On 2/21/2024 at 2:57 PM, DJL44 said:

#1 -WAR is a zero-sum game so if Yadier Molina has value that isn't being properly attributed, that means his teammates (especially his pitchers) were actually worse than they appear. It also means that other MLB catchers were a lot worse than they appear.

#2 Are you saying all the good defensive catchers are underrated? That would mean WAR is not giving enough defensive value to various defensive events. Blocked pitches, caught stealing, wild pitches and fielding bunts are pretty easy to quantify the run value so that can't be where the error lies. It would have to be that credit for strikeouts and walks that should go to the catcher is instead going to the pitcher.

#1) Thanks for responding DJL44. I have no idea how WAR is calculated. Are you saying each (1) win (of the winning team) is distributed to each person who contributed according to the value that the person who came up with WAR?

#2) When looking at total WAR, I do think that not enough WAR is distributed to defense compared to offense which a slugging catcher is given too much value over a defensive catcher at a defensive premium position. CCOF5yrstoolate mentioned that Fangraph is more favorable to defense but at catching, there are too many intangibles. A good catcher adds a lot to the success of a pitcher should he share in that WAR? And how do you determine that & how do you calculate that? It gets very complicated. 

I like to dabble at Baseball Trading Values. They calculate a player's trading value by subtracting his salary from his ATV coming up with a surplus value. This is a tool to help you come up with a reasonable trade offer and we can evaluate actual trades. When making an offer you try to balance the total surplus value of one team to be equal to the other. Should FO use this tool to determine whether to accept or reject a trade? Absolute not, Even if a trade offer is fair according to BTV or even an overpay, I reject 10 offers for every one I accept because the offer doesn't meet my needs or I don't agree with the value they give.

My point is that BTV is a tool but it's not absolute. So also total WAR can be an useful tool (I look at WAR) but it's not absolute & should not be used entirely to affect how we make our decisions.

12 hours ago, Richie the Rally Goat said:

https://library.fangraphs.com/misc/war/
 

“You should always use more than one metric at a time when evaluating players, but WAR is all-inclusive and provides a useful reference point for comparing players.”

 

Thank you, Ritchie. This site was very useful & I believe Fangraph could be more accurate than the others. They gave a good explanation. IMO offensive WAR is more straightforward to calculate than defense (especially catching many intangibles) so defense isn't given enough credit (ex, DP is credited as 1 out). And baserunning, is a player that steals 2B & 3B given the same credit as a hitter that hits a triple?

When considering trading a player or trading for a player, do we look only at the WAR from those players from the year before? W/o looking at any underlying conditions (either favorable or unfavorable)  that no longer exist? WAR is a valuable tool but we shouldn't blindly ignore the underlying conditions that existed.

16 minutes ago, Doctor Gast said:

Thank you, Ritchie. This site was very useful & I believe Fangraph could be more accurate than the others. They gave a good explanation. IMO offensive WAR is more straightforward to calculate than defense (especially catching many intangibles) so defense isn't given enough credit (ex, DP is credited as 1 out). And baserunning, is a player that steals 2B & 3B given the same credit as a hitter that hits a triple?

When considering trading a player or trading for a player, do we look only at the WAR from those players from the year before? W/o looking at any underlying conditions (either favorable or unfavorable)  that no longer exist? WAR is a valuable tool but we shouldn't blindly ignore the underlying conditions that existed.

100% agree on context being extremely important. 
 

Sample size is key on any metric, and comparing apples to apples. So WAR isn’t helpful at all for current season in April, I bristle seeing it quoted in May, but when you’re coming up on half a season and discussing trade deadline, it can be useful.

As others have mentioned, WAR is a cumulative stat, so I usually try to pair it with rate stats, to try to bring context. Hitters: OPS is my go to. Pitchers, I like WHIP. I tend to measure fielding separately, baseball savant outs above average (also a cumulative stat). Is a nice comparison.

My opinion is simple.  Advanced statistics have been created for the sole purpose of providing "factual data" that supports the developer's point of view.  For example, if I believe that HRs are more important than anything else, I would create something like WAR in order to provide "proof" that my point of view is the only one that matters.  That's why there are a million advanced statistics now and even more being created as I write this. 

