Fangraphs Q&A - Recent questions and answers in Sabermetrics

Answered: How can I find a list of linear weights for wOBA beyond 2010?

Tue, 18 Jun 2013 15:03:12 +0000

http://www.fangraphs.com/library/offense/woba/

For 2012 weights.

Predictive K%

Sun, 16 Jun 2013 14:54:39 +0000

Is there any way to convert Mike Podhorzer's xK% formula into a formula that only uses stats available on fangraphs? Obviously, it would get a bit more complicated, but I imagine the stats Looking strike, swinging strike and foul strike can be found out by doing a little math to the plate discipline stats available on fangraphs. I attempted to do this myself, but was unsuccesful. I figure someone better with stats may have better luck. Thanks in advance.

xK% = -0.61 + (L/Str * 1.1538) + (S/Str * 1.4696) + (F/Str * 0.9417)

Answered: What is Starlin Castro's batting average by pitch type?

Sun, 16 Jun 2013 05:26:09 +0000

I'd say your observation agrees with PITCHf/x data frm 2013. Nice job.

Looking at Castro's batting average on contact (which would include home runs), Starlin is hitting .307 on sliders, change-ups, curves, and knuckle curves.

His combined BA/CON for four-seamers, two-seamers, cutters, and sinkers is just .281.

Here is the breakdown for each pitch type:

https://docs.google.com/spreadsheet/ccc?key=0AiJWR6a0B5vndDFhdzA0Zk1RSUNLMVZZN1YtMllWSmc&usp=sharing

Answered: Is there anywhere to find a stikezone map with BABIP in each quadrant?

Fri, 14 Jun 2013 16:53:58 +0000

Brooksbaseball.net has this for pitchers. I've never seen it for hitters, but that doesn't mean it isn't out there. Here's a link to Verlander's

http://www.brooksbaseball.net/profile.php?player=434378&gFilt=&time=month&minmax=ci&var=babip&s_type=2&startDate=03/30/2007&endDate=06/14/2013&balls=-1&strikes=-1&b_hand=-1

Answered: Rookie query: Is there a consensus on best predictive stat for hitters and pitchers?

Thu, 13 Jun 2013 05:16:46 +0000

wOBA correlates year-to-year at r = .61

SIERA correlates year-to-year at .71

http://www.beyondtheboxscore.com/2012/1/9/2690405/what-starting-pitcher-metrics-correlate-year-to-year

http://www.beyondtheboxscore.com/2011/9/1/2393318/what-hitting-metrics-are-consistent-year-to-year

Answered: Is there a way to measure batting inconsistency?

Tue, 11 Jun 2013 14:51:10 +0000

I would argue the situation you describe has the hitter being highly consistent. Alternating 2/3 nights with 0/3 nights is the epidemy of constancy

Answered: But when you divide the wHR value by the wOBA scale you get something between 1.6-1.7?

Wed, 05 Jun 2013 01:26:01 +0000

True. But while the run value of the home run is 1.4, the run value of an out is -.25 or so. And because an out in the wOBA formula needs to equal zero (-.25+.25=0), the other events must add .25 or so as well (1.4+.25=1.65). Then the scale is added.

Here is a list of run values for just about all possible events:

http://www.insidethebook.com/ee/index.php/site/article/run_values_of_events/

Answered: Which teams lead the AL and NL in Solo Home Runs?

Mon, 03 Jun 2013 21:11:47 +0000

http://www.fangraphs.com/leaders.aspx?pos=all&stats=bat&lg=all&qual=0&type=0&season=2013&month=27&season1=2013&ind=0&team=0,ts&rost=0&age=0&filter=&players=0&sort=9,d

Orioles (49) and Braves (46)

Answered: What was the last game M. Cabrera popped out inside the dirt of the infield?

Mon, 03 Jun 2013 00:43:41 +0000

May 26th, 2013

Answered: why hard to find bb/k ratio? if flawed stat, - why?

Sun, 02 Jun 2013 15:21:50 +0000

http://www.beyondtheboxscore.com/2012/11/25/3686732/stop-using-k-bb

Answered: Has anyone done a study of RBI's by hit type?

Fri, 31 May 2013 03:29:22 +0000

So you want to see what percentage of RBI are the result of 1B, 2B, 3B, HR etc? And see how this pans out at the team level? Interesting...

Ran the percentages for 2012. The Yankees scored half their RBI via the long ball:

https://docs.google.com/spreadsheet/ccc?key=0AiJWR6a0B5vndE5nS2RYejBkRDVJSWI2N2hFRFQ0bUE&usp=sharing

Answered: Where does 1.4 run value for HR come from? Seems over 2 by LWTS on constants page.

