A guide to all the goofy stats I use to talk about the White Sox

A guide to all the goofy stats I use to talk about the White Sox

I am thrilled--THRILLED--to get to use this photo again

A lot of things have changed since I started writing about the White Sox on Chicago Now back in April.  I've gotten a lot better understanding of the game, other blogs actually acknowledge my existence nowadays, the notoriety of CN's White Sox coverage has plummeted (they used to have Jack McDowell on this site!), and a bunch of other changes that I don't want to really acknowledge because I'm embarrassed by my early work.

But one of the most noticeable changes has been a slow gradual embracing of sabermetrics, or as like to call them, "statistics that let you know more about the game than you would have ever figured out on your own" or "STLYKMATGTYWHEFOOYO" for short.

Of course, as some random kid on a site platformed by a major media conglomerate, I have a wide variety of readers.  Some of you are statheads who barely put up with the amount of opinions I make not based on recorded fact (in which case, you've probably figured out this piece isn't for you by now), and some of you, who are starting to get really peeved about how often I use a random acronym, stick it next to a percentage, and act like I just wrote a real sentence.

But the confusion is coming to an end right now.  In just a few minutes all the obscure terms and figures will be second-nature to you, you'll be rattling them off to your friends and co-workers...and oh how they will mourn what you have become.

The building-block: WAR

Pretty much the most important thing in the world; a lot more important than your family.  WAR is the means to assess the value of every player, and this is expressed in how many more wins a player is worth over a ho-hum call-up from AAA (think Alejandro De Aza, and think him hard).  It's actually a calculation of runs, and 10 runs = 1 win--but oh wait, I'm supposed to simplifying things.


There was a time where the WAR stat was very kind to Mark Kotsay....I'm sorry you missed it

Position players are graded for their offense and defense, and pitchers just for how many runs they yield.

A good breakdown is this

Position Player:
6+ Really Fantastic
5 Great
4 Quite Good
3 Effective Starter
2 It'll Do
1 Not good but belongs in MLB
0 Bad; could be a minor-leaguer
-1 Mark Teahen
-2 Satan on Earth

Starting Pitcher:
6+ Super great
5 Regular Great
4 Very Good
3 Established Rotation Member
2 Working But You Ain't Happy
1 You Keep Trying To Leave Him At The Hotel During Road Trips
-1 The Reason You're The Royals
-2 Stabbing Teammates In The Dugout

Relief Pitcher:
2.5 Nominated For Pope
2,0 Very Great
1.5 Very Good
1.0 Perfectly Suitable
0.5-1.0 Almost Every Reliever In The Damn League
0.0-0.5 The Linebrink Zone

There are multiple sources of WAR, but I tend to use FanGraphs's version because that's the blog I read more.  You can make a good argument for using Baseball Reference's WAR for pitchers, especially when evaluating a season in retrospect because BR bases their formula on actual runs allowed as opposed to FIP, which attempts to evaluate the objective quality of pitching.  Speaking of which.


FIP: This stat is an attempt to remove all incidents that a pitcher can't control completely from the process of evaluating him.  So what things are entirely the fault of the pitcher?  Walks, home runs, and strikeouts.  It's a crazy formula these totals are entered into, but it comes out like ERA.  Under 3.00 is fantastic, Under 4.00 is not good, Under 5.00 is really just ok, over 5.00 is time to retire.   Criticisms are that it overvalues strikeouts, and doesn't give a fair shake to guys like Jon Garland and Mark Buehrle, but it's a great prediction tool for future outcomes.

The elements of FIP are per-9 inning rates of homers, walks, and strikeouts, expressed as HR/9, BB/9, and K/9

HR/FB: This is the percentage of fly balls that become home runs.  It's assumed that the pitchers have little control over how many of their fly balls become dingers, but it is their fault how many total fly balls they give up.  So HR/FB really tells us how lucky the pitcher was.  If it was over 10.6%, they were unlucky and gave up more HRs than their skills dictate they should have, if it was under; they were lucky and it's possible they suck.


If you must stick to Wins and Losses, and ERA, please say that you're doing so out of loyalty to Jon Garland

GB%/FB%: Because fly balls are a risk to become home runs, a pitcher who forces grounders is desired over a flyball guy.  Good pitchers can keep their flyball rate under 40%, great ones can keep it under 30%, but a lot of guys make a living in the low-40s%.  If a player's fly-ball % is over 50, there's a good chance everyone who is a fan of the team he pitches for wants him dead.

