Orioles WAR ratings

[NelsonCruuuuuz]

Big surprises for me are m Chirinos, Nevin (less surprising but both soundly negstive) and Baker (higher than I thought ).

 

https://www.fangraphs.com/warleaders.aspx?season=2022&team=2&type=

[Filmstudy]

Santander's WAR is worse than I would expect for his OPS+, even factoring in for bad defense.

I'm not sure what methodology Fangraphs uses.  Do you happen to know the assumed replacement winning percentage?  Per BRef, Santader shows as a higher WAR (1.3) and a slightly above .500 player by WAA.  BRef uses a .294 replacement level for WAR which I don't see as a particularly useful measure for any baseball decision making.

[DrungoHazewood]

4 hours ago, Filmstudy said:

Santander's WAR is worse than I would expect for his OPS+, even factoring in for bad defense.

I'm not sure what methodology Fangraphs uses.  Do you happen to know the assumed replacement winning percentage?  Per BRef, Santader shows as a higher WAR (1.3) and a slightly above .500 player by WAA.  BRef uses a .294 replacement level for WAR which I don't see as a particularly useful measure for any baseball decision making.

Both Fangraphs and bb-ref lay out all the components.  For bb-ref it's in that second box labeled Player Value--Batting.

Santander is five runs above average in batting compared to all the batters in MLB.  Which makes sense, as he has a 113 OPS+, or that he's about 13% better than average.  Really good hitters are 130, 150 or higher.

He's a -1 baserunner taking into account both stealing, caught, and advancements on the bases.

He's 0 on double play frequency, right on average for his 6 GIDP in 341 PA.

He's a -1 fielder by DRS.

His positional adjustment for his combination of RF/LF/DH is -3, which makes sense.  Those are on the easier side of the defensive spectrum.

For 341 PAs the difference between average and replacement level (of .294, which both Fangraphs and bb-ref use) is 12 runs.

Add all that up and you get 5-1+0-1-3+12 = 12 runs above replacement.  In the '22 Orioles context a win is a touch over 10 runs, so that works out to 1.1 WAR.

Fangraphs would be similar.  The main difference for non-pitchers is they use Statcast OAA for the fielding component.  Which actually isn't a difference in this case, since he's a -1 defender by OAA, too.

.294 was chosen as replacement level because that works out to essentially the worst winning percentage a MLB team can be.  The 2018 Orioles had a .290 winning percentage in one of the worst years in modern baseball history.  It reasonable to say that a player who is a .294 winning percentage player is one who is kind of floating around AAA, the waiver wire, DFAs.  You can find .294 players and acquire them for free, more-or-less.  Every March when spring rosters are being culled there are .300 winning percentage players available for the taking.

So that's the zero value point in baseball.  You're measuring how much value a player has above talent that can usually or often be acquired for nothing.  You don't measure from zero, because the talent that gets you from zero to replacement level is also worth about nothing.  The Orioles just claimed Louis Head, a 32-year-old Marlins reliever who has a 7.23 ERA this year.  They didn't pay anything for him, he was available on waivers for nothing.  Louis Head is far better than any number of guys who're in A ball. Those A ball guys are .200 or .100 MLB players.  But there's no value in the difference between Head and them, except that you'd never claim the A ball guys but they're also not paying any more for Head.  He's the zero point.

You can also measure from average if you'd like.  Sometimes that's useful.  But you have to remember that an average MLB player is still a very good and valuable player who teams will pay substantial amounts of money for.  Jordan Lyles has a 92 ERA+, he's been a below-average player throughout his career, but because there's value in the distance from replacement level to below-average he's being paid $7M this year.

[Frobby]

4 hours ago, Filmstudy said:

Santander's WAR is worse than I would expect for his OPS+, even factoring in for bad defense.

I'm not sure what methodology Fangraphs uses.  Do you happen to know the assumed replacement winning percentage?  Per BRef, Santader shows as a higher WAR (1.3) and a slightly above .500 player by WAA.  BRef uses a .294 replacement level for WAR which I don't see as a particularly useful measure for any baseball decision making.

Several years ago, Fangraphs and BB-ref agreed to use the same assumed winning percentage.  And the reason they used .294 (47-48 wins) is that it is exceedingly rare for a major league team to win fewer games than that.  The 2018 O’s and 2019 Tigers won 47.  The 2003 Tigers won 43, the 1962 Mets won 40.   I think that’s about it for teams under 48 wins since the 162-game schedule was adopted.  

[Filmstudy]

7 hours ago, DrungoHazewood said:

Both Fangraphs and bb-ref lay out all the components.  For bb-ref it's in that second box labeled Player Value--Batting.

