Daniel Moroz: Understanding what stats mean is at the heart of diamond analysis

I started writing about the Orioles way back before the 2008 season (at the suggestion of my friend, who isn't a fan and so was probably tired of listening to me talk about baseball), and it's been a fun and rewarding experience. No way did I ever expect to be invited to participate in something like this, but I'm grateful to MASNsports.com for the opportunity.

Since my analysis tends to come more from a statistical perspective, I thought by way of an introduction, I'd discuss that a bit. Baseball is obviously a game that has a great deal of statistical depth - everything from the runs scored versus runs allowed that tells us who won the game, to a particular runner's success rate at stealing third-base (I really hope Brian Roberts comes back productive and not just healthy). One can get a great deal of enjoyment from seeing a majestic home run or (my favorite) a knee-buckling curveball, but to write about baseball and read about baseball, you kind of have to go to the numbers at some point. Once we've established that, we're just haggling over which stats to use.

I'm not going to - nor could I - go through the entire universe of baseball statistics, but just a few basics is a fine start.

Pitchers:

Wins and losses are the name of the game in baseball, but that's on a team level. Win-loss record is not a good way to judge individual pitchers - using it to say whether a pitcher's good or not doesn't work all that well. Why? Because there are too many conflating factors. What goes into whether a pitcher gets a W in a game? Well, there's the pitcher, of course. But also the defense, the offense, the bullpen and the manager. A pitcher can do his job, but if even one aspect that's totally out of his control doesn't cooperate, he might not get the win - or could even be saddled with a loss. Sometimes a pitcher will get the win but give up eight runs. Sometimes he'll give up one run - unearned at that - and get the loss. I think everyone tends to get this on a case-by-case basis, but it doesn't always carry over in general. Is the 15-5 pitcher better than the 5-15 pitcher? We don't really know.

ERA is better, but has a similar issue - defense. ERA is supposed to tell you how many (earned) runs a pitcher allows - and the construction is usually along the lines of "pitcher X gave up five runs" - but really it's the entire run prevention unit (pitcher + defense) that allowed those runs. Errors are a way of trying to suss out what's the pitcher's responsibility and what's on the defense, but it doesn't completely work. An error could result in a run when it should have been an out, but a ball that got past a third baseman with no range could result in a run when it should have been an out, too. One hurts the ERA, and the other doesn't. But we do consider the second one to be the pitcher's fault.

(You know, if Mark Reynolds wasn't be making so many errors this year, I think fans wouldn't realize he's not a good third baseman. But even if he only had one or two errors, he still wouldn't be a plus out there.)

What I tend to go by is what a pitcher controls more himself: strikeouts, walks (not intentional ones, which are the purview of the manager, but plus hit batters) and home runs. A pitcher with a lot of Ks and few walks or home runs is doing well, even if he's 5-15 with a 5.00 ERA. There are, of course, exceptions to the rule. But it holds strong the vast majority of the time at the major league level. And handily enough, there's a stat that's on the ERA scale that just looks at these factors; Fielding Independent Pitching (FIP). You can check out more about FIP here, but even just looking at strikeouts and walks by themselves gets you a lot of the way there.

The next step is to figure out why he's 5-15 with a 5.00 ERA, but in judging his performance individually I'd look at what he himself has done (well, along with the batters - and, I suppose, the ballpark, weather, etc.) rather than what he's done in a team context.

Batters:

The standard triple crown categories are batting average, home runs and RBIs, and that's giving these stats more weight than they probably deserve. Taking the last first, RBIs suffer from the same kind of problem as ERA - it tells us much about more than just the one hitter. RBIs are very dependent on opportunities; if a batter comes up with men on base all the time, he's bound to knock a fair number in even if he's not a particularly good hitter. On the other hand, Babe Ruth batting after Luis Hernandez, Cesar Izturis and Brandon Fahey is probably going to have a similar number in the RBIs column as he does in the home run column. Many batters with a lot of RBIs are good, but couldn't that be because good hitters get put into the middle of the lineup where they get a lot of RBI opportunities? On top of that, is a batter who flies out to the wall with the bases loaded and later hits a solo home run actually a worse hitter than the guy with the grand slam and the flyout with the bases empty?

Then there's batting average, which isn't bad in conjunction with other stuff, but doesn't give you a ton of information by itself. What significance does how often a batter gets a hit in plate appearances in which he doesn't walk or hit a sac fly have? The important things along these lines are actually how often a batter doesn't make an out overall (that is, on-base percentage) and the quality of the hits (that is, slugging percentage). A .290 batting average doesn't mean a whole lot if it's nothing but singles without any walks added on, but a .230 batting average can be just fine if the hitter is good at avoiding outs otherwise (by drawing lots of walks) and makes his hits count (extra bases - and it doesn't have to just be homers).

That's why the the triple slash line - batting average/on-base percentage/slugging percentage - is the way to go, I think. Not only does it give you a good picture of the quality of the hitter himself, but (unlike single measures such as OPS - on-base plus slugging - which isn't terrible for a quick look, but adds plate appearances and at-bats, two stats that are on different scales) it also provides an idea of the type of hitter we're talking about.

There are a lot of stats I use that go past these (and by no means have I done these justice), but the step toward statistics that actually tell you about what you want to know (and the question is usually, "How good of a season is this guy having?") instead of mucking things up with unrelated factors is an important one.

Daniel Moroz blogs about the Orioles for Camden Crazies and joins MASNsports.com as part of our season-long initiative to welcome guest bloggers to our site. All opinions expressed are those of the guest bloggers, who are not employed by MASNsports.com but are just as passionate about their baseball as our roster of writers.

blog comments powered by Disqus