McCovey Chronicles: An SB Nation Community

Navigation: Jump to content areas:

Around SBN: The Definitive Case For Will Muschamp, Part 1 Bar-right-arrows

All Stat Questions Answered

Hb307-1_tiny by marcello on

In another topic, some questions about stats were asked.  I thought it might be a good idea to have an open thread where people can ask any questions they have about stats, sabermetric or otherwise.  I'm hardly the most qualified to answer all questions, but I have a decent grasp of sabermetric stats and I'm hoping/expecting any and all people to contribute to the answering.  So, please, if you've wondered what some of us are referring to in other threads (xFIP, EqA, BABIP, etc.) please don't hesitate to ask.

 

P.S. Let's try to keep this thread relatively non-hostile, if possible.  I know that the subject of sabermetric stats can be a little controversial at times, and I know that I'm hardly the role model in keeping things mellow, but let's try.

 

P.P.S. See how easy it is to put together 75 words?  I'm a terrible writer and I managed it, even without the P.S. and P.P.S.

This FanPost is reader-generated, and it does not necessarily reflect the views of McCovey Chronicles. If the author uses filler to achieve the minimum word requirement, a moderator may edit the FanPost for his or her own amusement.

10 recs | Comment 96 comments

Story-email Email Printer Print

Comments

Display:

Ok, the first question that was asked in the other thread was:

What is the stat that takes fly balls to ground balls into consideration and what does it mean to pitchers?

I assume you’re talking about groundball to flyball ratio. I like it expressed as GB%, although some places will list GB/FB as a ratio. Basically, it is just how often a pitcher induces a groundball. To pull from another great site that offers a good overview of what’s important for pitchers, go here. I’ll pull out (insert Bocock joke here) the relevant part for this question:

GB% (Groundballs per Balls In Play) does a very good job of telling us how often a pitcher induces a groundball. 42% is league average, and anything over 50% is terrific, with the best sinkerball pitchers posting rates in the 60-65% range, while anything below 35% can be a problem if its not offset with a high strikeout rate.

Groundballs are great because they, obviously, never go for homeruns and rarely are even extra base hits. While a groundball will result in a base hit more often than a flyball, it is more than balanced out by how much more often flyballs are doubles, triples, and homers.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 6:02 PM PDT   0 recs

Groundballs are great because they, obviously, never go for homeruns…

Not most of the time, no.

My Dave Righetti is better than your Dave Righetti.

by howtheyscored on Jul 15, 2008 6:10 PM PDT to parent up   0 recs

Yeah, it’s probably happened a couple of times, but it shouldn’t happen without some piss poor defense.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 6:23 PM PDT to parent up   0 recs

Good thread! I’ll answer any question that I can.

Bay City Ball

by xanthan on Jul 15, 2008 6:05 PM PDT   0 recs

And you linked to it above...

But USSM has some fanastic primers on defense and evaluating pitching talent. Great reads.

Bay City Ball

by xanthan on Jul 15, 2008 6:09 PM PDT to parent up   0 recs

It looks like your links got messed up:

Fangraphs
The Hardball Times
Baseball Reference
The Baseball Cube
Baseball Think Factory

And I wanted to add:

First Inning

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 6:14 PM PDT to parent up   0 recs

I FAIL.

First Inning is a great site, too. Good call.

Bay City Ball

by xanthan on Jul 15, 2008 6:16 PM PDT to parent up   0 recs

I got it from the Matt Cain sucks – no wait he’s awesome thread, so I learned it here.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 6:22 PM PDT to parent up   0 recs

That is to say, I learned about it from you.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 6:30 PM PDT to parent up   0 recs

The hardball times and USS Mariner are the two best I think.

BBTF isn’t bad either.