Santander is five runs above average in batting compared to all the batters in MLB.  Which makes sense, as he has a 113 OPS+, or that he's about 13% better than average.  Really good hitters are 130, 150 or higher.

He's a -1 baserunner taking into account both stealing, caught, and advancements on the bases.

He's 0 on double play frequency, right on average for his 6 GIDP in 341 PA.

He's a -1 fielder by DRS.

His positional adjustment for his combination of RF/LF/DH is -3, which makes sense.  Those are on the easier side of the defensive spectrum.

For 341 PAs the difference between average and replacement level (of .294, which both Fangraphs and bb-ref use) is 12 runs.

Add all that up and you get 5-1+0-1-3+12 = 12 runs above replacement.  In the '22 Orioles context a win is a touch over 10 runs, so that works out to 1.1 WAR.

Fangraphs would be similar.  The main difference for non-pitchers is they use Statcast OAA for the fielding component.  Which actually isn't a difference in this case, since he's a -1 defender by OAA, too.

.294 was chosen as replacement level because that works out to essentially the worst winning percentage a MLB team can be.  The 2018 Orioles had a .290 winning percentage in one of the worst years in modern baseball history.  It reasonable to say that a player who is a .294 winning percentage player is one who is kind of floating around AAA, the waiver wire, DFAs.  You can find .294 players and acquire them for free, more-or-less.  Every March when spring rosters are being culled there are .300 winning percentage players available for the taking.

So that's the zero value point in baseball.  You're measuring how much value a player has above talent that can usually or often be acquired for nothing.  You don't measure from zero, because the talent that gets you from zero to replacement level is also worth about nothing.  The Orioles just claimed Louis Head, a 32-year-old Marlins reliever who has a 7.23 ERA this year.  They didn't pay anything for him, he was available on waivers for nothing.  Louis Head is far better than any number of guys who're in A ball. Those A ball guys are .200 or .100 MLB players.  But there's no value in the difference between Head and them, except that you'd never claim the A ball guys but they're also not paying any more for Head.  He's the zero point.

You can also measure from average if you'd like.  Sometimes that's useful.  But you have to remember that an average MLB player is still a very good and valuable player who teams will pay substantial amounts of money for.  Jordan Lyles has a 92 ERA+, he's been a below-average player throughout his career, but because there's value in the distance from replacement level to below-average he's being paid $7M this year.

Good sample calculation for Santander and good to know they both use .294.

The issues I have with WAR as a statistic is that I don't think it provides value one can apply to managing towards a championship and the sub .500 base fogs transparency.  For example, Rutschman and Santander have essentially the same WAR, but dramatically different value based on the playing time in which it was accumulated.  I have not seen anyone use WAR per PA to make a point, but it would be closer to the mark.

In terms of using it to manage towards a championship 1) Your target needs to be far above the replacement level (WAA is better) 2) The replacement level isn't typically relevant, because contending teams typically have a spate of choices well north of there if they are willing to simply take on salary and if trading a player or picks, I want an understanding of the relationship between that and championships, not the replacement level.

In thinking back over O's history, I can only think of 1 (I'm sure you can come up with others) starter the Orioles acquired in a serious playoff run where they were trying to fill an abyss of a hole with a starter who could only look good by WAR.  That was Todd Cruz acquired for the 1983 stretch run.  Glenn Gulliver brought up in 1982 could be another such example.

One of the other reasons I dislike WAR is for the same reason I hated OPS as soon as it was combined as the most common single stat for batters...most fans don't understand the nature and seriousness of the flaws which can create a lot of wasted breath.

[DrungoHazewood]

45 minutes ago, Filmstudy said:

Good sample calculation for Santander and good to know they both use .294.

The issues I have with WAR as a statistic is that I don't think it provides value one can apply to managing towards a championship and the sub .500 base fogs transparency.  For example, Rutschman and Santander have essentially the same WAR, but dramatically different value based on the playing time in which it was accumulated.  I have not seen anyone use WAR per PA to make a point, but it would be closer to the mark.

In terms of using it to manage towards a championship 1) Your target needs to be far above the replacement level (WAA is better) 2) The replacement level isn't typically relevant, because contending teams typically have a spate of choices well north of there if they are willing to simply take on salary and if trading a player or picks, I want an understanding of the relationship between that and championships, not the replacement level.

In thinking back over O's history, I can only think of 1 (I'm sure you can come up with others) starter the Orioles acquired in a serious playoff run where they were trying to fill an abyss of a hole with a starter who could only look good by WAR.  That was Todd Cruz acquired for the 1983 stretch run.  Glenn Gulliver brought up in 1982 could be another such example.

