Analyze This

I am a statistics geek. There, I said it, it's out there for everyone to see. I see PPG and OBP in my sleep. I'm obsessed with analyzing players through statistics. The sports world is a-changin, and if you want to keep up, you better know all about B/40 (Blocks per 40 minutes), YAC (Yards After Contact), OPS (On-base Plus Slugging), BABiP (Batting Average on Balls in Play), and eFG% (Effective Field Goal Percentage). So, with that in mind, I'm going to look over some of the common statistics in sports and find out what the strengths and weaknesses of those particular statistics are. In addition, I'll look at some of the less common statistics that you may not have heard of. Due to the vast amount of statistics there are to look at, I think I'll probably make this a running thing where I'll do a post every week or so with more statistics. That is, unless people think this is a stupid idea, then I won't do it and just go cry in my closet instead, so let me know what you think. Anyway, for this time I'm just going to look over some basic, commonly used statistics.

BA (Batting Average)

What Is It?
Batting Average is a baseball statistic that measures the ratio of hits to at bats.

How Is It Useful?
BA has long been the standard measurement of a hitter's performance, and it makes sense. The more hits a player gets when he's up to bat, the better a hitter he is, right? In general, it usually is a pretty good predictor of a how good a hitter is, with a BA over .300 being pretty good, over .350 being top-notch, under .250 being pretty bad, and under .200 (the Mendoza line, named for famously poor-hitting SS Mario Mendoza) being absolutely terrible.

What Are Its Flaws?
BA really has 2 major flaws. The first flaw is that it doesn't account for walks. For example, when evaluating a leadoff hitter, whose purpose is to get on base any way possible, BA tells you that Willy Taveras (.320 BA) and Dustin Pedroia (.322) are basically equal. However, Pedroia has drawn a walk 13 more times in 30 fewer ABs, meaning he gets on base at a significantly higher percentage than Taveras. The second problem is that BA treats all hits as equal, whether it's a bunt single or a home run. Lets take a look at the case of Prince Fielder (.278) vs. Juan Pierre (.280). Again, very similar batting averages. However, Fielder has 58 EBH (Extra Base Hits, i.e. doubles, triples, HRs) to Pierre's 17 EBH. Obviously, Fielder's power makes him the more effective hitter, but BA doesn't show that. The answer to these two problems, of course, is to use BA in conjunction with OBP (On Base Percentage) and SLG (Slugging Percentage) in order to analyze a player.

SB (Stolen Bases)

What Is It?
Pretty obvious, it's the number of bases a player has stolen.

How Is It Useful?
Well, it tells you which guys are speedy on the basepaths. Guys with lots of stolen bases usually make better top-of-the-order guys (if they can couple it with a decent OBP), and they can rattle pitchers when they're on base simply because of the chance they might steal.

What Are Its Flaws?
Stolen bases is kind of a sneaky statistic in that it looks very simple at first, but has a lot more flaws than it appears. The first flaw of the SB statistic is that if a player who isn't a good baserunner tries to steal a lot, he'll probably get a decent number of steals, but he'll also get caught a lot. For instance, would you rather have Juan Pierre (33 SB in 42 attempts) or Shane Victorino (27 SB in 29 attempts)? The best solution is to look at SB% (Stolen Base Percentage, SB/(SB+CS)) to see how useful a player is when stealing bases. Keep in mind, however, that an out is far more important than an advance from first to second, so the percentage needs to be much higher than 50% for a basestealer to actually be useful. In fact, statistics have shown that on average, if you're successful less than 75% of the time, you're actually hurting your team more than helping it.

BPG (Blocks Per Game)

What Is It?
A measurement for the number of shots a basketball player blocks on average each game (Total Blocks/Games Played)

How Is It Useful?
A high BPG average is usually indicative of a good post defender. An average above 2.5 is usually very good for a post player, and anything above the low-1s is good for a guard.

What Are Its Flaws
As with all per game statistics in basketball, it suffers from not taking into account the number of minutes played. For example, Jermaine O'Neal averaged 2.6 blocks last year compared to Alonzo Mourning's 2.3. Thos look pretty comparable, with O'Neal having the slight edge. However, O'Neal averaged 15 minutes more per game than Mourning did, which puts Mourning's per minute average at a whole different level. The other flaw with blocks is that, well, it's just not a very good measure of whether you're a good defender or not. Players who block a lot of shots often do so by losing rebounding position to try for a block, and they often get in foul trouble by attempting blocks. Also, very good shot blockers tend to see their block numbers drop as players realize how good they are and decide to try and avoid them. If you watched an Ohio State game last year, you could see this happen as teams were very hesitant to go into the post against Greg Oden. So, what do we make of the blocks statistic then? Well, whenever evaluating a player by their blocked shots, be sure to incorporate fouls and rebounds into your equation. If a player has a high block per foul rate and they still rebound the defensive glass well, then they're probably a pretty good post defender. It's a little simpler for guards, since any shotblocking you get from them is an added bonus. If you see a guard/forward with high block numbers (like Gerald Wallace), that probably means they're a pretty strong defender.

FG% (Field Goal Percentage)

What Is It?
A basketball measurement of how many field goals (any shot that isn't a free throw) you make per field goals you attempt (FGM/FGA)

How Is It Useful?
It's very useful when trying to analyze a player's effectiveness as a scorer. Players who have a higher FG% are typically better shooters and therefore more efficient scorers. For example, Kobe Bryant has averaged 24.6 points per game in his career to Allen Iverson's 27.9 . However, Kobe's FG% is 30 points higher than Iverson's, indicating that he is the more efficient scorer of the two.

What Are Its Flaws?
Mainly that it doesn't account for the difference between post players and guards. Post players regularly put up FG%s of over 60%, since the vast majority of their shots are right around the rim. Guards, on the other hand, are much more likely to be launching more difficult shots from further out. One answer to this is eFG% (Effective Field Goal Percentage), which weighs three-pointers as 50% more important than 2-pointers [(FG + 0.5*3P) / FGA]. eFG% is in general a much better all-around measurement than straight-up FG%, and generally what I use when analyzing players.

Sacks

What Is It?
A football measurement for how many times a player tackles an opposing quarterback behind the line of scrimmage.

How Is It Useful?
It's a good way to measure how well a player (usually a defensive end) can get into the backfield and make plays. Double-digit sacks are very good for a defensive end/outside linebacker, and anything in the upper single-digits is pretty good for any other defensive position.

What Are Its Flaws?
Hmmm, where to start? Well, first of all, it isn't really a very good indicator of whether or not a player is a good defender, since many rush ends concentrate on the sack at the expense of run defense. In general, TFL (Tackles For Loss) is a better statistic, and total tackles is also something you should take into account. For example, Mark Anderson of the Bears had 12 sacks, but only 28 total tackles. Compare that to Adalius Thomas, who had 11 sacks but 83 total tackles. Who would you rather have? The other thing to note is that sacks may also be the result of a strong secondary (good coverage resulting in the QB hanging on to the ball too long), or the result of a defensive scheme that emphasizes blitzing.

There you go, the first edition of Analyze This, maybe the first of many. Kudos to the first person who can name me all the players executing each statistic in the pictures.