The Simple Questions Thread

[Seamhead]

Have any questions that aren't relevant in any future/present thread? Then ask them here. Someone will either answer them, or better yet, link you somewhere where your question will be answered.

Oh, and there's already a good mailbag here:

http://www.tangotiger.net/wiki/index.php?title=Mailbags

Though, this is more advanced stuff.

[Jilly Bohnson]

Yeah, we're going to make this the simple question thread. What's OPS? What is UZR? Quick and easy stuff go in here, harder stuff can get it's own topic.

[Seamhead]

Agreed. Feel free to change the title to "Simple Questions Thread".

[Pavelb1]

Woba please. I consider myself fairly smart, but I can't wrap my head around it as described in The Book or by Rob Neyer, so in simple terms please.

[poodski]

I will start it off.

How is UZR/150 figured?

Because it sure as hell isnt current UZR figured over 150 games. At least not on fangraphs it isnt.

For example. Last year the Royals had a positive UZR, but a negative UZR/150. Or like this year Nick Swisher has a -.3 UZR in RF, but a 2.6 UZR/150.

[Jilly Bohnson]

Woba please. I consider myself fairly smart, but I can't wrap my head around it as described in The Book or by Rob Neyer, so in simple terms please.

Okay, so basically it goes like this. You know how stuff like a single, a double, a walk, and a homerun, etc. each have a run expectancy attached to them? That is, for example, a single is worth let's say half a run, meaning for every single that is hit it generally leads to half a run scored on average. Well what wOBA does, is it takes all of what a player did, all their singles, doubles, homers, etc., attaches their run expectancy for each of those events, and then multiplies them by certain mulitpliers that make wOBA on the same scale as OBP to make it easier to understand. So for example about .335 would be average, .350 would be pretty good, and .400 would f'ing awesome.

[Zep]

wOBA's actual formula looks like this in case you're ever bored and want to calculate for yourself:

wOBA = (0.72xNIBB + 0.75xHBP + 0.90x1B + 0.92xRBOE + 1.24x2B + 1.56x3B +1.95xHR) / PA

That is, (.72 x Non-Intentional Bases on balls + .75 x Hit-by-Pitch + .90 x Singles + .92 x Reached Base on Error + 1.24 x Doubles + 1.56 x Triples + 1.95 x Home Runs)/Total Plate appearances.

wOBA (from my understanding) is basically OBP, but each specific event that leads to a batter reaching base is given a specific value, or weight (a triple is worth more than a double, a double is worth more than a walk, etc.).

[Jilly Bohnson]

I will start it off.

How is UZR/150 figured?

Because it sure as hell isnt current UZR figured over 150 games. At least not on fangraphs it isnt.

For example. Last year the Royals had a positive UZR, but a negative UZR/150. Or like this year Nick Swisher has a -.3 UZR in RF, but a 2.6 UZR/150.

That's a really good question. I'm not sure, the only thing I'm thinking is maybe it is influenced by what they have done in the past? So Nick's got a -.3, but in the past he's done quite well in RF, a 9.3 in his career. So perhaps for partial seasons, a player's UZR/150 has a component of what would be expected given their previous performance?

[Pavelb1]

Okay, so basically it goes like this. You know how stuff like a single, a double, a walk, and a homerun, etc. each have a run expectancy attached to them? That is, for example, a single is worth let's say half a run, meaning for every single that is hit it generally leads to half a run scored on average. Well what wOBA does, is it takes all of what a player did, all their singles, doubles, homers, etc., attaches their run expectancy for each of those events, and then multiplies them by certain mulitpliers that make wOBA on the same scale as OBP to make it easier to understand. So for example about .335 would be average, .350 would be pretty good, and .400 would f'ing awesome.

ahhhh....got it. Thanks!

[homie564]

Ok Sabermetrics + me dont click.. except for the simple things... heres a question i want to know... what sabermetric stat is best to determine a players overall value and skill to a team?

[-Lavigne43-]

Ok Sabermetrics + me dont click.. except for the simple things... heres a question i want to know... what sabermetric stat is best to determine a players overall value and skill to a team?

I would look at a multitude of stats to determine a players value. The best available offensive stat in my opinion is wOBA with UZR being the best available defensive metric. For pitching I like to look at FIP, BABIP and their K/BB.

VORP tries to determine a players value to a team but it does not take into account defense.