BROCK BOND LIKES HIS MARTINIS PUNCHED IN THE FACE, NOT STIRRED.

by SloIsLonelyForTheOrange on Jul 15, 2008 6:37 PM PDT to parent up   0 recs

Fangraphs

is very useful (if you know what you’re looking for)

BROCK BOND LIKES HIS MARTINIS PUNCHED IN THE FACE, NOT STIRRED.

by SloIsLonelyForTheOrange on Jul 15, 2008 6:38 PM PDT to parent up   0 recs

What’s the deal with DICE? Baseball Mogul uses it as their “saber” stat for pitchers, but I’ve never seen anybody here even sneeze that stat out their butt (which is where sneezes come from). I kind of get the impression that it penalizes fly ball pitchers and values strikeouts… but I really never bothered to find out.

Anybody got the skinny?

My Dave Righetti is better than your Dave Righetti.

by howtheyscored on Jul 15, 2008 6:10 PM PDT   0 recs

Basically, like FIP, DICE estimates a pitcher’s ERA using only three outcomes: BB (and HPB), HR and K. DICE is suppose to be better than ERA because it’s not dependent on defense or luck (like if hits are allowed consecutively or not).

..so allow me to present Tim Lincecum and Matt Cain as two sweet, sweet bottles of warming hooch.

by Cookyman on Jul 15, 2008 6:18 PM PDT to parent up   0 recs

Originally it was thought that pitchers had control over their HR rates. However, it has since been “discovered” that they really have control over their GB rates and that this has influence on their HR rate. So, while DICE is better than ERA, it’s also a bit limited since it incorporates HR’s instead of GB%. Still, it’s a good shorthand stat.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 6:27 PM PDT to parent up   0 recs

I’ve never seen anyone use it though.

..so allow me to present Tim Lincecum and Matt Cain as two sweet, sweet bottles of warming hooch.

by Cookyman on Jul 15, 2008 6:30 PM PDT to parent up   0 recs

It sounds familiar, but I can’t specifically remember anyone using it either. It’s probably because there are better stats out there that are just as easy to get.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 6:31 PM PDT to parent up   0 recs

Awesome

thanks for the info ans links.

I'm young but i didn't fall off the truck yesterday!

by jbowl on Jul 15, 2008 7:32 PM PDT   0 recs

So

correct me if I am wron, an ops+ of 100 would be league average?

I'm young but i didn't fall off the truck yesterday!

by jbowl on Jul 15, 2008 7:37 PM PDT   0 recs

An OPS+ of 100 means that you are exactly a league average hitter. But remember, OPS+ doesn’t adjust for position, only whether or not you are a league average hitter.

So for example, if a hitter has an OPS+ of 150, it means that he’s 50% better than your league average hitter.

Bay City Ball

by xanthan on Jul 15, 2008 8:00 PM PDT to parent up   0 recs

when using ops+

is an ops of 800 with obp 400 and slg 400 the same as an ops with obp 300 and slg 500?

proud father of the newly acquired Brandon Crawford..

by Azmanz on Jul 15, 2008 11:16 PM PDT to parent up   0 recs

Yes, it has the same limitations, in that regard, as OPS.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 11:24 PM PDT to parent up   0 recs

Yeah, that’s one of the big problems with OPS+, but it’s pretty good as a quick and dirty tool.

Matt Cain: He'll save children, but not the Dodger children.

by jponry on Jul 15, 2008 11:28 PM PDT to parent up   0 recs

There’s a that’s what she said joke hiding in there somewhere…

My Dave Righetti is better than your Dave Righetti.

by howtheyscored on Jul 15, 2008 11:55 PM PDT to parent up   0 recs

That’s what she said!

"I been waitin' a long time for this! I been waitin' since the f**kin' amateurs!" --WILL "THE THRILL" CLARK

by Josh from Hollywood on Jul 16, 2008 12:01 AM PDT to parent up   0 recs

Actually no.

Look at the formula for OPS+. It DOESN”T weight OBP and SLG equally. It gives OBP greater weight. Maybe not enough, but it DOES give OBP greater weight.

