Somebody explain WAR already...

[Goody]

What is it?

Is it some sabermetric thing?

Is it something that would benefit replay leagues?

Why is there such a huge calling for it from many people here?

Does it really take into account real life physics, mental states, and intangiables? (I think a random number generator does that just fine with the stats we already have available).

Sabermetrics and all these extra stats are, IMO, just a bunch of fluff that overanalyzes things in numbers...when real world baseball is not just a bunch of numbers and calculations.

You can't predict how Milton Bradley will perform this year by using a spreadsheet or a formula. Fans know this, scouts know this, writers know this....

No decision in baseball is made on purely numbers and formulas.

So how will WAR benefit OOTP? And what the heck is it? I had enough stats in the game back in 2003 or so.

It did the job and moreso. I could see where more robust calculations behind the scenes (in the code) would make OOTP better...but that happens every year. What's the obsession with this WAR?

[Goody]

In short, I'd rather have the outcomes be a bit mysterious, in the background code...rather than have everything be decided but some sabermetric stat and predictable.

[ms2002]

Wins Above Replacement (WAR) Sabermetrics Library

I recommend that entire site to you. Hope this helps..

[TribeFanInNC]

Basically it sums a player's contribution into one number. Kinda like VORP does except that WAR includes fielding.

So yes, it is a sabermetric thing.

[CBL-Commish]

I'm sure this is going to be a fun thread.

In short, WAR is a one number summary of a player's contribution (hitting, pitching, fielding) above and beyond that of a replacement-level player.

Has nothing to do with calculating outcomes within the game. Has nothing to do with projecting Milton Bradley's performance in 2010. Has nothing to do with physics or mental states or intangibles. No moreso than those things impact RBI or Runs or saves or whatever. It would be calculated after the fact.

It gives us a way to simply and easily compare the value of all kinds of players across all kinds of contexts. It allows us to say that Mark Belanger was about 33 wins over replacement for his career, while Eric Karros was about nine.

One big problem with WAR in OOTP is how you define replacement level in a game that allows an almost infinite varity of configurations within a universe? I guess you could just set replacement level as a .300-ish (winning percentage) player in any league, but OOTP certainly allows one to create leagues in a universe where .300-ish players are either very hard to obtain, or very easy.

[JackRules1]

Baseball, more so than every other sport, is about the numbers. But does a person need to be up on the latest stats to enjoy the game? Absolutely not. I happen to enjoy reading about WAR, FIP, Winshares, etc but it really is only to help me to enjoy our national passtime in a different way. The sabrmaticians believe that the old standbys: ERA, AVG, RBI were not sufficiently telling us the whole story concerning a players productivity, so they have come up with new ways to help us to measure how much a player actually contributes to wins and losses. But the real story is in the eye of the beholder. I don't need a bunch of fancy stats to tell me that Albert Pujols is a great hitter, but if I want, I can look at the new stats to decide just how great of a hitter Albert really is.

I like having these new stats in OOTP, but I agree that the game should not become reliant on the stats to determine how players will perform. That's why we play the game, because we don't know for sure that any player will perform to match the stats. Just like in real life.

And thanks for the link, ms2002.

[RandyTiger]

Originally Posted by jackrules1

i like having these new stats in ootp, but i agree that the game should not become reliant on the stats to determine how players will perform. That's why we play the game, because we don't know for sure that any player will perform to match the stats. Just like in real life.

+1.

[Goody]

It still doesn't take into account that maybe player A happened to face tougher pitching than player B.

Player A might have worn out the opposing pitcher for player B by batting before him in the line up.

Pitcher A might have had a better catcher calling his games than pitcher B.

A defensive player might have more pitchers that get groundouts than another team's staff.

When a player is replaced, moved or something changes in any player's life the stats will be affected for various different players. There can be no one all inclusive stat that sums up worth to a team.

[JackRules1]

Originally Posted by Goody

It still doesn't take into account that maybe player A happened to face tougher pitching than player B.

