|
||||
|
09-08-2019, 05:36 AM | #21 |
OOTP Developer
Join Date: Jun 2009
Location: Here and there
Posts: 14,145
|
WAR is useful for whatever you feel it's useful for.
To me, right now, it's the best single metric we have to determine a player's value. Nothing else combines all the other aspects of a player's value together, although if someone creates one I'm all ears. It's complicated, yes, but it has to be, if you want to "properly" compare players.the vast majority of the adjustments are because it's whole purpose is to calculate things relative to a mythical replacement player, and done so that the league as a whole has an essentially fixed total WAR. That being said, it should never be used as the only metric. But I treat it kind of like na initial sort. So WAR says that Verlander, Greinke, and Kershaw are all slightly under the average mark for HOF starters (by JAWS, which is a metric based on WAR). That's not saying they're not HOF material - in fact, it's the opposite. They're close enough on WAR that it's worth digging into the other values to see their exact qualifications. Of course, this brings up a whole other issue - namely, one thing I will agree with the "WAR-deniers" if you might call them that, that using WAR to compare pitchers of different eras is entirely wrong. Verlander is considered a modern workhorse and he's only going to barely reach 220 IP this year. It's patently unfair to compare him to an era where SP used to start 40 games a year, and where teams didn't have an 8-man pen to work with. But to me, that's more something that probably needs fixing in the JAWS system which is all about comparing players of different eras, rather than any sort of flaw with WAR. |
09-08-2019, 01:38 PM | #22 | |
Banned
Join Date: Aug 2019
Location: Ban land in 3...2...
Posts: 2,943
|
Quote:
((AB*(H+BB+HBP))+((H+2B+(2*3B)+(3*HR))*(AB+BB+SF+H BP))/((AB*(AB+BB+SF+HBP)) I don't really think all the "I can calculate it by hand" people are really calculating it by hand. Or ever have. I think they are adding OPS and SLG Which is kinda like making toast after someone has harvested the wheat, baked the bread, made the toaster, and wired the house for you. Similarly, WAR is just BR+FR+BSR+PR+Rep+PosAdj+League All if those concepts are pretty easy to understand for a casual baseball fan. Even if the exact formulas aren't. |
|
09-08-2019, 01:49 PM | #23 |
Hall Of Famer
Join Date: Apr 2009
Location: United States
Posts: 10,456
|
Good comments, Matt. An interesting situation arose this morning in my OOTPB game concerning WAR and I immediately thought of this thread.
I routinely "Ask Computer" for its suggestions as to major awards. Focus on Brendan Estave: What? I don't recall him having that great a year: Indeed, there were several other offensive standouts in the National League this year: Particularly this guy, who is my MVP choice: What is the point of all this? The game chose Estave on the basis of WAR. That's not to criticize it; I am sure that, given OOTPD's formula for WAR, it is accurate and he did have the highest WAR. But he did not have an MVP season, IMO. Thanks to the link to FanGraph's definition of WAR, I was reminded that WAR is based on other aspects of performance, including baserunning and defensive skill. Look again at Estave's season stats. He had a decent OPS (it's 1906 in my game) and he led the league in PA and runs. His stolen bases were less than half of the league leader, but they counted. Moreover, he's a Golden Glove winner again this year. I believe that it's because he was decent-to-good in a bunch of things that accounts for his high WAR. He would have won the MVP if I had not intervened. Or should have I let him win it? I look at my MVP award as primarily hitting; indeed, I call it the "Batting Excellence Award." Yet, based on OOTPD's definition of WAR, Estave was the best all-around player in the league. Leave aside the arcane and variable nature of WAR. Maybe a third, more subtle criticism of WAR arises: It's too scientific. It may be technically correct but intuitively wrong. Now, you may say, MVP is "best all-around player" by definition. Still, would you give the MVP award to Estave with his .274 batting average because he also was a good baserunner and fielder? Or would you award Pennypacker, who batted .315 and had a league-leading OPS of .821?
__________________
- Bru |
09-08-2019, 02:11 PM | #24 |
Hall Of Famer
Join Date: Apr 2009
Location: United States
Posts: 10,456
|
Wow. Before I move on, I want to point out that Pennypacker does not even appear on the list of award candidates. Leading the league in OBP and OPS, and he has a WAR of 1.5! All of these guys had a better season than Pennypacker:
No doubt the calculations would check out. Wrong results in actuality, though. EDIT: No, I was wrong; that's Pennypacker juuust peeking his head above the edge of the window on the bottom right. So alright, he was in consideration at least. But find Meraz and Dittman, the game's number 2 and 3 MVP candidates. They had better stats than Estave! And here is Pennypacker, who has a better slash line and wRC+ (I'll look that up in a minute ) than all three of them but the lowest WAR:
__________________
- Bru Last edited by Déjà Bru; 09-08-2019 at 02:27 PM. |
09-08-2019, 02:43 PM | #25 |
Hall Of Famer
Join Date: Apr 2009
Location: United States
Posts: 10,456
|
That really was a timely and cogent example, I must say. All the more reason why I really cannot put much stock and faith in WAR, I'm sorry to say.