There is also WAR which weights wOBA and UZR in a formula to calculate how many wins a position player adds to a team. There is also a formula to calculate WAR for pitchers. I am not too familiar with this stat though

[Driven]

Yeah, we're going to make this the simple question thread. What's OPS? What is UZR? Quick and easy stuff go in here, harder stuff can get it's own topic.

Maybe a stickied thread of a glossary of terms would be helpful?

EDIT: Nevermind you guys are way ahead of me.

[Rylinkus]

wOBA's actual formula looks like this in case you're ever bored and want to calculate for yourself:



That is, (.72 x Non-Intentional Bases on balls + .75 x Hit-by-Pitch + .90 x Singles + .92 x Reached Base on Error + 1.24 x Doubles + 1.56 x Triples + 1.95 x Home Runs)/Total Plate appearances.

wOBA (from my understanding) is basically OBP, but each specific event that leads to a batter reaching base is given a specific value, or weight (a triple is worth more than a double, a double is worth more than a walk, etc.).

Is that the fangraph formula or the SC one?

Actually, I'll expand. Fangraphs uses a different wOBA formula than Statcorner. Which one is the better one in people's opinions? And why? Perhaps this isn't a simple question and warrants it's own thread.

It does bring up the issue with Saber stuff. .BA is, well .BA. Everyone computes it the same way. The fact that wOBA fluctuates in the way it's acquired probably turns some people off to the concept.

[Gigantes4Life]

I will start it off.

How is UZR/150 figured?

Because it sure as hell isnt current UZR figured over 150 games. At least not on fangraphs it isnt.

For example. Last year the Royals had a positive UZR, but a negative UZR/150. Or like this year Nick Swisher has a -.3 UZR in RF, but a 2.6 UZR/150.

It is UZR over 150 games. But it's 150 defensive games. It's based upon the expected chances at their relative position. Meaning if a 2B plays 9 innings but gets no balls hit to him, he really didn't even play a defensive game.

http://www.fangraphs.com/statss.aspx...on=OF#fielding

Right at the beginning of the right columns, it says "DG," that of course stands for Defensive Games.

The reason the UZR/150 can come out so weird like that is because they simply don't extrapolate UZR. They extrapolate errors, range, arm, etc.

Ok Sabermetrics + me dont click.. except for the simple things... heres a question i want to know... what sabermetric stat is best to determine a players overall value and skill to a team?

Lavigne pretty much had it.

WAR is the best stat in measuring a player's overall value to his team at the moment. Of course, you need to use the best stats in order to measure it correctly.

A combination of UZR and +/- (you must pay for this though) for defense (some use the Fans' Scouting Report, probably go over this later).

wOBA is the way to go for offense.

And FIP or tRA for pitchers. Some like to use ERA with FIP, but tRA is probably the way to go as well.

The whole concept of WAR takes a while to explain, and honestly probably deserves its own thread. I believe seamhead made a post about it in the MLB forum, maybe he could repost it?

Is that the fangraph formula or the SC one?

Actually, I'll expand. Fangraphs uses a different wOBA formula than Statcorner. Which one is the better one in people's opinions? And why? Perhaps this isn't a simple question and warrants it's own thread.

It does bring up the issue with Saber stuff. .BA is, well .BA. Everyone computes it the same way. The fact that wOBA fluctuates in the way it's acquired probably turns some people off to the concept.

I believe the major difference is that fangraphs incorporates stolen bases into its equation. Everything else is just alterations of the historical data. In the longrun, the difference shouldn't be large. I prefer SC simply because of wOBA*.

[Seamhead]

Okay, so basically it goes like this. You know how stuff like a single, a double, a walk, and a homerun, etc. each have a run expectancy attached to them? That is, for example, a single is worth let's say half a run, meaning for every single that is hit it generally leads to half a run scored on average. Well what wOBA does, is it takes all of what a player did, all their singles, doubles, homers, etc., attaches their run expectancy for each of those events, and then multiplies them by certain mulitpliers that make wOBA on the same scale as OBP to make it easier to understand. So for example about .335 would be average, .350 would be pretty good, and .400 would f'ing awesome.

And run expectancy = move over value (slugging in a way) + get on-base value (OBP).

It's context-neutral, meaning it doesn't take into account any type of situation. It's just the marginal value of the event given a neutral environment.