ZIPS: Milledge: 466 HR, 485 2B, 2282 hits, 278-379-524

by rfloh on Jul 16, 2008 12:28 AM PDT to parent up   0 recs

Huh, I never realized that. Thanks

Neglectful father of David Quinowski

by marcello on Jul 16, 2008 8:42 AM PDT to parent up   0 recs

Duh, I knew that. Oops.

It should be weighting OBP more though.

Matt Cain: He'll save children, but not the Dodger children.

by jponry on Jul 16, 2008 9:11 AM PDT to parent up   0 recs

do you know the formula?

proud father of the newly acquired Brandon Crawford..

by Azmanz on Jul 16, 2008 11:18 AM PDT to parent up   0 recs

From bb-ref’s glossary:

Adjusted OPS+

This value is calculated differently from the Total Baseball PRO+ statistic. I chose OPS+ to make this difference more clear. PRO+ as best I can tell is

PRO+ = 100 x ( OBP/lgOBP + SLG/lgSLG – 1)/BPF

Where lgOBP and lgSLG are the slugging and on-base percentage of a league-average player, and BPF is the batting park factor. This takes into account the difference in runs scored in a team’s home and road games, so it doesn’t depend on how good an offense or defense a team has.

My method is slightly more complicated, but I think it is more correct. The BPF is set up for runs and the way it is implemented in PRO+ applies it to something other than runs.

You can see his whole method in the glossary, it’s not copying correctly here.

Matt Cain: He'll save children, but not the Dodger children.

by jponry on Jul 16, 2008 11:30 AM PDT to parent up   0 recs

just in case

it’s not obvious from the formula. It weights OBP higher because the lgOBP* is typically (basically always) lower than the lgSLG. So if thel lgOBP is .330 and the lgSLG* is .415, compare 2 players:

A OBP: 400 SLG 415 (OPS* .815)
B OBP: 330 SLG 485 (OPS* .815)

A OBP/lgOBP* = .4/.330 = 1.21 (SLG/lgSLG = 1)
B SLG/lgSLG* = .485/.415 = 1.16 (OBP/lgOBP = 1)

So A would have an OPS+ of 121 and B would have one of 116.

FIRE BRIAN SABEAN

by zenbitz on Jul 16, 2008 12:25 PM PDT to parent up   0 recs

thanks

makes perfect sense

proud father of the newly acquired Brandon Crawford..

by Azmanz on Jul 16, 2008 12:36 PM PDT to parent up   0 recs

Wow I actually understand this. If the world is ending I apologize.

"While conservatives tell you 'leave things alone and no one will lose,' and liberals tell you 'interfere a lot and no one will lose,' baseball says 'someone will lose.' Not only says it - but insists upon it! ... Democracy is lovely, but baseball's more mature." BVCE supports SF Dugout and Manny Burriss.

by BaronVonCurrentEvents on Jul 16, 2008 12:45 PM PDT to parent up   0 recs

Eh, it’s alright, the world is pretty overrated. It’s definitely no purgatory; now there’s a place with some upside.

Neglectful father of David Quinowski

by marcello on Jul 16, 2008 1:42 PM PDT to parent up   0 recs

Crap

sorry guys I am having a heck of a time typing tonight.

I'm young but i didn't fall off the truck yesterday!

by jbowl on Jul 15, 2008 7:37 PM PDT   0 recs

Please explain FIP and BABIP

And please give some sense of the norms (i.e., bad, average, pretty good, and great). Many thanks!

I'm adopting a true Giant and an awe-inspring gamer: tk. "Atta babe."

by Mayor of 311 on Jul 15, 2008 9:01 PM PDT   0 recs

FIP – Fielding Independent Pitching: About the same as DICE above, the formula is FIP=(13HR + 3BB – 2K/IP)+3.20. The +3.2 is in there so the final number approximates ERA.

BABIP – Batting Average on Balls In Play: Just what it says. A player’s batting average when home runs and strikeouts are taken out of the equation. This stat is mostly used to see how “lucky” a player is. League average is pretty stable around .300. An individual’s BABIP is highly dependent on the type of contact they make, and a players xBABIP (expected BABIP) can be approximated by: LD% + .12, or more accurately by (GB%.24+FB%.15+LD%*.73). If a player is significantly over/underperforming their xBABIP, they have likely been lucky/unlucky on their balls in play, respectively.