Fri, 31 May 2013 02:59:41 +0000

The linear weights value of a home run is around 1.4 as you've probably heard. The values in the 'seasonal constants' table are actually 'wOBA weights.' These values in the wOBA formula are linear weights scaled in such a way so that an average wOBA will resemble an average OBP.

Answered: What pitches does Robert Coello throw?

Wed, 29 May 2013 21:29:44 +0000

He throws it like a fork ball, but it behaves like a knuckler:

http://sports.yahoo.com/news/angels-pitcher-robert-coello-s-mystery-pitch-leaves-hitters-muttering-under-their-breath-185910466.html

Answered: Is there a way to see splits by count?

Wed, 29 May 2013 04:02:30 +0000

Count splits are available in the player page 'splits' tab:

http://www.fangraphs.com/statsplits.aspx?playerid=4881&position=OF&season=2013

Answered: Aoki WAR discrepancy

Mon, 27 May 2013 20:05:36 +0000

DRS values Aoki at 11 runs above average (rWAR), UZR at around 2 runs above average (fWAR)

Answered: Which leads to 3+ run innings more leadoff HR or double?

Sun, 26 May 2013 04:22:48 +0000

A quick query from 2002-2012 suggests that both lead to 3-run innings 13% of the time.

Innings with leadoff home runs scored an average of 1.5 runs. Innings with leadoff doubles scored an average of 1.1 runs.

I can double check the data later.

Answered: Question on WRC+

Sun, 26 May 2013 03:01:58 +0000

wRC+ is available under the player splits which includes LHP/RHP and Home/Away splits:

http://www.fangraphs.com/statsplits.aspx?playerid=9368&position=3B&season=2013#advanced

mlb team winning percentage when scoring first

Thu, 23 May 2013 14:24:48 +0000

Answered: streamer R v U, what does this mean?

Wed, 22 May 2013 14:52:59 +0000

Steamer is a projection system. You may have googled the term "streaming".

"R" refers to that player's projection for the Rest of Season. "U" refers to that player's Updated projection.

Answered: Where can I get actively updated baseball databases

Tue, 21 May 2013 02:56:31 +0000

Can you just download Fangraphs 2013 data in the 'leaderboards' to spreadsheet then import it to your database as csv? That's what I do.

Answered: Can someone tell Joe Mauer to stop striking out? Please?

Sat, 18 May 2013 02:30:14 +0000

I'll tell him this weekend when I see him at Uncle Al's retirement party.

Answered: what is Dan uggla's batting average when the pitcher throws a change up?

Sat, 18 May 2013 02:28:11 +0000

It's a bit difficult to calculate batting average (H/AB) for a single pitch type because you have so many pitches that are neither 'hits' nor 'at-bats.' If Uggla fouls off a Change-up do you count that as an 'out' in his change-up batting average? Or ignore it entirely? What about balls? Called strikes in 3-0 counts?

Therefore, Fangraphs instead uses run values to look at what pitch types batters might have trouble with. Those numbers are located here:

http://www.fangraphs.com/statss.aspx?playerid=3442&position=2B#pitchvalues

Uggla, for the record, has a negative wCH (run value on Change-ups) currently at -3.4 this season. It is second only to the Slider (-4.3) as his most troublesome pitch.

Answered: Component ERA

Sat, 18 May 2013 02:02:55 +0000

Pitcher's total base estimate, and the formula is right there.

PTB = 0.89×(1.255×(H − HR) + 4×HR) + 0.56×(BB + HBP − IBB)

Answered: Why doesn't OBP include every time a player reaches base?

Fri, 17 May 2013 02:48:03 +0000

Sure, he's on base, but that's not really what you care about. You care about how many outs he's making. OBP is really meant to be a "non-out percent" but that doesn't really roll off the tongue so well.

ROE, yes, I see your point. That should probably be included in OBP. In the original wOBA formula ROE it was included.

Answered: Will you combine batting, pitching, fielding WAR levels to make one team WAR stat?

Wed, 15 May 2013 00:11:48 +0000

The batting WAR leaders incorperates fielding so there is really just Batting+Pitching.

This doesn't change the crux of your question just that you only need to combine 2 different WAR values not 3.

Answered: Where I can find stealing percentages against by pitcher?