(And, if you haven't already guessed, I'm old.  But not senile. 😁).

2 hours ago, Doctor Gast said:

#1) Thanks for responding DJL44. I have no idea how WAR is calculated. Are you saying each (1) win (of the winning team) is distributed to each person who contributed according to the value that the person who came up with WAR?

#2) When looking at total WAR, I do think that not enough WAR is distributed to defense compared to offense which a slugging catcher is given too much value over a defensive catcher at a defensive premium position. CCOF5yrstoolate mentioned that Fangraph is more favorable to defense but at catching, there are too many intangibles. A good catcher adds a lot to the success of a pitcher should he share in that WAR? And how do you determine that & how do you calculate that? It gets very complicated. 

It's pretty easy to look up how it is calculated. There is a set amount of WAR for the whole league. Pretty much all of the calculators agree on the run value of offensive events. (a single, walk, stolen base, etc). The amount of WAR allocated to run creation is the same amount allocated to run prevention (pitching and defense). If the amount allocated to defense is too low then the amount allocated to pitching has to be too high.

There are only two ways WAR could undercount Molina's defense:

1) The best catchers are better than they appear and the worst catchers are worse than they appear. In other words, RField has the right average but the standard deviation is wrong. Catcher defense would be regressed too much to the mean. This gets problematic because it would suggest the worst 1/3 of catchers shouldn't be catching at all. A team starting Matt LeCroy or Ryan Doumit would put the team in a 1 run hole every game they caught.

2) The division of run prevention is incorrect between pitchers and catchers. Catchers should be getting more credit for strikeouts (and debit for walks) than they currently do. This gets into how much pitch framing and game calling are worth.

There is a separate argument that catcher replacement value is wrong because it's based on position switching but that's secondary to the other points above and would affect all catchers equally based on games caught.

Just now, terrydactyls said:

My opinion is simple.  Advanced statistics have been created for the sole purpose of providing "factual data" that supports the developer's point of view.  For example, if I believe that HRs are more important than anything else, I would create something like WAR in order to provide "proof" that my point of view is the only one that matters.  That's why there are a million advanced statistics now and even more being created as I write this. 

(And, if you haven't already guessed, I'm old.  But not senile. 😁).

A huge difference in the last 20 plus years is the publishing of data from various people and sources. Baseball teams have always kept data and used formulas, but it seemed like most of it was private. As a coach and when I played I used a whole (fairly complex) system of ratings that eventually were broken into numerical values to attempt to keep an objective idea of things such as fielding, running, batting, and pitching. These models were used for predictions, assessments, and devising optimum strategies. They were just my way of making sense of what was in front of me. The ability to make some money by putting together everything in a fashion that someone else could understand and use never occurred to me or others. I'm not sure anyone would understand the maze I created. I don't think I could simplify it.

I think it might be correct to say that all of the analytics today give a more complete picture to a consuming public but it is the data and information produced from recent technology that has made a more substantial difference for players (trackman and that sort of thing). When managers used platooning and shuffled certain pitchers to specific roles in the bullpen 50 years ago, it wasn't just a hunch.  There were systems in place. The change (my opinion) has been in the vast collection of published data which may be homogenizing how teams evaluate and use players. The one constant is change.

1 hour ago, Doctor Gast said:

Thank you, Ritchie. This site was very useful & I believe Fangraph could be more accurate than the others. They gave a good explanation. IMO offensive WAR is more straightforward to calculate than defense (especially catching many intangibles) so defense isn't given enough credit (ex, DP is credited as 1 out). And baserunning, is a player that steals 2B & 3B given the same credit as a hitter that hits a triple?

When considering trading a player or trading for a player, do we look only at the WAR from those players from the year before? W/o looking at any underlying conditions (either favorable or unfavorable)  that no longer exist? WAR is a valuable tool but we shouldn't blindly ignore the underlying conditions that existed.