Another note: Often in statistical discussions, when players are drastically over/underperforming their career averages, someone will undoubtedly say that they will “regress to the mean.” Really, more accurately, the player will likely regress TOWARD their career mean, but statistics have no memory, and to get TO the mean, the player would have to under/overperform their career averages by the same amount for the same amount of time.

For example, if a player is hitting 50 points above their career average halfway through the season, to reach their career average, they would have to hit 50 points BELOW their career average for the rest of the season. It is more likely they will preform at their average for the rest of the season, resulting in a season total 25 points above their career average.

by Viliphied on Jul 15, 2008 9:59 PM PDT to parent up   0 recs

Just a small note on FIP:

Personally, I like xFIP better, since it tries to normalize the HR rate to what should be expected based on your GB rate and home park. Both are fine though. When looking at FIP or xFIP, just think of the number as ERA.

On what you wrote above about xBABIP, it looks like the auto-formatting screwed it up. So, just to clarify for others, the equation you wrote is (GB% x .24 + FB% x .15 + LD% x .73). But really, as you said, approximating it as LD% + .12 is fine.

Lastly, because it’s such a good point, I want to quote this:

Often in statistical discussions, when players are drastically over/underperforming their career averages, someone will undoubtedly say that they will "regress to the mean." Really, more accurately, the player will likely regress TOWARD their career mean, but statistics have no memory, and to get TO the mean, the player would have to under/overperform their career averages by the same amount for the same amount of time.

For example, if a player is hitting 50 points above their career average halfway through the season, to reach their career average, they would have to hit 50 points BELOW their career average for the rest of the season. It is more likely they will preform at their average for the rest of the season, resulting in a season total 25 points above their career average.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 10:11 PM PDT to parent up   0 recs

And also, because it’s so nice to see, I’d like to quote this from that:

TOWARD

Ah… it’s rejuvenates me.

My Dave Righetti is better than your Dave Righetti.

by howtheyscored on Jul 15, 2008 11:03 PM PDT to parent up   0 recs

And people who have been seeing (not necessarily reading) my comments for a while now will realize that, just this once, I’m not even joking.

My Dave Righetti is better than your Dave Righetti.

by howtheyscored on Jul 15, 2008 11:04 PM PDT to parent up   0 recs

Hold it-- there are times you HAVE been joking???

I'm adopting a true Giant and an awe-inspring gamer: tk. "Atta babe."

by Mayor of 311 on Jul 15, 2008 11:18 PM PDT to parent up   0 recs

xFIP hates Giant pitchers.

Organizationally, the Giants encourage their pitchers to work up in the zone, leading to more flyballs (and, in my opinion, more K’s and walks). xFIP really hates flyball pitchers, because flyballs tend go for extra bases/home runs.

you can't block the Bocock

by oldjacket on Jul 15, 2008 11:04 PM PDT to parent up   0 recs

FIP assumes that pitchers have full

control over their HRs allowed per flyball. xFIP assumes that they have no control over that at all, it regresses that 100% towards the league mean.

Realistically, it’s probably somewhere in between the 2.

ZIPS: Milledge: 466 HR, 485 2B, 2282 hits, 278-379-524

by rfloh on Jul 16, 2008 12:30 AM PDT to parent up   0 recs

What exactly is RZR? Why does it seem to rate outfielders higher than infielders?

you can't block the Bocock

by oldjacket on Jul 15, 2008 10:11 PM PDT   0 recs

RZR is the proportion of balls hit into a fielder’s zone that he successfully converted into an out. From this article, it is further described as:

The central idea is to evaluate the fielding of individual players by analyzing only those zones in which the average fielder at that player’s position fields at least 50% of balls for outs. This method allows you to split the playing field between fielders and assign responsibility for many batted balls.