Technically correct, I will grant you. If I had the time and energy to sit down and follow all of those calculations (and accept some subjective assessments), then I could grasp WAR, sure. But taken altogether; the arcane nature, the varying formulae, the overly scientific approach that can leave behind intuition and common sense; all of these prevent me from ever warming up to WAR. So, cephasjames, that's my last word on the matter. My advice is, don't knock yourself out trying to understand it. Accept it or not and move on.
__________________
- Bru |
09-08-2019, 02:54 PM | #26 | |
Banned
Join Date: Aug 2019
Location: Ban land in 3...2...
Posts: 2,943
|
Quote:
Yes, I think it's possible for a worse hitter to be a better overall player than a better hitter when one remembers that base running and defense also count. That said, it is possible that OOTP has a poor WAR model. But, that wouldn't invalidate WAR as a concept anymore than OOTP miscalculating batting average would invalidate batting average. You haven't given enough information on these to players to comment on who had the better season . Nor do I know enough about your league's setup or OOTP's WAR calculation to come to any strong conclusions. That said, at first glance, I'm surprised there's that much difference between them. Pennypacker is the better hitter. But he's not THAT much better. 150 to 120 wRC+. Then there's baserunning. Both were poor base stealers. Each probably provided negative value with stolen bases. Pennypacker certainly did. Bet there's also getting thrown out trying out trying to advance and taking extra bases, which, I have no information for. Then, defense. Estave looks like an excellent defender while Pennypacker is terrible. By ratings. Again, there's no information about how they actually performed. That's still pretty big gap to be explained by defense. But, looking at these batting lines and seeing how few strikeouts there are, we have to remember that means a lot more balls in play. Which makes defense more valuable. So, yes, it's certainly possible that 1) Estave had a 6 WAR season and Pennypacker a 1.5 WAR season. AND that Estave was the more productive player. 2) That OOTP's WAR calculation isn't very good. And specifically that it's not calibrated for the much different conditions in your game compared to 2019 real life baseball Not enough information to say. Last edited by CBeisbol; 09-08-2019 at 02:55 PM. Reason: Typos |
|
09-08-2019, 04:03 PM | #27 |
All Star Starter
Join Date: Oct 2013
Posts: 1,947
|
Estave also had more rbi’s and runs. A lot more stolen bases and apparently a much better defensive player. If you look primarily just at their batting then yes Pennypacker is the better player. If you look at their overall play???
IRL the player with the best combination of average, home runs, rbi’s almost always wins the MVP but until recently, they had no concept of overall value or defensive value or know the difference between batting average and on base percentage. Last edited by Reed; 09-08-2019 at 04:17 PM. Reason: Spelling |
09-08-2019, 04:12 PM | #28 |
Hall Of Famer
Join Date: Apr 2009
Location: United States
Posts: 10,456
|
I hear you, CBeisbol (and Reed). We have differing opinions, and that's fine. Baseball is big enough to support lots of different opinions. Some caveats:
I think Gibson was the best all-around batter and baserunner in the league this year. As far as fielding, that's for my Golden Glove Award. Thus, the dark horse for the Batting Excellence Award (plus, he's an interesting 2-way player). And a new approach to manually picking my awards, thanks to participating in this thread.
__________________
- Bru Last edited by Déjà Bru; 09-08-2019 at 04:26 PM. |
09-08-2019, 04:38 PM | #29 | ||||
Banned
Join Date: Aug 2019
Location: Ban land in 3...2...