Tue, 14 May 2013 20:40:18 +0000

Here is a spredsheet with SB Against for all pitching-seasons since 2009 with at least 150IP if you still feel like browsing.

https://docs.google.com/spreadsheet/ccc?key=0AiJWR6a0B5vndE1VWlVYVGw1OGVqSFNXOWU0WjFZOUE#gid=0

And the article fom whence it came:

http://www.beyondtheboxscore.com/2013/5/13/4320312/stolen-base-rates-for-pitchers-and-johnny-cueto-pickoffs-sabermetrics

Closers in non-save situations

Fri, 10 May 2013 12:56:38 +0000

Has anything been written (on FG or elsewhere) about closers in non-save situations, and how their performance in those situations compares to their overall performance? I feel like over the last 3 or so years, every time (almost literally) I see a manager bring in a closer in a tied game or a 4+ run lead situation, runs are given up. Jason Grilli vs. the Mets last night was the latest case, and he's been unhittable in save situations.

Any way to look at this?

Matt Cain and his Sliders??

Thu, 09 May 2013 21:47:13 +0000

What is up with Matt Cain throwing sliders this year. Im kind of new to sabremetrics but ive realized that he is throwing 10% more sliders than he has over the last 6 years or so. Could this be credited to his recent lack of success? Or will this even out as the year progresses? He through above 20% last year also.

Answered: Why is fangraphs WAR different from Baseball Reference WAR?

Wed, 08 May 2013 21:02:19 +0000

Niether is 'correct'. The two sites have agreed upon the definition of replacement level, but the two sites still use different methods to value players-- particularly with pitchers.

Basically, Fangraphs rates pitchers according to their FIP, while B-R uses a defense adjusted RA9. Moore's ERA/RA9 is incredibly low right now, compared to his defense-independent numbers like, K%, BB% and HR%, hence the large discrepency.

Answered: can you compare pitcher and batter war at the same time?

Mon, 06 May 2013 11:17:28 +0000

That's the idea. WAR attempts to put a price tag on a players value, adjusting for position, league, and season so that we should be able to compare different players at different positions on different years.

That said, I wouldn't want to automatically assume a pitcher with a 3.5 WAR is absolutely more valuable than a 3.2 hitter with so many variables involved, but it should get you pretty close.

Answered: Would anyone be willing to help my calculate my high school baseball batting stats from this season?

Mon, 06 May 2013 03:37:59 +0000

Which stats were you hoping to calculate?

Answered: What does the 'dol' mean in the value section of the leaderboard

Sun, 05 May 2013 23:28:13 +0000

'Dollars'.

It basically represents the amount of $ the player's WAR numbers would price him at in the Free Agent market.

For instance, last year's 10 WAR performance from Mike Trout would cost a team 44.8 million dollars to buy it.

According to 'Dollars', 1 Win is worth something like 4.5 million dollars. By that logic, an average ballplayer (2 wins per season or so) would get something like (roughly) 9 mil per year, which works out intuitively.

Answered: Where can I find the wOBA against for a pitcher?

Sat, 04 May 2013 14:37:19 +0000

You may have to do that manually in excel. I could give you Retrosheet wOBA splits but not for 2013.

Answered: Brett Anderson QS in relief?

Thu, 02 May 2013 18:15:49 +0000

It'd be a relief appearance, not a QS.

Answered: Anything that predicts the probability that a player gets at least 1, at least 2, etc hits for the next game?

Mon, 29 Apr 2013 18:37:31 +0000

One game is way to small of a sample size to be able to generate probabilities for performance.

Answered: Crazy Fact or BS?

Mon, 29 Apr 2013 05:27:33 +0000

I'm gonna say B.S. I don't have the time to look at that exact query, but I did quickly check more recent years with pitchers reaching first and advancing and saw nothing remotely close to that 81%.

Who reported that figure/ made the tweet? A reputable source or not?

If it's a reputable source I'll check my numbers.

Answered: Is there an available stat for batters to determine the effectiveness of the pitchers they've faced?

Thu, 25 Apr 2013 14:26:49 +0000

I don't think there is that does it, but a couple thing scome close. wRC+ is good becasue it takes run enviorment inot accoutn, and beacues the quality of pitcher effects the run enviornment this stat paretially takes into acound what you are talking about. Also on baseball reference you can look at batter slpits and see how they do agains oponets with a record over .500 vs ones under .500, this is obviouls a rough proxy but you can asume teams that winn more have better pitchers. On average batter an OPS 46 points lower vs. teams over .500.

Answered: Is there a statistic for bunt success?

Thu, 25 Apr 2013 14:20:23 +0000

Baseball referance has got you covered on this one, under hitting stats go to situational hitting, The have Sac hit (aka bunt) attemps and successe for earch player. http://www.baseball-reference.com/leagues/NL/2012-situational-batting.shtml this is the 2012 NL leader board

Answered: What's the average fastball speed of all (and starting) left-handed pitchers in MLB in 2012?

