What is it good for?

Absolutely not evaluating the value of relievers

I would argue it excellent for that very thing. "Closer" is the most overvalued position in the game.

I would agree with the second part. The issue is that WAR is a compiling statistic. Relievers only pitch ~70 innings per season, so an elite reliever can often fail to accumulate more WAR than a mediocre starter.

But that reflects- correctly- the limited impact of the reliever.

Until you see a bad one! A bad closer can just kill a season.

(Not sure if you mean closer or relievers in general.)

Is that perception or reality though? Of course when you lose a game in the ninth inning it stands out more than when you blow it in the fifth.

I agree that the last three outs seem to be difficult. And it appears that certain guys do not have the makeup to consistently and successfully get those three outs. But whether my perception is based upon small samples and anecdotes, I can't be certain.

But that reflects- correctly- the limited impact of the reliever.

Not all innings pitched are equally valuable. Mike Leake getting three outs with a 5 run lead isn't as important as Josh Hader coming in to strike out the side with the bases loaded. Yet, both go down as scoreless innings.

I would agree with the second part. The issue is that WAR is a compiling statistic. Relievers only pitch ~70 innings per season, so an elite reliever can often fail to accumulate more WAR than a mediocre starter.

But that reflects- correctly- the limited impact of the reliever.

Adjusted WPA/out would do it, but that would lose too many people mathematically. But even then, you come in and pitch a clean 9th in three run game, you've added a whole 0.03 WPA over 3 outs, or 0.01/out.

If we're talking about changes to the game, one thing that I believe is a positive development over the past couple seasons is more first pitch swinging. The first pitch is often the best pitch of an at-bat and if you're taking it in the interest of merely "seeing pitches" you're helping create a boring game dominated by walks and strikeouts. It got to a point where pitchers could impudently throw a get-me-over fastball on the first pitch and start many counts 0-1. The increase in first pitch swings is clearly an adjustment to this.

Another thing I think is interesting is that all these teams have their detailed spray charts and proprietary data that informs their shifts, yet BABIP remains right at .300 no matter what the defense does. Tip of the cap to Voros McCracken.

WAR has always been a flawed stat designed from the beginning to find the answers that were wanted. It starts off with a bunch of assumptions that fall apart so much people have to say "Well, it's only good for comparing players who play the same position" even though a positional adjustment is one of the whole points for this. It then throws in defense without adjusting for how little defense matters at some positions.

It is an easily digestible stat for internet arguments and sports radio because it puts a number value on a player so you can say "No, Player X is better because he has a WAR of 3.1 and Player Y has a WAR of 2.0!" without having to think about anything else in that equation.

the moment a "should have" factor is included into an equation, it no longer becomes scientific

If we're talking about changes to the game, one thing that I believe is a positive development over the past couple seasons is more first pitch swinging.

whats BABIP mean?

I'm still stuck at BA, OBP, and SLG, only.

Batting Average for Balls In Play.

Even though "batting" is in the title, it's really a pitching stat. The theory is that a pitcher can only control three outcomes: strikeouts, walks, and home runs. All other balls that are put in play are considered good luck (out) or bad luck (hit) and not a function of a pitcher's skill. This is because someone discovered that all pitchers at the MLB level seem to have about the same BABIP. Assuming that's true, then all you would need to do to judge a pitcher is look at his strikeouts, walks, and home runs against. You can also look at a pitcher's recent BABIP to see if he's been lucky (his BABIP is lower than average), or unlucky (BABIP higher than average) because over time it should come back to the average.

Of course, many disagree with the notion that a pitcher has no effect on whether a ball in play is a hit or out. Some pitchers would seem to be good enough to induce weak contact, while others get pounded.

BABIP doesn't work as well as a metric for batters, in that there is more variability depending on the hitting style. For example, Ichiro had a very high BABIP because of the many infield hits he had.

I was never a Cub fan, but Lee Smith was my favorite Cubs player of all time.

When he came in, you knew the game was over for the other team. He threw fastballs past everyone.

Harry would get all worked up about him. It was very exciting to watch as a youngster.