One of the other reasons I dislike WAR is for the same reason I hated OPS as soon as it was combined as the most common single stat for batters...most fans don't understand the nature and seriousness of the flaws which can create a lot of wasted breath.

Championship caliber teams routinely use players who are below average but above replacement.  The Braves last year didn't really have a regular LFer, they had nine different players who were below average in 100+ PAs.  Perfect example was Dansby Swanson, who played almost every game, was a 1.9 rWAR/-0.4 WAA player. On the World Champions.

Tom Tango has a system that converts WAR into what he calls the Indies, which is a win-loss record instead of a single number.  The single number is very useful in a lot of cases, but for comparing player value 15-3 vs 15-13 means a lot more than just saying six wins vs. five.  That gives you playing time context.  Santander might be a 5-4 player, while Rutschman is a 4-1 player (numbers made up for this example).

People don't understand a lot of things.  I think WAR helps eliminate some of the BS arguments you'd get when people would haphazardly combine numbers.  They'd take a guy who hit .320 with 28 homers but a brick glove and terrible baserunning and pick and choose how to value the total player, usually on the basis of how much the like the guy.  WAR provides a consistent framework and data sources so you can't do that.  You see that in old MVP votes, where the guy with 2.6 WAR wins the trophy and the guy with 9.0 finishes 11th because there was nothing systematic about combining all the different aspects of player value.

[Roll Tide]

13 hours ago, Filmstudy said:

Santander's WAR is worse than I would expect for his OPS+, even factoring in for bad defense.

I'm not sure what methodology Fangraphs uses.  Do you happen to know the assumed replacement winning percentage?  Per BRef, Santader shows as a higher WAR (1.3) and a slightly above .500 player by WAA.  BRef uses a .294 replacement level for WAR which I don't see as a particularly useful measure for any baseball decision making.

Santander’s WAR according to baseball reference is 1.1.

[Oriole1940]

When I was growing  up   on baseball before others tried to  tell you what to think, the rule of thumb was that to be a winning, good team, you needed to beat up on the poor teams, and play  at least .500 ball against the good teams.  Worked out good, you had a winning record and that was that.  

 

 

[DrungoHazewood]

44 minutes ago, Oriole1940 said:

When I was growing  up   on baseball before others tried to  tell you what to think, the rule of thumb was that to be a winning, good team, you needed to beat up on the poor teams, and play  at least .500 ball against the good teams.  Worked out good, you had a winning record and that was that. 

Just curious, who's telling you what to think?

[Ohfan67]

11 hours ago, DrungoHazewood said:

Both Fangraphs and bb-ref lay out all the components.  For bb-ref it's in that second box labeled Player Value--Batting.

Santander is five runs above average in batting compared to all the batters in MLB.  Which makes sense, as he has a 113 OPS+, or that he's about 13% better than average.  Really good hitters are 130, 150 or higher.

He's a -1 baserunner taking into account both stealing, caught, and advancements on the bases.

He's 0 on double play frequency, right on average for his 6 GIDP in 341 PA.

He's a -1 fielder by DRS.

His positional adjustment for his combination of RF/LF/DH is -3, which makes sense.  Those are on the easier side of the defensive spectrum.

For 341 PAs the difference between average and replacement level (of .294, which both Fangraphs and bb-ref use) is 12 runs.

Add all that up and you get 5-1+0-1-3+12 = 12 runs above replacement.  In the '22 Orioles context a win is a touch over 10 runs, so that works out to 1.1 WAR.

Fangraphs would be similar.  The main difference for non-pitchers is they use Statcast OAA for the fielding component.  Which actually isn't a difference in this case, since he's a -1 defender by OAA, too.

.294 was chosen as replacement level because that works out to essentially the worst winning percentage a MLB team can be.  The 2018 Orioles had a .290 winning percentage in one of the worst years in modern baseball history.  It reasonable to say that a player who is a .294 winning percentage player is one who is kind of floating around AAA, the waiver wire, DFAs.  You can find .294 players and acquire them for free, more-or-less.  Every March when spring rosters are being culled there are .300 winning percentage players available for the taking.

So that's the zero value point in baseball.  You're measuring how much value a player has above talent that can usually or often be acquired for nothing.  You don't measure from zero, because the talent that gets you from zero to replacement level is also worth about nothing.  The Orioles just claimed Louis Head, a 32-year-old Marlins reliever who has a 7.23 ERA this year.  They didn't pay anything for him, he was available on waivers for nothing.  Louis Head is far better than any number of guys who're in A ball. Those A ball guys are .200 or .100 MLB players.  But there's no value in the difference between Head and them, except that you'd never claim the A ball guys but they're also not paying any more for Head.  He's the zero point.