Also… I like fangraphs, and it’s my go-to, but baseball-reference has a glossary too

https://www.baseball-reference.com/about/

https://www.baseball-reference.com/about/war_explained.shtml

3 hours ago, DJL44 said:

It's pretty easy to look up how it is calculated. There is a set amount of WAR for the whole league. Pretty much all of the calculators agree on the run value of offensive events. (a single, walk, stolen base, etc). The amount of WAR allocated to run creation is the same amount allocated to run prevention (pitching and defense). If the amount allocated to defense is too low then the amount allocated to pitching has to be too high.

There are only two ways WAR could undercount Molina's defense:

1) The best catchers are better than they appear and the worst catchers are worse than they appear. In other words, RField has the right average but the standard deviation is wrong. Catcher defense would be regressed too much to the mean. This gets problematic because it would suggest the worst 1/3 of catchers shouldn't be catching at all. A team starting Matt LeCroy or Ryan Doumit would put the team in a 1 run hole every game they caught.

2) The division of run prevention is incorrect between pitchers and catchers. Catchers should be getting more credit for strikeouts (and debit for walks) than they currently do. This gets into how much pitch framing and game calling are worth.

There is a separate argument that catcher replacement value is wrong because it's based on position switching but that's secondary to the other points above and would affect all catchers equally based on games caught.

Thanks, IMO under these circumstances, I'd prefer to have 1st catchers, then SS & CFers have a great bulk of the defensive WAR & other positions have a lesser % depending on importance. & inferior defenders receive less defensive WAR.

But better yet, Offense WAR & Defensive WAR should be equal & Pitching should have a separate stat to measure their proficiency. This way the better defenders especially premium positions will be better represented. Because Defensive WAR is being diluted by sharing with Pitching WAR. The player with higher Offensive WAR has an advantage over a defensive player in Total WAR.

14 minutes ago, Doctor Gast said:

Thanks, IMO under these circumstances, I'd prefer to have 1st catchers, then SS & CFers have a great bulk of the defensive WAR & other positions have a lesser % depending on importance. & inferior defenders receive less defensive WAR.

But better yet, Offense WAR & Defensive WAR should be equal & Pitching should have a separate stat to measure their proficiency. This way the better defenders especially premium positions will be better represented. Because Defensive WAR is being diluted by sharing with Pitching WAR 

WAR is assigned defensively by the number of events at the position. Players get credit for making plays and debited if they miss them. RField is calibrated versus the average player at the position. Do you want to just make up events that didn't happen in the games to give SS and CF more credit? That would be credit for standing on the field watching someone else make the play.

The second paragraph suggestion would fundamentally miscalculate WAR. Runs created and runs allowed have to balance. Your accounting would credit the offense for (example) creating 4 runs in a game but debit the defense for 8 runs allowed. Fielding and pitching HAVE to share defensive runs allowed. It's simple accounting; a credit in one place has to be a debit somewhere else.

The great Bill James weighs in:

https://www.billjamesonline.com/judge_and_altuve/

Absolutely Nothing! 

4 hours ago, Richie the Rally Goat said:

The great Bill James weighs in:

https://www.billjamesonline.com/judge_and_altuve/

Bill James and his work is well known as are his arguments of the means now used. Of note, but perhaps even having relevance, is that James finds disagreement with systems separate from his own. James has a special ego. I always found his work interesting but I hope I'm not the only one who doesn't believe or think that James and his generation were the first to consider data or actually believe that baseball people from before 1970 had no idea what they were doing or how to develop means to winning games.

Every time new civilizations are "discovered" from long ago, people are astonished to learn that some ancient crude group of humans had means of production and sophistication in their lives which seem impossible because our current era is the best ever. How about the science uncovered in the last fifty years that animals are intelligent and can communicate. We are learning .... slowly. Perhaps some day someone will find John McGraws complex notebooks that he devised to identify which player worked best in each specific situation over 100 years ago.

14 hours ago, tony&rodney said:

Bill James and his work is well known as are his arguments of the means now used. Of note, but perhaps even having relevance, is that James finds disagreement with systems separate from his own. James has a special ego. I always found his work interesting but I hope I'm not the only one who doesn't believe or think that James and his generation were the first to consider data or actually believe that baseball people from before 1970 had no idea what they were doing or how to develop means to winning games.