Posts: 2,943
|
Quote:
But it's factually inaccurate to say that, for example, measuring distances by pacing them off is more accurate than using a meter stick - assuming you have a decent meter stick (which is not just a glib comment). Quote:
Again, the loudest critics of WAR show the least understanding of it. People who understand WAR realize it is a accurate, but imperfect, tool. Quote:
Again, most any WAR is calculation is simply going to be Batting runs+fielding runs+defense runs+replacement level. How those are calculated, and what inputs are used to calculate them, can vary. This is why looking at different WARs is valuable. When you understand the differences in the inputs you can understand why different WARs may rate players differently. Quote:
Like, say, OPS A guy has a .750 OPS. Is that good or bad? Well, if it was in 1875 when the league OPS was .571, it was pretty good If it was in 1894 when it was .819, it was not as good. You could do the same with batting average, ERA, OBP, etc... Last edited by CBeisbol; 09-08-2019 at 04:44 PM. |
||||
09-08-2019, 05:16 PM | #30 |
Hall Of Famer
Join Date: Apr 2009
Location: United States
Posts: 10,456
|
If I am silent from this point on, please do not think that you have won your points or have convinced me as to their accuracy or applicability. I merely wish to be done with WAR and move on to other business. Further back-and-forth will be fruitless and frankly, some of your comments sound a bit condescending so we will stop our conversation right here. You may have the last word, however, since you seem to need it. Follow up after this post.
EDIT: Heh, apparently I haven't changed much in my opinion of WAR in six and a half years. I'm at war with WAR!
__________________
- Bru Last edited by Déjà Bru; 09-08-2019 at 05:30 PM. |
09-08-2019, 06:28 PM | #31 | |
Banned
Join Date: Aug 2019
Location: Ban land in 3...2...
Posts: 2,943
|
Quote:
I didn't suppose you to be the type of person who changed your opinion when presented with further evidence. Even 6 years' worth. WAR works, and is quite accurate, even if you don't like it, and don't want to take the time to understand it. For anyone else who does, information has been presented in this thread. |
|
09-08-2019, 09:16 PM | #32 | |
Hall Of Famer
Join Date: Jul 2015
Location: Parts unknown
Posts: 6,397
|
Quote:
If you call what you posted "easy" congrats. You are a math wunderkind. I'm telling you to the general fan it is complicated. Didn't say it was wrong. Didn't say it was invalid. I'm saying it is complicated. What is BR? What is FR? What is BSR? etc.......... Yes fangraps "explains" each one....... but what is wRAA? What is wOBA? what is lgwOBA? etc, etc etc. What is the league adjustment? What is the park adjustment? How did they figure that out? How do I know the figure they came up with is accurate? How do I know it has any merit? And after all that, how do I know all those numbers really did come up with a true depiction of that player's value in that stat? Now I have to all those questions answered similarly in every other formula that is used to create the formula components that make up the formula. How much time is that going to take? After I explain it to myself, will I agree with it, or call it a bunch of hooey? How is that not complicated? As I've said before on this forum, I applaud the effort the sabermetric fanboys have made. It would be nice to have definitive stat that measures a player's value completely instead of just "eyeballin' " it like we have for years. Keep working on it. Hope you get 'er done. But if I don't understand it, if there is no consensus on it, if the creators of each one each year have to adjust it because it wasn't "just right", then how am I going to put faith in it? Because that is just what it is...faith. I have to hope that it's correct. Not know that it is correct. That complicates things. I don't dismiss WAR in any sports debate. But I also don't automatically accept it either. It is one more spice in the gumbo.
__________________
If a man is guilty 4 what goes on inside of his mind, then let me get the electric chair 4 all my future crimes. - Prince Batdance June 7, 1958 - Apr 21, 2016 |
|
09-08-2019, 10:18 PM | #33 | |||||
Banned
Join Date: Aug 2019
Location: Ban land in 3...2...
Posts: 2,943
|
Quote:
That was the point about the toaster. OPS is easy-once someone has already figured out OBP and SLG for you. Quote:
Quote:
If one can understand OBP and SLG and add them together (even if they can't, or don't, figure them out for themselves and just look them up) then one can understand Batting Runs (how many runs the player produced by hitting), Fielding Runs (how many runs the player saved defensively), Base Running Runs (how many runs the player was worth running the bases) and add them together (even if they can't, or don't, figure them out for themselves and just look them up. Quote:
https://library.fangraphs.com/offense/woba/ lgwOBA (league wOBA) is the average wOBA for the entire league. wRAA (weighted Runs Above Average) is based on wOBA and is how many runs above (or below) average a player was worth based off of their hitting (not including base running (except in how base running effects a player getting (or not getting) an extra base hit). https://library.fangraphs.com/offense/wraa/ There's also wRC+ (adjusted weighted Runs Created) which is like OPS+ in that it compares to league average how many runs a player's batting was worth on a rate basis. 100=average. >100 is above average. <100 is below average. Quote:
It's not faith. Not at all. You can prove that WAR works. And since WAR works you can extrapolate that the component parts work. Here's the proof. https://www.beyondtheboxscore.com/20...ird-order-wins You can do this yourself to confirm. Remember WAR measures what each player does without considering what his teammates have done (for the most part). So, if a batter homers, or a defender makes an error, or a pitcher walks a hitter WAR doesn't know if the bases were loaded, or empty, or how many outs there were, or what the score was, etc. And looking at each players' individual WAR and adding them up for each team, and comparing how many wins a team 'should' have by WAR and how many wins they actually have, you see that WAR very closely reflects the number of actual wins. There are, of course, differences. Because a solo home run in a 10-1 game impacts who wins the game a lot less than a 9th inning walk-off home run does. But try looking at a team's collective batting average, home runs, put outs, errors, ERAs, and walks given up and try to guess how many wins they have. WAR works It's a fact. Yes, people are always trying to improve it. It's a good thing that no one decided that it was good enough when they first thought of it. We should change our opinions as we get more information. Last edited by CBeisbol; 09-08-2019 at 10:19 PM. |
|||||
09-08-2019, 10:47 PM | #34 |
Hall Of Famer
Join Date: Jul 2015
Location: Parts unknown
Posts: 6,397
|
That right there is why I can't stand the sabermetric cultists.