An important part of RZR is the Out of Zone (OOZ) plays, which is what it sounds like: the number of plays made by a fielder outside of his defined zone.

RZR, in conjunction with OOZ, gives a decent idea of how much range a player has. To quote the linked article again:

Revised Zone Rating is only a measure of fielding range. There are other things to consider when judging fielders, such as the ability to turn the double play, handle bunts, throw out runners, back up plays, etc.

In regards to your second question, I’m not sure, sorry.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 10:45 PM PDT to parent up   0 recs

The reason OFs

get higher ratings in RZR, and also zone rating, is that balls hit into the OF are usually easier to catch.

But, a missed play in the OF is also usually going to cost more runs.

When using RZR or zone rating, you shouldn’t compare players in different positions. If you’re looking at Aaron Rowand’s RZR, compare him to other CFs.

ZIPS: Milledge: 466 HR, 485 2B, 2282 hits, 278-379-524

by rfloh on Jul 16, 2008 12:33 AM PDT to parent up   0 recs

can we ask for editorials?

It may be a lot to ask, but what’s everyone’s favorite defensive statistic and why?

you can't block the Bocock

by oldjacket on Jul 15, 2008 10:59 PM PDT   0 recs

RZR is nice because it’s so readily available at all times.

What I think the best way to deal with defensive stats overall, though, is to look at as wide a range of them as you can and identify trends. Of course, the problem there is that some of the best ones (Dewan’s +/-, Dial’s Defensive Runs Saved, MGL’s UZR, etc.) aren’t available on a day to day basis.

Dial’s got AL values for starters up and should have NL values up soon though, which is nice.

Matt Cain: He'll save children, but not the Dodger children.

by jponry on Jul 15, 2008 11:30 PM PDT to parent up   0 recs

This.

Neglectful father of David Quinowski

by marcello on Jul 15, 2008 11:34 PM PDT to parent up   0 recs

I’ll just add a small amendment to this (which I +1 all the way). Make sure that when look at the different defensive stats that you also that they might be measuring different things, some measure runs and others measure outs.

Bay City Ball

by xanthan on Jul 16, 2008 8:50 AM PDT to parent up   0 recs

Wow, that was a horrible bit of writing.

I’ll just add a small amendment to this (which I +1 all the way). Make sure that when you are looking at the different defensive stats, that you also understand that they might be measuring different things, some measure runs and others measure outs.

Bay City Ball

by xanthan on Jul 16, 2008 8:51 AM PDT to parent up   0 recs

To get a good rough estimate of a player’s range, I suggest looking at the player’s RZR and OOZ. But if you’re looking for the more accurate information, I suggest looking at Chris Dial’s work (which is just working with ZR to find runs saved, etc.) and John Dewan’s Fielding Bible. I also like Chone Smith’s tweaked zone rating, which is pretty easy to calculate on your own. Mitchel Lichtman’s work is liked by a lot of people . . .IIRC, he was hired by the Cardinals and doesn’t post his ratings anymore. But his work is essentially the same as Dewan’s.

Actually, I have some questions of my own. How does Dial figure out “average opportunities” and “AvgZROps”? I think he mentions that comes from years of data, but if someone knows how he does it, that’d be great.

Also, how does Baseball Prospectus figure out their RATE and RATE 2 metrics?

Proud adoptive parent of Tim Alderson.

by Anticon23 on Jul 15, 2008 11:36 PM PDT to parent up   0 recs

AFAIK, MGL no longer works for the Cardinals and sometimes posts his data on Tom Tango’s Book Blog.

Matt Cain: He'll save children, but not the Dodger children.

by jponry on Jul 16, 2008 12:23 AM PDT to parent up   0 recs

What’s RATE?

..so allow me to present Tim Lincecum and Matt Cain as two sweet, sweet bottles of warming hooch.

by Cookyman on Jul 16, 2008 7:45 AM PDT to parent up   0 recs

RATE is a Baseball Prospectus fielding measure.

