Baseball Statistics Glossary

[Skipper Steve]

As requested in one of our threads, I've added a Stat Glossary to NetShrine. It can be found at:

http://www.netshrine.com/statglossary.html

Please check it out - - if you noticed anything that I got wrong (very possible! I did this quickly) or if you have something that you think should be added, post it in this thread. Thanks!

[Skipper Steve]

BTW, yes, I know that not every formula is shown and some of the definitions are broad - - - file this effort under "something is better than nothing."

[Skipper Steve]

No comments?

[BuzzBuzzard]

I didn't get very far before being presented with a question. I had never really looked at AERA before. Based upon your example, does that imply that in 1999, Pedro was 2.4 times better than average? If so, what exactly does that tell you?

Nice work on the compilation of information. I am sure to have more questions as I work through it.

[Skipper Steve]

Originally posted by BuzzBuzzard
Based upon your example, does that imply that in 1999, Pedro was 2.4 times better than average? If so, what exactly does that tell you?

Nice work on the compilation of information. I am sure to have more questions as I work through it.

Thanks Buzz - appreciate the look and the kind words.

Pedro? Hmmm - - let me put it this way, if his 1999 season was a "test" where a grade of 65 was passing, he got a 156. Not too shabby.

[timconnelly]

I thought I might give a very detailed explanation on how some of the sabermetric formulas work and more importantly, why they work.

Runs Created is a formula that Bill James worked out a number of years ago. There are a number of variations to it as it can be fine tuned to adjust for some of the extra statistical information that is not available from one year to another. But it is a far more reliable method of measuring offensive productivity than batting average, slugging average, on base average, or any other regular statistical method using even the simplest method of figuring it.

Before I give you the formula, the weakness of batting average is that it fails to take into account the amount of power that a hitter has, plus it ignores walks. Slugging average measures power but fails to address the batters ability to get on base. And on base average measures the ability to get on base by looking at walks but fails to look at power.

The purpose of the batter is to create runs for his team. So why won't RBI and runs scored adequately measure that? The answer is that both stats are team dependent. If you have a lot of runners on base, your chances for a high RBI total is greatly enhanced. With good batters batting behind you, your chances of scoring runs is greatly enhanced. So you are only measuring the productivity of the hitter a little bit .

But there is still another problem with RBI and runs scored: They fail to measure important offensive contributions that often lead to runs. The single that moves the runner from 1st to 3rd gets you nothing. The walk that loads the bases will only count if you happen to score. The problem with that is that these things have offensive value that simply isn't measured.

Runs created solves this whole problem as it looks at the 3 most important offensive things in just the right proportion: 1) Times reached base 2) The amount of power that the batter hit with 3) The number of opportunities that the batter had.

1) Take the number of times the batter got on base. So you are counting walks, HBP, and base hits.
2) Multiply that by the number of total bases. So the more power the batter had, the higher this total will be.
3) Take this number and divide it by plate appearances. This is at bats plus walks plus HBP.

A more complex Runs Created formula takes stolen bases, caught stealing, grounded into double plays, and other similar things, and factors it in.

This will give you an accurate estimation of the number of runs that this would create for a team! How do we know that it works? We can use the formula on a whole team and it will accurately estimate the number of runs that a team would score from that set of offensive factors. We are only breaking it down individually when we turn from a team to a player.

Ok, so now we have the number of runs the player created: Runs Created per game is a way of breaking that down to approximate how many runs that would add up to per 27 outs.

So you take the # of at bats and subtract base hits: That will give you the number of outs. (Again I am simplifying it, you can add grounded into double plays and caught stealing when available)

You then take the total # of outs and divide it by 27. That gives you the figure for exactly how many games the players outs added up to.

Divide the runs created by the total above and that will give you runs created per 27 outs.

There are 2 ways of looking at this: One is that it is what a team made up of players with exactly this offensive productivity would score if the entire line up was this player. But it is actually what that player is producing spread around through the entire season! Barry Bonds was actually creating more than 18 runs for his team for every 27 outs he was making!

But this still has the problem of measuring a raw total that fails to take into account the value those numbers have. Averaging 6 runs per game when the league average is 4 runs a game makes you incredibly valuable, average 6 runs a game when the league average is 6 runs makes you average.

A famous mathematician named Pythagoras came up with a famous formula. Amazingly, it works incredibly well with wins and losses relative to runs scored and runs allowed.

You take the runs created and square it and then divide this by (runs created squared plus the league average of runs squared)
Actually, Lee uses 1.83 rather than squared.

This will give you the offensive winning percentage. It is an accurate reflection of what that set of offensive contributions would mean to a team in terms of helping them to win.

That is pretty long and I'm not sure anybody will get through this. But maybe that'll help somebody a bit.

I don't happen to see "WHIP" that you guys use all the time. I don't know all of Lee's terms that are used but are all of them in that html file?

For terms like "ASB", is that your own? The one about "BK" seemed like it was straight out of the MLB terms (see the "Lingo" thread in the "History" section).

[Skipper Steve]

None of these are my own.

"WHIP" is more of a rotisserie term than a sabermetric. It's walks (W) plus hits (H) divided by innings pitched (IP). The closest thing in the glossary to it is:

Baserunners Per Nine Innings [BR/9]

The total number of batters reaching base against a pitcher divided by the number of innings pitched and multiplied by nine. It measures how many batters reach base on a per game basis against a pitcher.

A league best figure for this category is typically between 9 and 10.

Presuming this is an appropriate place to discuss this (if not, please let me know), if a pitcher "scatters" 9 hits over 7 innings, gave up 5 walks but the opposing team never took advantage of this, would the pitcher who never got hurt by these hits but got the win (perhaps a 4-2 score) be seen as less capable due to a higher stat?

In both WHIP and BR/9, a pitcher's measured on how many people got on base (measured in one way or another). I guess this is like the WHIP vs ERA, but if the pitcher never let those runners hurt him, especially if no one ever got past 2B or even got that far, I'd say this pitcher controlled the situation.

I'm trying to find out all of them and their definitions. Is this available here, or are they on Lee's site?

[Skipper Steve]

Originally posted by Yogi#8Fan
seen as less capable due to a higher stat?

I would say that a pitcher with a high WHIP would be viewed as not having great stuff.

As far as Lee's stats, if it's not in the Stat Glossary here, let me know and I will get you the answer.

I didn't see WHIP in your glossary. If you can let me know where it and all the other sabermetric terms and definitions are, please do so.

[Skipper Steve]

Originally posted by Yogi#8Fan
I didn't see WHIP in your glossary. If you can let me know where it and all the other sabermetric terms and definitions are, please do so.

Re-read this entire thread.

. I simply asked where terms like "WHIP" are all found. You'd mentioned it being a rotisserie term. All I asked for was a link, not having to re-read the entire thread. Link, please, if possible.

[Skipper Steve]

Not possible Yog.