Every time new civilizations are "discovered" from long ago, people are astonished to learn that some ancient crude group of humans had means of production and sophistication in their lives which seem impossible because our current era is the best ever. How about the science uncovered in the last fifty years that animals are intelligent and can communicate. We are learning .... slowly. Perhaps some day someone will find John McGraws complex notebooks that he devised to identify which player worked best in each specific situation over 100 years ago.

 

100% agreed, I thought James was well reasoned, and he fits with the OP, but I disagree with him, on the history and his qualms with WAR.

its not a perfect metric, but its good, and more importantly its easy and accessible.

baseball has always been a stat heavy sport, from the earliest days. Our stats have become more complex with access to computers, but keeping the box scores in the paper books, scouting reports… it’s all analysis of data with the tools you have.

Taylor Walls in 2022. 

466 PA's - .172 BA - .553 OPS

What does Baseball Reference list for bWAR and what does Fangraphs list for fWAR? 

Just asking for a friend. 

2 hours ago, Riverbrian said:

What does Baseball Reference list for bWAR and what does Fangraphs list for fWAR? 

Just asking for a friend. 

Does your friend not have an internet connection?

22 minutes ago, DJL44 said:

Does your friend not have an internet connection?

I suspect that my friend already looked up the answer on the internet and is just being an ass. 😄

As a poster stated earlier: 

7 minutes ago, RpR said:

As a poster stated earlier: 

War, huh, yeah
What is it good for?
Absolutely nothing, uhh
War, huh, yeah
What is it good for?
Absolutely nothing
Say it again, y'all
War, huh (good God)
What is it good for?
Absolutely nothing, listen to me, oh

I am so tired of people bringing up this song when discussing analytics. It's not clever the 5000th time.

5 hours ago, DJL44 said:

I am so tired of people bringing up this song when discussing analytics. It's not clever the 5000th time.

I seen you around for a long, long time
I remembered you when you drank my wine

Why can't we be friends?
Why can't we be friends?
Why can't we be friends?
Why can't we be friends?

Sorry, the defensive side of it is too full of holes. The concept of one number that “says it all” is cool…unfortunately, you can’t count on the defense part of the math to render a number that isn’t distorted somewhat…or one that is even consistent year to year for the same player.

I go by OPS+ and/or wRC+ for offense, and eyeballs (mostly) for defense. And in today’s game, with fewer and fewer balls put in play, the offensive side has more weight than ever in assessing the total value of the player of any position (on a relative basis). I don’t necessarily like that. But it is what it is…and more balls in play doesn’t appear to be anywhere on the horizon.

Defensive metrics are tough to calculate, so I think what you said has merit. Let's say for example, one catcher sees that a hitter is very anxious and aggressive at the plate, so he calls for a slider off the plate and the batter whiffs. Another catcher in the same situation does not realize this and so calls for a fastball and the hitter cranks a homerun to win the game. In a sense, the catcher has determined the outcome more than the pitcher, but that is almost impossible to measure. That is a simplistic example, but I think it demonstrates just one of the problems measuring defensive metrics.

12 hours ago, Aerodeliria said:

Defensive metrics are tough to calculate, so I think what you said has merit. Let's say for example, one catcher sees that a hitter is very anxious and aggressive at the plate, so he calls for a slider off the plate and the batter whiffs. Another catcher in the same situation does not realize this and so calls for a fastball and the hitter cranks a homerun to win the game. In a sense, the catcher has determined the outcome more than the pitcher, but that is almost impossible to measure. That is a simplistic example, but I think it demonstrates just one of the problems measuring defensive metrics.

Lets say a pitcher's best pitches are FB/ slider. The catcher calls for a CU & the batter whiffs, the batter get's 2 strikes on him & the batter's on the FB/ slider. The catcher keeps calling FB/ slider & the batter keeps fouling them off driving up the pitch count. The best pitch on certain situations might not be the pitcher's best pitch. If the batter is on your FB/ slider, you change things up with a change up to get a SO even if its not your best pitch.