Thu, 25 Apr 2013 14:09:46 +0000

90.39 MPH for left handers and 92.08 MPH for righties acording to texas leaguer

http://pitchfx.texasleaguers.com/league-averages.php

Is a great pitch fx site, they don't have the league averages up for 2013, but all you have to do is go to the player page for a left handed pitcher and float your mouse over there fasball vellocity and it will give you the league average and stdev. ex. http://pitchfx.texasleaguers.com/pitcher/282332/ just hover your mouse over FF velocity.

Could you write an article about what is going on with Victor Martinez?

Tue, 23 Apr 2013 20:24:24 +0000

I can look at BABIP, but I'd like a deeper look into how he's making contact and other factors that might suggest or discount something other than just bad luck. Thanks!

Answered: Judging pitcher efficiency with batters faced?

Mon, 15 Apr 2013 05:47:15 +0000

It's not identical but it sounds very similar to WHIP and looking at the leaderboard it seems to be fairly similar. The main difference seems to be that your proposed metric would eliminate the negative impact the pitcher gets for allowing a hit / walking a batter if the batter is later eliminated on the basepaths.

Answered: It seems WARs have been changed recently. What is the difference?

Sat, 13 Apr 2013 14:43:40 +0000

Please see these two posts:

http://www.fangraphs.com/blogs/war-changes-and-updates/

http://www.fangraphs.com/blogs/unifying-replacement-level/

Answered: What does "RoS" mean? (as in "ZIPS (RoS)")

Thu, 11 Apr 2013 17:29:05 +0000

Rest of Season.

Answered: Are there plans to update park factors to include 2012 data?

Wed, 10 Apr 2013 14:42:33 +0000

http://www.fangraphs.com/qa/1332/theres-problem-with-park-factors-2010-and-when-will-have-2012

What is going on with Chris Carter's splits?

Tue, 09 Apr 2013 21:14:15 +0000

In reviewing Chris Carter's splits, I am sort of perplexed.

His career OPS is almost the same across L and R

However how that happens is totally different

http://www.fangraphs.com/statsplits.aspx?playerid=9911&position=1B&season=0

Somehow his BB% vs L is quite high (18.3%), where it is bad (7.3%) vs. R - so his OBP is skewed heavily towards batting left.

However, his SLG is the opposite - .359 vs L and .436 vs R

His average is similarly bad against both.

Last year his walk percentage was even more insane vs L (22%), but his slugging percentage was similarly skewed towards righties

I can't remember seeing players like this before.

Is this common?

I would guess this is just a small sample, but it seemed to be true in the minors as well.

Fangraphs doesn't have minor league splits information, but I found a site that does

http://mlsplits.drivelinebaseball.com/mlsplits/playerinfo/452080

Maybe he sees the ball exceptionally well vs left handers, so ends up taking a lot of walks?

Or he knows he will crush it against righties, so he swings at more pitches?

I don't know enough about this guy, can someone explain how this happens?

Are previous season's WAR Projections available?

Tue, 09 Apr 2013 14:47:57 +0000

I'm curious if you have previous year's WAR Projections availalbe. IE - just like you have Zips, Marcel, FANS, etc projections for 2013, do you have those values for previous years as well?

Answered: why is total war in the leaders tab for all teams exactly 1000 every year?

Sat, 06 Apr 2013 02:38:31 +0000

Both Fangraphs and Baseball-Reference changed their WAR calculation recently to sync replacement levels just a few weeks ago. It was pretty big news:

http://www.fangraphs.com/blogs/war-changes-and-updates/

Answered: Is there any way I can look at statistics for just interleague play?

Thu, 04 Apr 2013 20:45:17 +0000

I may be able to help you with that, if there is something in particular you are looking for?

Answered: Does Team Batting WAR include the fielding contribution?

Thu, 04 Apr 2013 20:39:56 +0000

Yes. Assuming you are talking about pages like this:

http://www.fangraphs.com/leaders.aspx?pos=all&stats=bat&lg=all&qual=0&type=8&season=2013&month=0&season1=2013&ind=0&team=0,ts&rost=0&age=0&filter=&players=0

I don't believe a defense neutral version of WAR is available at fangraphs. You would have to manually calculate that.

Answered: Noob question on how RC+ works

Thu, 04 Apr 2013 10:53:38 +0000

I believe the chart you are referring to is for "wRC" and not "wRC+". wRC would be the counting stat (where 60 runs a year is close to average-ish) that the rate stat wRC+ (where 100 is average) is based on.