You can also measure from average if you'd like.  Sometimes that's useful.  But you have to remember that an average MLB player is still a very good and valuable player who teams will pay substantial amounts of money for.  Jordan Lyles has a 92 ERA+, he's been a below-average player throughout his career, but because there's value in the distance from replacement level to below-average he's being paid $7M this year.

Ahhh. Nice to read a classic Drungo post. The baseball professor is in the house. 

[owknows]

1 hour ago, Oriole1940 said:

When I was growing  up   on baseball before others tried to  tell you what to think, the rule of thumb was that to be a winning, good team, you needed to beat up on the poor teams, and play  at least .500 ball against the good teams.  Worked out good, you had a winning record and that was that.  

 

 

“We can’t bust heads like we used to—but we have our ways. One trick is to tell them stories that don’t go anywhere like the time I caught the ferry over to Shelbyville. I needed a new heel for my shoe, so I decided to go to Morganville which is what they called Shelbyville in those days. So, I tied an onion to my belt which was the style at the time. Now, to take the ferry cost a nickel. And in those days, nickels had pictures of bumblebees on ‘em. ‘Give me five bees for a quarter,’ you’d say. Now, where were we? Oh, yeah! The important thing was that I had an onion on my belt which was the style at the time. They didn’t have white onions because of the war. The only thing you could get was those big yellow ones.”

[yark14]

2 hours ago, Oriole1940 said:

When I was growing  up   on baseball before others tried to  tell you what to think, the rule of thumb was that to be a winning, good team, you needed to beat up on the poor teams, and play  at least .500 ball against the good teams.  Worked out good, you had a winning record and that was that.  

 

 

Let me guess, you also walked to school uphill both ways? 

[oldfan]

I hate to think what the HOF will look like 20 years from now. Their already putting very good players in, it's not for the very good, it's for the no doubters. When you have to start digging down deep into these new metrics, that's a red flag. If your not a first ballot guy, your not a true Hall of Famer.

[Hallas]

3 minutes ago, oldfan said:

I hate to think what the HOF will look like 20 years from now. Their already putting very good players in, it's not for the very good, it's for the no doubters. When you have to start digging down deep into these new metrics, that's a red flag. If your not a first ballot guy, your not a true Hall of Famer.

With the exception of relievers (who i think need to be limited to the Lee Smiths and Mariano Riveras of the world) I think "very good over ~15 years, or phenomenal over ~10" is basically a collection of the best players in baseball.  While I think you get bonus points for having an MVP level peak, I think it's perfectly acceptable to make it into the hall as a long-duration stat stuffer.  When you're talking about the magnitude of stats that are being accumulated it still ends up being extremely valuable, skilled players in the hall.  We might have to revisit this policy if we start getting players who are league average and continue to perform as such over extremely long durations, like on the order of 25-30 years, but we're nowhere near that point.

[Hallas]

10 hours ago, Filmstudy said:

Good sample calculation for Santander and good to know they both use .294.

The issues I have with WAR as a statistic is that I don't think it provides value one can apply to managing towards a championship and the sub .500 base fogs transparency.  For example, Rutschman and Santander have essentially the same WAR, but dramatically different value based on the playing time in which it was accumulated.  I have not seen anyone use WAR per PA to make a point, but it would be closer to the mark.

In terms of using it to manage towards a championship 1) Your target needs to be far above the replacement level (WAA is better) 2) The replacement level isn't typically relevant, because contending teams typically have a spate of choices well north of there if they are willing to simply take on salary and if trading a player or picks, I want an understanding of the relationship between that and championships, not the replacement level.

In thinking back over O's history, I can only think of 1 (I'm sure you can come up with others) starter the Orioles acquired in a serious playoff run where they were trying to fill an abyss of a hole with a starter who could only look good by WAR.  That was Todd Cruz acquired for the 1983 stretch run.  Glenn Gulliver brought up in 1982 could be another such example.

One of the other reasons I dislike WAR is for the same reason I hated OPS as soon as it was combined as the most common single stat for batters...most fans don't understand the nature and seriousness of the flaws which can create a lot of wasted breath.

I think this is a perfectly acceptable shortcoming for a counting stat. Combining wOBA and any of the per-year fielding stats would get you what you're looking for.  It might be slightly unfortunate that you can't get holistic measures of contributions on any currently-available rate stat without calculating one yourself, but I think that's not the end of the world.  The issue only really comes up with mid-season rookie call-ups and players that spend a lot of time on the IL.