__________________
If a man is guilty 4 what goes on inside of his mind, then let me get the electric chair 4 all my future crimes. - Prince Batdance June 7, 1958 - Apr 21, 2016 |
09-09-2019, 12:33 AM | #35 | |
Banned
Join Date: Aug 2019
Location: Ban land in 3...2...
Posts: 2,943
|
Quote:
Someone upthread mentioned a "math wunderkid". Understanding correlation and slope-intercept equations (middle school level math) might make some a "math wunderkid" in comparison to others. |
|
09-09-2019, 07:05 AM | #36 | |
Hall Of Famer
Join Date: Jul 2015
Location: Parts unknown
Posts: 6,397
|
Quote:
The sun rises every day. That is a fact. No one is changing that. Michael Jordan won 6 NBA titles. That is a fact. No one is coming up evidence that will contradict that statement. Gravity is a fact. No one is switching it off and stuff is going to go floating in the air. WAR cannot be a fact if it changes from year to year and there is no consensus on how to come up with it. fangraps changes theirs. baseball-reference changes theirs. Bill James changes his. They have created theories. Estimations. NOT FACTS.
__________________
If a man is guilty 4 what goes on inside of his mind, then let me get the electric chair 4 all my future crimes. - Prince Batdance June 7, 1958 - Apr 21, 2016 |
|
09-09-2019, 08:39 AM | #37 |
All Star Reserve
Join Date: Mar 2015
Location: Australia
Posts: 651
|
I am not going to get in a back and forth, but just because the formula changes slightly does not mean the underlining reasoning, theory and calculations are not sound.
|
09-09-2019, 09:04 AM | #38 | |
Hall Of Famer
Join Date: Jul 2015
Location: Parts unknown
Posts: 6,397
|
Quote:
The science of car design changes. New gadgets are invented. New technology is created. Someone can find a way to make them fly. Someone else can find a way for it to go underwater. Someone can make it go faster. Someone else can make it more efficient. Someone may be able to put weapons on it. Whatever. The point is it can't be the "ultimate" because it will always change. It can always be better. Until the endpoint is reached, until all are agreed that there is no way it can be bettered, any claim that it is the standard is an overstatement. WAR is not a fact. It is an opinion. A step forward. But the destination to "factdom" has not been reached.
__________________
If a man is guilty 4 what goes on inside of his mind, then let me get the electric chair 4 all my future crimes. - Prince Batdance June 7, 1958 - Apr 21, 2016 |
|
09-09-2019, 09:08 AM | #39 | |
OOTP Developer
Join Date: Jun 2009
Location: Here and there
Posts: 14,145
|
Quote:
Or, who was worth more to the Jays this year: Galvis in 115 games, or Bichette in 37? Comparing using other metrics is impossible, since it's so hard to compare a few games of great production vs many games of lesser production. But by WAR, they're even. To me, that's the power, in that it can help sort things out for players who have a different profile, who have varying skills, and gives a nice neat value to everyone. |
|
09-09-2019, 09:36 AM | #40 | |
Hall Of Famer
Join Date: Apr 2010
Posts: 2,152
|
Quote:
The top 3 players by war this season are mike trout, christian yelich, and cody bellinger who everyone would consider the three best players in baseball so it seems like it is doing something right. No one who knows what they are talking about will proclaim that it is perfect (no statistic tells the whole story) or that a 2B with 3.5 war is definitively more valuable than a pitcher with 3.3 but it helps to start putting things into perspective. Last edited by dkgo; 09-09-2019 at 09:41 AM. |
|
Bookmarks |
|
|