What stats do consider the least indicative of a player’s/team’s true ability/value? The two I consider the worst: Football team “D”, because it’s based on yards allowed. not points allowed. Second would be closer ERA in baseball. I don’t care how many runs/9 innings a guy surrenders, I only care about save % and WHIP.
I’ll let others come in and complain of wins and saves. The pitching stat I’ll submit as being overrated is K/9. I think this is a product of needing pitchers to throw 97mph. That’s nice and all, but you don’t need 8+ K/9 to be a great pitcher, just like you don’t need to throw 97mph to be a great pitcher.
Time of possession in American football. A team with a good offense doesn’t need to control the ball that long because they can score very quickly.
By now, it’s practically a cliche among number-minded baseball fans that wins and RBIs are outmoded stats.
Wins are also a silly way to rate quarterbacks, in my book. Over the long run, quarterbacks who “don’t do anything but win” don’t win.
And it drives me boners when a football announcer reads (as if this were interesting, or even important) “You know what/ The New York Giants are 28-1 when Brandon Jacobs carries the ball 30 times.”
(Pssst… I made that up!)
That’s supposed to prove that the Giants need to hand the ball to Jacobs, and “establish the run” early. In reality, it means nothing of the kind. ALL teams have stellar records when their top back has 30 carries
If you have a big lead, you keep handing off the ball to your running backs i nthe second half, to kill the clock. Teams don’t win because they run a lot! They run a lot because they’re winning!
Rating football defense based on points allowed, not points per possession. A team with a slow moving, ball control offense is going to give the opponents fewer possessions per game, thus fewer points, without changing the defense at all. Similarly, rating basketball defense/offense on total points instead of points per possession.
I agree about save percentage, the save is a terrible statistic since it’s so dependent on the exact score during the 9th inning. A guy can get a lot of saves just by being on a team that happened to wind up in a lot of save situations.
I actually like wins as a stat. It tells you a lot of things in one number, and at least gives you a general idea of the pitcher’s overall performance. Sure, you can’t say that 20 wins is better than 17 without more review, but that is true of many statistics.
“Closer ERA” is not a statistic. ERA is, and it’s certainly more important than save% for a closer. At the very least, they’re inextricably tied together. And while I like WHIP a lot, if I had to pick one useful stat to evaluate closers, I’d go with FIP.
Defensive yards allowed in the NFL is worthless.
Huh, and here I thought it was the steering wheel in your pants that did that. 
Teams at the bottom of the standings have a surprising number of save opportunties throughout the season - that’s not the reason saves are terrible.
Nope. Your faulty reasoning for not liking saves would be of great help here, since one of the top 2 or 3 reasons for getting a lot of wins would be to pitch for a team that gets a lot of wins. If Felix Hernandez pitched for the Yankees, he’d probably have won 25 games in 2010. But he only won 13 with the Mariners, in an absolutely dominating season where his offense didn’t do shit for him. People who like wins would tell you that 25 wins is WAAAAAY better for a pitcher than 13, but the only difference here would be his team’s offensive performance, which he has absolutely no control over.
Keeping track of blown saves would make sense. If a guy can’t close the game with a lead, you want to know about that. The save statistic doesn’t tell you anything. One guy can rack up a lot of saves for holding a 3 run lead for one at bat.
Actually the real baseball statheads care most of all about strikeout, walk, and HR percentage for closers, because BABIP (batting average on balls in play) is pretty much out of the control of the pitcher as long as you’re talking about fastball-dependent pitchers. So there’s a large element of luck in WHIP. As far as save % goes, that’s dependent on how the manager decides to use the closer. You can easily inflate closer save % by using him in suboptimal situations like “2 out in the ninth, three-run lead” when it’s better to save him for the following night in case of a more optimal situation.
I’ll disagree with Barkis that K/9 is an overrated stat. The Baseball Prospectus team found it to be very useful simply because the fewer times the ball was in play, the fewer times something good could happen for the batter. But you have to take K/9 in the context of BB/9; guys that both struck out and walked a lot of guys, like Nolan Ryan, were overrated. Guys that strike out a lot of batters and walk very few batters, like Roy Halladay, are headed for the Hall of Fame.
I’ll agree with astorian that wins and RBI are just about useless stats.
Pitcher wins is the most overrated stat in baseball, followed closely by RBI.
SB is also rather overrated unless also accompanied by caught stealing.
Holds are stupid, but not really rated so they can’t be overrated.
Of course, a guy can rack up a lot of blown saves by constantly being called on in nearly impossible situations.
In the end neither saves nor blown saves tell you what you want to know about a relief pitcher - can he get outs without hard contact. The best way to measure that is walk rate, strikeout rate, HR rate, and ERA (or better a defense-independent version of ERA).
It’s not good vs. bad teams, it’s sheer number of opportunities that fluctuates wildly. Who did better, the guy with 40 saves who blew 10, or the guy with 30 saves who blew 2?
I’m allowed to be inconsistent and have faulty reasoning, I’m a baseball fan, dammit! I stand proudly by my ridiculous affection for the Win.
First, I nominate (or second the nomination of) the RBI stat. RBI tells you more about the abilities of players in front of you in the lineup than it does about the player with the RBI. A person could get an impressive number of RBI just by virtue of being behind a good on-base % and base stealer in the lineup, even if his own batting is only mediocre. On the other hand, a hypothetical person batting a perfect 1.000 would have a terrible RBI if he cleaned up for an especially poor lineup of hitters.
Another awful stat is the already-mentioned time of possession in football. You can’t even make a better-than-average guess on the outcome of a game if you only knew that one stat from it. K’s in a baseball game (while also a pretty useless stat) are a better predictor of the outcome than time of possession in a football game.
In the NBA, the last few years they’ve started using +/- points when a player is in the game. I think for non-superstars, those are useless.
I heard the stat came from hockey, which makes more sense as scoring in hockey requires more of a team effort than in basketball. In basketball, any scrub who is out there with a superstar is going to get pretty high +/- ratings, so its pointless for them.
Rebounding margin in basketball is right up there. For that matter, so are PPG and points allowed, because both are so related to pace.
I don’t know how useful it is for individual players, but for various lineups I think it’s very interesting. I like plus-minus and some of the newish scoring and defensive rate stats that you see applied to the NBA these days. The player usage stats, too.
It’s not interesting on it’s own, but it’s interesting in certain contexts. Yes, a team with a high powered offense can score quickly and have a lower time of possession. Similarly, I’ve seen terrible offenses that heavily favor the run have very high ToP and still lose. But, generally speaking, there’s a strong correlation between ToP and winning. A team that has more ToP likely runs more offensive plays, gets more first downs, gets better field possession, and likely eventually scores more.
I too agree that normalizing defensive stats per possession is generally going to give you a better idea. But it still needs an element of number of possessions per game because pace of the game is going to be interesting in analyzing the stats. For instance, imagine you have a defense that gives up roughly the same number of points per possession that the opposing offense scores, but the offense has more possessions per game. I’d think that means they typically run a faster paced offense, may be inclined to wear out the defense and get some advantage.
QB rating. It’s a good general idea, but not much more.
Red-zone TD%. Why not red-zone points? You also score by FGs.
FG%. It’s be more interesting and useful if it was paired with average-distance/FG, average success and average miss.