A way to look at the fielder’s rate of production, equal to 100 plus the number of runs above or below average this fielder is per 100 games. A player with a rate of 110 is 10 runs above average per 100 games, a player with an 87 is 13 runs below average per 100 games, etc.

RATE2 incorporates adjustments for league difficulty and normalizes defensive statistics over time.

I have no idea why they decided to base it around 100, instead of 0. Anyways, BP’s fielding stats aren’t that great. They’re probably useful in tandem with other stats, but I certainly wouldn’t use them as a stand alone.

Neglectful father of David Quinowski

by marcello on Jul 16, 2008 8:53 AM PDT to parent up   0 recs

Huh. You learn something new every day. Anyway I usually go to BP only for hitting stats (except for maybe pitchers VORP).

..so allow me to present Tim Lincecum and Matt Cain as two sweet, sweet bottles of warming hooch.

by Cookyman on Jul 16, 2008 9:50 AM PDT to parent up   0 recs

Also, how does Baseball Prospectus figure out their RATE and RATE 2 metrics?

I’m not sure, it might be proprietary information.

Neglectful father of David Quinowski

by marcello on Jul 16, 2008 8:46 AM PDT to parent up   0 recs

Exactly. ;)

Neglectful father of David Quinowski

by marcello on Jul 16, 2008 8:53 AM PDT to parent up   0 recs

Also all of the other defensive stats closely track each other, but BP’s is off in the Gamma quadrant.

you can't block the Bocock

by oldjacket on Jul 16, 2008 9:00 AM PDT to parent up   0 recs

with the Founders?

by tyrannoman on Jul 16, 2008 9:09 AM PDT to parent up   0 recs

what's a what?

you can't block the Bocock

by oldjacket on Jul 16, 2008 9:15 AM PDT to parent up   0 recs

nerd

Eugeniooooooo!!!!

by FairweatherFan on Jul 16, 2008 9:28 AM PDT to parent up   0 recs

Founders? Revive the John Adams thread?

I'm adopting a true Giant and an awe-inspring gamer: tk. "Atta babe."

by Mayor of 311 on Jul 16, 2008 10:08 AM PDT to parent up   0 recs

Adams was a Solid

Fred Lewis can stand under my umbrella.

by S.F. Giangst on Jul 16, 2008 10:46 PM PDT to parent up   0 recs

Thanks. =)

I’m just glad to know that I’m not the only one that isn’t sure how BP figures out this stuff. I wish they’d come out with a manual or something.

Proud adoptive parent of Tim Alderson.

by Anticon23 on Jul 16, 2008 11:05 AM PDT to parent up   0 recs

Awesome thread.

I wish it could be stickied or something.

Billy Hayes: His job is better than yours.

by delorean on Jul 16, 2008 8:46 AM PDT   0 recs

2 Questions

What is EQA, and is there a non-subscription place where people can get park factors for stats like ERA+?

¿Julio is tourist in San Francisco? Harper's Bizarre!

by hairball on Jul 16, 2008 9:06 AM PDT   0 recs

EQA

From BP:

Equivalent Average. A measure of total offensive value per out, with corrections for league offensive level, home park, and team pitching. EQA considers batting as well as baserunning, but not the value of a position player’s defense. The EqA adjusted for all-time also has a correction for league difficulty. The scale is deliberately set to approximate that of batting average. League average EqA is always equal to .260. EqA is derived from Raw EqA, which is (H + TB + 1.5*(BB + HBP + SB) + SH + SF) divided by (AB + BB + HBP + SH + SF + CS + SB). REqA is then normalized to account for league difficulty and scale to create EqA.

Basically, it is another attempt at an all encompassing offensive metric, like OPS+.

Again, like defensive metrics, it is best to consider several ( OPS+, EQA, AND Slash) when evaluating a player. Very seldom in life will you find one number that is the complete answer. They each have their strengths and weaknesses.

No 42 jokes, please.

Eugeniooooooo!!!!

by FairweatherFan on Jul 16, 2008 9:31 AM PDT to parent up