Contact Quality (CQ)
Set forth below is a description of a short-hand stat designed to measure how well a batter hits when he hits a fair ball. I call it “Contact Quality” or “CQ.”
Obviously, BABIP is the most common measure of hitting balls-in-play. But BABIP excludes home runs, and doesn’t otherwise distinguish between singles and extra-base hits. One possible approach would be to compute slugging average on fair balls (“fair balls” refers balls-in-play plus home runs). However, the weights used in slugging average do not accurately reflect the value of different types of base hits.
CQ uses the weights from wOBA, in a simplified fashion, before the adjustment to normalize the scale to OBP. Rounding the weights, I assign 0.8 for a single, 1.1 for a double, 1.4 for a triple, and 1.7 for a home run. Or, in other words, 0.5 for a hit and 0.3 per total base.
So, the formula is: (0.5*H+0.3*TB)/(AB-K)
Interestingly, the average CQ in 2014 was .312, as compared to an average OBP of .314.
CQ is intended to be a simple stat that, together with K% and BB%, gives a pretty good profile of any batter.
If a batter’s CQ is above .400, he is considered a member of Mensa (contact quality division). In 2014, among qualified batters, the following were Mensa members:
1. Giancarlo Stanton (.453)
2. Mike Trout (.450)
3. Jose Abreu (.435)
4. Chris Carter (.407)
5. Justin Upton (.405)
6. Andrew McCutchen (.404)
7. Matt Kemp (.403)
Rounding out the top 10:
8. Miguel Cabrera (.388)
9. Anthony Rizzo (.387)
10. Marlon Byrd (.386)
Why would you make up a stat like CQ when there exists linear weights which uses weights determined by the actual value of a PA instead of something made up in your head? And why do you think it measures contact quality when many well contacted balls end up being outs and more than a few poorly contacted balls end up being hits?
Peter —
As I wrote, it was intended as a short-hand stat, nothing more. It utilizes readily available stats, and is all based on linear weights (albeit rounding the numbers for simplicity). It doesn’t use batted ball information, but neither do many other stats (e.g., wOBA).
I think it could serve two related purposes. First, I often look at BB% and K% when evaluating batters (and pitchers). I think CQ captures what’s left out. Second, BABIP seems to be the default option to measure performance on balls in play; I think CQ paints a more complete picture.
Best, Craig
Nice article. A couple of things:
1) I think CQ is skewed in favor of guys with all or nothing, uppercut swings. Have you considered adding a coefficient to strikeouts?
2) Did you test it against a lot of players/notice anyone who looked like a bounce-back or regression candidate for 2015?
Thanks. On your first question, you’re correct. CQ (like BABIP) does not take into account strikeouts (or walks). It’s not intended to be an all-in measure of performance – rather, it’s meant to give an idea of how productive a batter was when he hit a fair ball.
So, for example, Victor Martinez had a CQ of .364 last year — good, but not great. But his K% was a lot lower than those of the leaders.
On the second question, no.
Craig
http://tangotiger.com/index.php/site/comments/wobacon
Yep, that’s pretty much the same — better acronym too!
http://www.draysbay.com/2011/3/28/2075213/know-thine-enemy-baltimore-orioles
I like it! When I evaluate hitters, I look at BB%, K%, BABIP, and ISO. This effectively combines BABIP and ISO, leading to a much more natural delineation into “patience”, “contact”, and “contact quality,” with the latter being a major improvement over current measures of “power”, all of which ignore some aspect of on the field performance. Thanks for posting!
Contact quality is so unexplored it seems, perhaps because we know Hit f/x will make it easier and thus it’s easier to just push the study of contact quality into the future. Anyhow, here what I think should factor into a contact quality stat: LD% (because liners are hit hard more often than not), XBH/(AB-K) (I don’t like differentiating different types of extra base hits; all extra base hits are almost always hit hard), flyball distance (because the harder it’s hit the farther it goes), and possibly a lesser explored concept I thought up: Zone Contact Ratio (ZCR).
ZCR=Contact on Pitches in the Zone divided by all contact. Reasoning: It’s easier to make solid contact on pitches in the zone than on pitches out of the zone. Therefore, the higher percentage of your contact is on strikes, the higher percentage will be well-struck.
Just my two cents.