BABIP: The measure of opposing batter's average on all balls in play or simply, what percentage of at-bats that don't go for strikeouts nor home runs become hits.  Like HR/FB, this is really a measure of how lucky the pitcher is.  Some guys (groundball pitchers) will tend to have a slightly higher BABIP than others (flyball pitchers), but this...

BABIP over .300 = unlucky
BABIP under .300 = Lucky!

...is a good rule of thumb to start out.


Compared to pitching, offense isn't that complex to evaluate.  There aren't nearly as many traditionally embraced stats that just need to be disregarded, and you can still trust that the guy who consistently hits .350 is good, whereas the pitcher who consistently wins 15 games a season could just play for the Yankees.  Even the pretty commonplace OPS stat (slugging pct + on-base pct) is useful in a pinch because it's really easy to calculate and gives you a pretty good idea of what's going on.  These are just going to make things a little easier.

wOBA: It takes the on-base percentage of a player and weights all of the outcomes to account for the power and slugging as well.  It's literally an all-in-one statistic for evaluating offense.  It's also weighted so that the average wOBA is the same as the average OBP for the league every year, so you can value it in the same way. 

.400+ is MVP level
.350+ is very good
~.300 is bad,
and under .300 is Mike Caruso


Now that you've embraced sabermetrics, you can now be even more overbearing about your love for Frank Thomas

wRC+: Based on wOBA, it expresses a hitter's performance (in this case, his 'run creation', hence the acronym) in terms of whether its below or above average.  100 is perfectly average, 125 is very nice, 150 is excellent, and 75 and under is dirt-crap-Royce Clayton-awful.

Other important rates are K% and BB%, which really just say what percentage of the hitter's plate appearances end in strikeouts and walks.  For walks, anyone over 10% is to be admired, over 15% is to be revered, and over 20% is to be bronzed.  Under 5% is very impatient.

Strikeouts more just tell you about the hitter's approach and style.  Someone like Adam Dunn strikes out over 30% of the time but makes up for it in power, while A.J. Pierzynski struck out less than 10% of the time in 2010, and was really pretty bad.  Significant changes in player's career rates are to be observed and taken note of, and if a player is really striking out waaay too much, then it will probably show up in other stats.


There are a lot of defensive statistics out there, and most of them are pretty good.  DRS is good, Total Zone is just fine, and baseballanalysts.com featured a really interesting report where they just calculated the percentage of chances in players zones that became outs (it was not kind to A.J. Pierzynski).  I just happen to cite UZR. 

The goal of any of these systems is to measure their defenders with a strong emphasis on the amount of field coverage they provide, as opposed to just measuring the number of obvious mistakes they make.  A player who gets to 900 balls, and makes 30 errors is more valuable than someone who gets to 500 balls and screw-ups 10.  These metrics certainly take errors into account, but as just one of many factors.

UZR measures in really fancy-pants ways the amount of runs compared to the average defender at the position the player is responsible for allowing.  I could say that someone who is -10 (Mark Teahen) is really bad or +10 (Alexei Ramirez) is really good and that players who stay within 3 runs either way of 0 are negligibly different from the average, but UZR is more complex than that.  It varies a lot from season to season, and the more years of data that are drawn from, the more accurate it will be.

For example, if you really think Jermaine Dye is a good defender, one awful season of UZR--though worrisome--shouldn't be enough to dismiss this notion.  Four crappy years in a row?  Then you need to re-evaluate things.  In most cases, UZR should be a supplement to your assessments of player defense, not the end-all be-all.  Using multiple metrics is also recommended.

In the world of baseball stats, what I just broke down was actually pretty basic, and I still don't understand a LOT of what's out there.  But this should keep you pretty up-to-speed on anything I say on this blog, and make you worlds smarter than the average drunk in the next row at U.S. Cellular.  I'll be posting it up on the navigation bar for constant reference at a later time. 

Follow White Sox Observer on Twitter @ JFegan_WSO_ODTL and on Facebook.  

It's never too late for links.

J.J. has cautiously positive news about Jake Peavy.
Jim at Sox Machine is apparently taking over South Side Sox, the most popular and powerful White Sox blog, meaning it's time to start being really nice to Jim

Leave a comment