Apologies, but to project WAR for a player/lineup for an upcoming season, and equate that to a win total is dumb. There are so many more variables that factor in to wins and losses. You saber guys are excessively black & white sometimes. Just my HSO

yeah WAR is great for individuals, like, say, saying mike trout had the best season of anyone in baseball this year, but to then say the angels would have won 10 fewer games without him is silly, i think

WAR is also, according to the people who publish it, an arbitrary number with a variable as much as 2.0 when looking at player vs player. It is a weighted system used for comparing past performances and NOT any better than whatever you might choose to use as a predictor for future performance. The great one, Michael Young had a 2.1 in 2011 and a -2.4 in 2012. I think that comparing his years and obvious performance, it would have been hard to say that he was a true 2.1 player in 2011, so if you swing the variable in the negative direction could you say he was a .1 in 2011 and a -4.4 last year. I think so.

Well hey, at least you still have batting average to judge what a good hitter is.

At least you still have wins and losses to know who's a good pitcher vs. a bad one.

At least you still have errors to know who's a good fielder or not.

At least you still have Gold Gloves to know who flashes some leather.

At least you still have ESPN to know which teams are good or not.

At least you have MLB Network to know which players are good or not.

We shouldn't have traded Michael Young. He was our leader.

Wins Above Replacement is a concept for nerds and nerds are stupid.

WAR is stupid.

WAR ia just a bunch of made up stats. I don't believe them. I would rather just use my own perceptions to justify my HSO's instead of backing up my observations with actual objective data.

^i don't think anyone here is against WAR, just against using it to evaluate a team's wins/losses and/or the future, rather than an individual and/or the past

though stranger did say "you saber guys"

So, Elvis must be a crappy fielder, he was second in the AL with fielding errors...

Felix Hernandez won the cy young in 2010 with a record of 13-12. Carl Pavano was 17-11. I guess Pavano was the better pitcher...

Year after year there are Gold Glove winners who are a joke.. this year, Adam Jones, Jimmy Rollins, Carlos Gonzalez....

It's hard to lead from the bench... if you're looking at the eyeball test and not WAR, then you have to know that MY was a terrible liability in the field and last year at bat, as well...

Wow-be careful-you're gonna make Eric mad. He REALLY thinks WAR is important.

OK, WAR does have issues. So, what's the better alternative?

Why, emotions, feelings, and that little tickle in the gut of course!!

I'm not convinced the Angels wouldn't have lost even more than 10 games without Trout. The guy was a difference maker in so many of their games. WAR is a quick reference point; if you're doing in-depth analysis, you should rely mostly on things more precise than WAR.

As I understand it, everything that contributes to WAR happens in a vacuum. It doesn't take into account the context; whether or not that home run Beltre hit made the difference in the game (it did). It considers all home runs equal, which they aren't (remember that 18-3 game in Boston back in April? Just a few superfluous homers.)

I'm convinced that between their pitching and Pujols slumping, the Angels might have been below .500 if Wells had played all those games instead of Trout. It's possible.

WAR is important, IMO, but we do need to be careful about adding/subtracting WAR from actual win totals in an attempt to determine how many games we'll win next year.

The current crop of statistics aren't perfect, which is why they are constantly being improved. The bottom line is that projections based on the state of the art in baseball statistics are better than the projections based the guts of fans, which are biased and based on incomplete information. If you want to criticize WAR or any other statistic, that's fine. But your criticism isn't helpful unless it's constructive. In other words, you need to explain exactly how the statistic is flawed and suggest how it can be improved. That requires thoroughly understanding how it's calculated.

FYI, if your goal is to make projections, projection systems such as PECOTA and ZIPS are better than past WAR. We use WAR on this site b/c it's publicly available and the PECOTA and ZIPS projections for 2013 aren't out yet.

a .1 WAR for MY in 2011...really?? I know it isn't perfect, and you can give and take some, but its obious he added more thn .1 WAR in 2011.

"As I understand it, everything that contributes to WAR happens in a vacuum. It doesn't take into account the context; whether or not that home run Beltre hit made the difference in the game (it did). It considers all home runs equal, which they aren't (remember that 18-3 game in Boston back in April? Just a few superfluous homers.)"

The stat you're looking for is WPA-win probablility added. It's not very predictive, but it's pretty fun as a "storytelling" statistic.

"Why can't we be friends? ..." When I see this thread i get this tune running through my head over and over, lol....

@Fajita...just using the example of how a plus/minus variable of 2 is supposed to be considered when examining a player. It was supposed to look flawed for 2011 but palatable for 2012. That example takes the additional step of showing that even from year to year you cannot assign the same possible error to even the same individual. Just an example, and while I do agree that PECOTA and ZIPS are good approaches to a tool, the cannot, nor will they ever, predict injuries, a contract year or just falling off the shelf.

Being an "Old Coot', you'd think I may be the last one to put my head around SABR.
Not so. I've always seen the discrepancies of "base stats" over 50+ years. Especially
with pitchers CyYoung Awards and fielders Gold Glove Awards. Certainly more, but those
two usually signify elitism.

I take into effect all data that is available. I certainly embrassed Bill James methodology.
Mostly, it makes sence to use all stats, in a subjective manner. My mind stays always open,
but I feel I can judge or note with subjectivity.

There will always be that element of judging worthiness and average, below average and
elitism. Memory/Gut and Data sometimes do conflict, but to rid either is not appropriate.
Mickey Mantle was my hero. I have the memory and data, to prove he was a worthy hero.
The ALWAYS question on "The Mick"? Yeah, but what could he have really accomplished, IF...

@kilroy: yes, thank you. WPA is helpful too, but neither does it tell the whole story, which is why you have to use many stats to evaluate a player.

@RFan: yes to that, too. Do those projection systems offer defensive stats (UZR, DRS, etc.) too?

Tangent: Here are ZIPS and Oliver's projections for Yu for 2012. And here are his stats for the year. Not a terrible job considering they were based on his NBP data. His ERA and BB/9, for the whole season, weren't quite as good as predicted, but his IP and Oliver's prediction for his K/9, were spot on.

"while I do agree that PECOTA and ZIPS are good approaches to a tool, the cannot, nor will they ever, predict injuries, a contract year or just falling off the shelf."

No one can predict these things. If something is totally unpredictable, then you should just ignore it. Of course, we can predict that most teams will suffer some key injuries in any given season (even though we can't predict exactly which players will be injured). So in making projections, it probably helps to notice whether a team is deep or not. But even that's hard to quantify.

"Do those projection systems offer defensive stats (UZR, DRS, etc.) too?"

I'm not sure. I've never seen them. I usually just look at the past 3 seasons of UZR, and assume some age related decline.