Player A might have worn out the opposing pitcher for player B by batting before him in the line up.

Pitcher A might have had a better catcher calling his games than pitcher B.

A defensive player might have more pitchers that get groundouts than another team's staff.

When a player is replaced, moved or something changes in any player's life the stats will be affected for various different players. There can be no one all inclusive stat that sums up worth to a team.

I can't argue with any of that. Why was the Yankee's team ERA 1.2 runs higher with Posada behind the plate compared to the other catchers in 2009? If it is all Posada's doing I imagine he would be out of a job. That's a stat that tells us something, but certainly not the whole story.

[Biggio509]

Originally Posted by CBL-Commish

I'm sure this is going to be a fun thread.
One big problem with WAR in OOTP is how you define replacement level in a game that allows an almost infinite varity of configurations within a universe? I guess you could just set replacement level as a .300-ish (winning percentage) player in any league, but OOTP certainly allows one to create leagues in a universe where .300-ish players are either very hard to obtain, or very easy.

I don't think it really matters in the end. It may get a few people saying WAR is not coming out realistically but a higher WAR would still be a higher WAR. What you might find is that in some universes everyone has a high WAR and in another exterme every one has a negative WAR. The stat still says the same thing if two WARs are 60 and 50 or if they are -20 and -10. The stat works but some will complain neither extreme is realistic.

WAR could be tied to the sabremetric player creation stats so the replacement player is the average player created by your settings. So a WAR in the California league maybe not be perfectly comparable to a WAR in the Carolina league but they both tell you where the player in relation to the average player of the league.

I think it might be an excellent tool for the minors because it would give a real quick reference to see if a player should be at the level he is at. A negative WAR might mean he needs to be demoted while a high WAR means he should move up. Granted you would not want to rely on one stat but it could be a good first check.

[CBL-Commish]

Originally Posted by Goody

It still doesn't take into account that maybe player A happened to face tougher pitching than player B.

It certainly could. Many WAR, or WAR-like systems like the ones at BP do adjust for strength of schedule or even individual opponents.

Player A might have worn out the opposing pitcher for player B by batting before him in the line up.

Pitcher A might have had a better catcher calling his games than pitcher B.

If anyone could prove this happens to any great degree or finds a way to quantify it, you could certainly adjust for it. Since nobody has figured out whether this stuff even exists in a meaningful way, no, WAR won't account for it.

A defensive player might have more pitchers that get groundouts than another team's staff.

Again, any good system takes this into account. Bill James' win shares was adjusting for this 10 years ago.

When a player is replaced, moved or something changes in any player's life the stats will be affected for various different players. There can be no one all inclusive stat that sums up worth to a team.

Ok. Let's go back to the original point - why have WAR in OOTP? Because it's a great tool to use to evaulate players. Of course it doesn't capture everything. All of your criticisms and many, many more could be used to argue ERA and RBI shouldn't be in the game, either.

If perfection is a prerequisite for including a stat in a text-based baseball sim then we're going to have to rework the engine to just tell long stories about each game, and maybe judge players only through facegen images of happy/sad/angry coaches.

[Malleus Dei]

Everything you always wanted to know about WAR

[Goody]

Originally Posted by Malleus Dei

Everything you always wanted to know about WAR

That doesn't even tell me how the "replacement level player" baseline is calculated.

Is it based off of assumptions?

How would it be implemented in OOTP? League ERA settings?

This would take too much of the mystery and strategy out of it. In the real world there's so many factors, so stats can only get you so far.

We're stuck with just numbers simulating the actions and decisions of human beings in OOTP, so the more you make stats available that show the probability of these behind the scenes calculations in the code...the less random it becomes.

Why not just make a game that gives each team a rating of 1 to 100 and roles a random number to modify the outcome a bit to see who wins?

