The Most Important Baseball Stat?

[RaychelSnr]

Operation Sports' Chris Sanner checks in today with an article that is sure to have its fair share of opinions from the baseball world as he talks about the most important stat in baseball.

""For years now a war has been brewing among baseball fans and professionals. What's the best tool to evaluate a baseball player's talent? Your gut and eyes, or a computer munching on the latest data."

[NYJets]

"But before you call me a Sabremetrician, let me first say a few other things about Sabremetrics. I think that the system is fairly black and white, because judging a player by one stat and one stat alone is foolish. Plus it would be silly to say there isn't a human element involved. Some guys do not perform well in big game situations or aren't good clubhouse guys."

Sabremetrics doesn't say you should judge players on one stat alone. And I think most people into stats don't feel that being a good clubhouse guy and intangible stuff like that doesn't exist, just that it is tremendously overvalued.

[catcatch22]

What happened to speed? I would put my lineup this way.

Top - Speed
Upper Middle - On base
Lower Middle - Power
Bottom - Speed

[JoeRyan33]

If I'm going to use one stat to judge a hitter, it's not any of those in current use. Basically, you must allocate value to each result of a plate appearance. You do that by breaking it down, walks (intentional and unintentional), singles, doubles, triples, home runs. Each has their value to the scoring of a run. I could elaborate on this, but I'd end up writing an article, which I have done many times on this same topic. Good post.

[RaychelSnr]

Only if that top guy is getting on base to use his speed then you have something with speed. However, having guys at the top who get on base (because they see more plate appearances than guys down the lineup) is far more important than speed. Speed would be looked at as a nice bonus, that old tired notion that a speedy guy at the top of the lineup is very much false in my eyes. You want guys that can get on base first and foremost. Of course, like I said in the article there WILL be dissenting opinions to my theory

[krishna]

Im a bit of a statistics God.

If I had to choose one, I'd choose EqA.

http://shades-of-wrigley.blogspot.com/2008/02/equivalent-average-unmasked.html

[krishna]

but only looking at traditional stats, some combination of obp and slg is the best way to go. Somewhere around k*obp + slg where 1 < k < 2.

[bcruise]

In real life I'd agree that OBP is very important, but in video games I think it's a little different. Gamers are generally a lot less patient and don't walk (or the games don't allow them to), which takes a good-sized chunk out of the OBP formula. I'd say power (or statistically Slugging percentage) reigns supreme in a video game environment - a good player can make up for a hitter's other deficiencies simply by making a good swing. But power is one thing they have less control over - a weak-hitting AAAA guy isn't going to hit HR's no matter how good your timing is.

JMO.

[Cletus]

Well in real life OBP is a good starting point, but having a high OBP guy batting 4th might not be what you want. Sure he will get on base, but what if you need 3 rbi's and there is 3 runners on? Would you rather the guy walk or hit the 3 run homerun? I would take the guy who could hit me a 3 run homer because the more batters you bring up in a clutch situation the more they are destined to fail. The more I think about it the more it is personal and personnel. I would rather have one guy that I trust in to bring in runs than spread it around. Walks are good and fine if you're a 1 or 2 hitter, but sometimes you need a guy to drive in runs. I think RBI's are the most important stat, or more specifically runners left on. It just depends on how you want your lineup and what order they are batting in.

[JoeRyan33]

OBP only tells you how much a hitter gets on base. It tells you nothing about how he got there. Any statistic that rates a walk and a home run identically is poor measure of a hitter. You can't walk around the bases, someone has to drive runners in. Also, take a look at your average singles hitter and you'll see the lack of actual run production.

Bank on your extra base hits. Doubles hitters keep rallies alive.

SLG tells you a lot more about a hitter, and correlates better with actual production. But then I prefer to use the raw numbers to make assessments, giving each possible result their own value.

The question becomes, what is the value of a a double, a home run, etc?

[Bigtonyclark]

Re: The Most Important Baseball Stat?

These two groups, for the most part, make up a club's philosophy when it comes to evaluation. You rarely find a team that does things as a combination, which to the casual observer, would seem to be the smart thing to do.

I do not agree with that all. It'd be hard to find a team in baseball that does not integrate both scouts and statistical analysis. Really hard. Maybe the White Sox.

Also, there is no one single most important statistic, though I do think every conversation about it should begin with OBP and slugging percentage, since they correlate directly with winning. I like to use the slash stats, along with OPS+, EqA, wOBA, and some others. All of those have flaws, like every statistic, but when used in unison, you can really get a good picture of a player's offensive value.

Same goes for pitching. ERA has a lot of flaws most people don't tend to see, but I still like it, especially when looking at a player's career as a whole. If you like ERA, then take it up a notch, and use park-adjusted ERA: ERA+. But, you still can't just look at ERA+ without looking at a pitcher's peripherals. I really don't like ERA when looking at relievers.

There are some really neat things been done with fielding analysis at the moment, too.

[JoeRyan33]

Re: The Most Important Baseball Stat?

Measuring pitchers performance should fall into two main categories:

1. The direct opposite of how one should measure hitters - tendency to give up certain types of hits, equating a value with each hit. This would do a better job of replacing ERA as a measuring tool. It does rely on defense to some degree, which is why #2 is necessary.

2. Strikeout/walk ratio.

Statistics won't tell you everything. Lewis' Moneyball stressed the idea of "not trusting your eyes". I couldn't disagree more with this idea. Baseball is a game of physical competition. One must assess a hitter/pitcher's physical abilities and tendencies in assessing their overall performance.

Don't overlook the importance of good coaching. Again, this flies in the face of the Moneyball theory that you cannot change a player, that they are what they are.

[catcatch22]

From Lou Brock to Ricky Henderson to Vince Coleman to Jose Reyes. Speed can change the game because those players when on can turn a single into a triple. They make pitchers stay in the stretch and slow the game down completely making the pitchers focus on them and not the hitter. Lots of Speed in a lineup can kill a pitcher who can't work well in the stretch. Yes they have to get on but when on they change the game. Mariano with Dave Roberts on first stealing second comes to mind in that classic Red Sox - Yankees series where the Yanks choked it up and lost 4 straight.

[BlyGilmore]

Re: The Most Important Baseball Stat?

I agree Joe. That's part of the problem i have with strict SABR folks - they completely take the humans out of the equation and instead see everything as numbers.

sure - lineups don't matter if your game plan is completely devoid of any tactics or strategy. but add those and its a different story.

The teams that are going to be successful are the teams that marry the traditional points of view with the SABR points of view. In a lot of ways this is what the Red Sox have done - taking some of the SABR teachings but applying them to more traditional beliefs.

Personally I think that's why for all their success the A's never did much in the playoffs. Their entire team was built almost like a softball team, with players who had no real skills other than being patient, working the count and having some pop in their bats.

Over a season, everything evened out and more times than not this was an advantage. But when you get into the playoffs you don't have the luxury of time. You don't always have the ability to just wait until next inning - sometimes the difference between winning and losing is manufacturing runs.

Take Game 4 in the 2004 ALCS. A strict SABR team wouldn't have won that game. Roberts never would have stolen second because stats tell you its not a safe gamble - he's more likely to score not stealing over time than he is by stealing and risking getting thrown out. Hell - a player like Roberts might not have even been on that team.

[dkgojackets]

Re: The Most Important Baseball Stat?

Operation Sports' Chris Sanner checks in today with an article that is sure to have its fair share of opinions from the baseball world as he talks about the most important stat in baseball.

Depends on what you are looking for.

Best pure hitter? OPS+

Most valuable offensive player? VORP easily.