and BBs.  

https://www.fangraphs.com/leaders.aspx?pos=all&stats=pit&lg=all&qual=y&type=8&season=2023&month=0&season1=1901&ind=0&team=0&rost=0&age=0&filter=&players=0&startdate=1901-01-01&enddate=2023-12-31

[Inside Pitch]

18 minutes ago, Angelsjunky said:

 

EDIT: Meaning, sites don't use it incorrectly, but people - media, fans, etc - do, because they generally don't know what it represents and WAR has become so ubiquitous as a catchall stat.

 

I may end up deleting the previous response because I think this is really the important part of what you're saying.  Fans do really horrible jobs of using stats -- we are in total agreement.   But I do think it's inclusion is pitcher WAR makes total sense.  Like I said -- I wasn't sure what you were trying to say.

[AngelsLakersFan]

3 minutes ago, Inside Pitch said:

It's possible I'm not understanding what you mean by totalizing.

It's being used to try to establish a baseline.  You quoted Bill James, so again...  I remind you his entire belief system was based on finding the league average and moving on from there.   

If you're looking at pitchers league wide, any attempt to normalize the defensive end of the pitching equation seems pretty pivotal.  Only thing I can think of that would be more impactful is park tendencies..  So I don't see how it's possibly being misused as part of a greater formula.

The biggest argument one could mount against using FIP when talking about pitcher WAR IMO is the messy nature of defensive metrics in general.

 

Well I think the issue is that the fWar formula is disconnected from the actual runs scored / run allowed numbers. I don't have a problem with this, and if I remember correctly, this may have been an issue with the original win shares calculation.

Obviously you can create WAR equivalents at different levels, the win level, the run level, the play outcome level or the batted ball level. It really just depends on the goal of the metric. fWar is pretty much entirely done at the play outcome level, while bWar is a mix of run/play outcome which is why it is a fundamentally poor stat.

[Inside Pitch]

2 minutes ago, AngelsLakersFan said:

Well I think the issue is that the fWar formula is disconnected from the actual runs scored / run allowed numbers. I don't have a problem with this, and if I remember correctly, this may have been an issue with the original win shares calculation.

Obviously you can create WAR equivalents at different levels, the win level, the run level, the play outcome level or the batted ball level. It really just depends on the goal of the metric. fWar is pretty much entirely done at the play outcome level, while bWar is a mix of run/play outcome which is why it is a fundamentally poor stat.

I bought and memorized that damn Win Shares book only to have it fade away.  Still kind of bitter, lol...

[AngelsLakersFan]

2 minutes ago, Inside Pitch said:

I bought and memorized that damn Win Shares book only to have it fade away.  Still kind of bitter, lol...

It was conceptually ground breaking but the stat never caught on because no one really grasps what an 18 win share season means.

[Inside Pitch]

22 minutes ago, AngelsLakersFan said:

It was conceptually ground breaking but the stat never caught on because no one really grasps what an 18 win share season means.

There were so many moving parts and people just weren't ready.  And kind of like we have been talking about here, there was so much work put into finding those baselines that I think it was too much for most people.

[Angelsjunky]

19 hours ago, Inside Pitch said:

Mussina was a significantly better pitcher than either Glavine or Greinke, but yeah, he was never that truly dominate force that the others were on occasion.  

Dude was just consistently very good.   You look at the all time leaderboard for pitcher WAR and the only guy to compile more WAR while having pitched the same or less innings is Pedro.  Like Schilling, the dude was just great at limited HRs and BBs.  

https://www.fangraphs.com/leaders.aspx?pos=all&stats=pit&lg=all&qual=y&type=8&season=2023&month=0&season1=1901&ind=0&team=0&rost=0&age=0&filter=&players=0&startdate=1901-01-01&enddate=2023-12-31

I'll have to take a look at the numbers, as I'm mostly going by memory, which holds that Glavine was a bit overrated (not to mention the generous Atlanta Braves rotation strike zone) and Mussina under the radar. And of course Mussina was of an era that led to an un-sexy career ERA of 3.68. Andy Pettitte is another guy - 3.85 ERA but 68.2 WAR in 3316 IP. Not as good as Mussina, clearly, but consistently good to very good for a long time.

But yeah, so consistent. Those guys are so rare these days. 

Edited May 7, 2023 by Angelsjunky

[Chuck]

This is a great thread, @Angelsjunky. Submitted to the Blog! 

[Chuck]

On 5/5/2023 at 5:02 PM, Angelsjunky said:

A thread for all things WAR. Nerd!

Anyhow, it was actually Clayton Kershaw who inspired this, not Trout. If you haven't noticed, Kershaw is off to a great start, with a 1.89 ERA through 6 starts, including 1.2 fWAR and a 9.71 K to 1.18 BB rate in 38 IP. Kershaw may be a bit under-appreciated these days, as he hasn't been among the very best pitchers since his peak in 2011-16, after which he dropped a level from best pitcher in the world to merely top 10 or so. But even then, since 2017 he's had a 2.68 ERA and averaged 4.9 WAR per 200 IP...not quite his 2011-16 level (2.06 ERA, 6.8 fWAR per 200 IP) but still among the better pitchers. What I find particularly impressive about "latter day Kershaw" is that he seems to overpower batters less, while still knowing how to keep the runs down...meaning, he's gotten smarter.

Meaning, Kershaw is under-appreciated if you over-emphasize WAR. He's not dominating in the same way he used to on pure stuff and peripherals, but he's still putting up stellar ERAs.

As far as career fWAR is concerned, with 74.5 he's now tied with Warren Spahn in just about exactly half the innings at 27th all-time. He's almost certainly going to pass Robin Roberts and Kevin Brown to finish the year at 25th, behind Justin Verlander. If he has another decent year next year, he'll be in the top 20, though will be hard-pressed to reach the top 10 (95.7 fWAR).

OK, Trout update. He's been pretty good this year, but clearly not peak level. He's at 1.5 fWAR through 29 games, good for 9th in the majors. But he hasn't really gotten hot with the bat yet, so expect a spike at some point and with a chance at finishing 1st for the first time since 2019.

Trout passed Joe DiMaggio earlier in the year and is at 83.6 for his career, 34th all-time. He's going to pass a bunch of guys this year:

25. Cap Anson 91.2

26. Al Kaline 88.9

27. Albert Pujols 88.8

28. Wade Boggs 88.3

29. Roger Connor 86.2

30. George Brett 84.6

31. George Davis 84.6

32. Chipper Jones 84.6

33. Adrian Beltre 83.8

34. Mike Trout 83.6

35. Joe DiMaggio 82.6

 

As you can see, Trout needs 5.3 WAR to pass Albert Pujols, which he should do--barring injury--sometime in the second half. He has a chance at becoming the 26th hitter in major league history to reach 90 WAR by year's end.

@Angelsjunky,explain again why you're using fWAR for their respective careers and not bWAR? 

[Angelsjunky]

5 hours ago, Chuckster70 said:

@Angelsjunky,explain again why you're using fWAR for their respective careers and not bWAR? 

I prefer fWAR, at least for hitters - or at least am just used to it. They mostly differ on defense, though. I haven't decided which version (or Fangraphs RA9-WAR) is best, because @Inside Pitch won't tell me.

[stormngt]

On 5/6/2023 at 11:01 AM, itsKnoppUitsme said:

I only remember that as an Angel, he was a complete disappointment!

How was he a disappointment?

[Blarg]

48 minutes ago, stormngt said:

How was he a disappointment?

He pitched to a very pedestrian 3.53 ERA and 108 ERA+ for the Angels. The following three seasons for the Dodgers he was a combined 2.30 era and 156 ERA+.