[Goody]

Originally Posted by JackRules1

I can't argue with any of that. Why was the Yankee's team ERA 1.2 runs higher with Posada behind the plate compared to the other catchers in 2009? If it is all Posada's doing I imagine he would be out of a job. That's a stat that tells us something, but certainly not the whole story.

Something similar happened with Kenji Johjima's stats last year. And he is out of a job.

[attackemu]

Dude, showing WAR wouldn't "show the probability of the behind the scenes calculations," whatever that tortured sentence means. It certainly doesn't take the mystery and strategy out of it. It's just a stat like any other. It'd be a reflection of what already happened in the game. In game terms, it wouldn't give you any insight into whether the player had been lucky or just good, or whether he'd had a ratings boost, or anything else. Why do we want WAR in the game? Why do you want RBI and Wins in the game? Just another way to evaluate our players. If you feel like knowing how many bases your player steal, or what their ERA is, tells you too much, then I guess you should just play that Random Number game of yours.

[Go Tribe]

i think first you have to blame someone for stealing yellow cake uranium

or something like that

[Malleus Dei]

Replacement level defined and five hundred other places.

[robc]

Originally Posted by Goody

A defensive player might have more pitchers that get groundouts than another team's staff.

I believe WAR uses the ultimate zone rating for defense, which looks at the percentage of balls a fielder gets to in his zones. So, total fielded balls don't matter. It is the % of possible balls fielded.

[Antonin]

Originally Posted by attackemu

Dude, showing WAR wouldn't "show the probability of the behind the scenes calculations," whatever that tortured sentence means. It certainly doesn't take the mystery and strategy out of it. It's just a stat like any other. It'd be a reflection of what already happened in the game...

Exactly. And it would help you decide whether to put player A or player B into today's lineup, or who to tap as your pinch hitter in the bottom of the 9th. In real life, managers faced with a situation try to use whatever player might be best for that situation. Stats help real life managers make that decision and they can also help OOTP players make that decision.

In Cleveland, where I live, former Indians manager Mike Hargrove was well known for playing the stats game. He relied on lefty-righty and other statistically-based matchups to such an extent that he would sometimes bring in one pitcher to face one single batter.

It sometimes drove Indians fans crazy.

Originally Posted by attackemu

Why do we want WAR in the game? Why do you want RBI and Wins in the game? Just another way to evaluate our players. If you feel like knowing how many bases your player steal, or what their ERA is, tells you too much, then I guess you should just play that Random Number game of yours.

For me, stats and the study of same are part of the fun of baseball. I play OOTP because it lets me fantasize about being a big-league manager. Evaluating players is what big-league managers spend a lot of time doing, and new ways to evaluate increases the fun [for me]. I sometimes print out stat reports and study them when I can't actually play OOTP. They fascinate me.

I became interested in VORP after reading about it on this forum. Now, I'm really looking forward to the addition of WAR to this game.

I think the addition of new stats to OOTP will do a lot more to make it a better baseball game than some of the other features that have been requested.

[Goody]

Originally Posted by Malleus Dei

Replacement level defined and five hundred other places.

Exactly how bad of a hitter do you have to be for consideration as a replacement level player? Tango Tiger has recently updated his position adjustments. These are based on defensive statistics, not offense. He looks at how players who play multiple positions perform when switching for one to another, with some consideration for handedness, as lefties do not play second, third, or short.

Tango Tiger's Position Adjustments
POSITION POSITION ADJUSTMENT
Catcher 12.5
Shortstop 7.5
2B, 3B, CF 2.5
Corner Outfield -7.5
First base -12.5


If we consider a replacement level player to be 20 runs below major league average, you get this:

POSITION REPLACEMENT LEVEL
Catcher -32.5
Shortstop -27.5
2B, 3B, CF -22.5
Corner Outfield -12.5
First base -7.5"

Who is Tango Tiger and why does his opinion get to have an effect on a statistic?

Statistics should be just that, compiled outcomes of what happened. It seems WAR is constantly changing based on human input and evaluation.