Trying to Measure Contact Management

This past weekend I had the pleasure of attending SaberSeminar in Boston. This is the second consecutive year I have been able to go, and I would highly recommend that you attend in future years if at all possible.

There were many great presentations, but one in particular stood out to me because of the potential relevance to fantasy baseball. Our very own Tony Blengino gave a spectacular presentation on the best and worst pitchers in the history of baseball at contact management (aka inducing weak contact). As far as I can tell, Tony took the HITf/x data, to which us normal people don’t have access, and calculated how each pitcher performed when allowing the various batted ball types. He then combined the performance on various batted ball types and scaled to 100 like we do here with things like ERA- and wRC+. I’m positive I’m simultaneously butchering the methodology while leaving significant portions of it out. Forgive me. For a little more insight, read Blengino’s recent posts on limiting hard contact for AL pitchers and NL pitchers.

This got me all fired up to get back home and see if I could calculate something like what Tony came up with so that we could use this as a fantasy tool. I was thinking this could be a new mechanism by which we could determine a player’s ability to induce weak contact. That’s a drum that Michael Salfino has long been beating by looking at ISO allowed. Salfino has rightly pointed out that hit quality (measured by ISO allowed) is more meaningful than the hit itself (measured by BABIP).

As noted in the link above, Salfino got turned on to the idea of ISO allowed when trying to explain why Johnny Cueto is so good at preventing runs despite variations in his BABIP. So when I saw Blengino’s presentation, I kept thinking that if we’d had this information available in 2011-12, we could have used his contact management numbers to show that Cueto was good at inducing weak contact. Matt Cain also crossed my mind. Could Blengino’s contact management work have convinced us that Cain induced weak contact and could thus continue to outperform his peripherals long before we were willing to accept it?

Cueto and Cain are a really good pairing to test something like this out. They both started out by outperforming their peripherals in an unspectacular fashion. They each had seasons with an ERA in the mid-to-high threes while their ERA estimators were over 4.00. But then they both took it to the next level by posting sub-3.00 ERAs while their ERA estimators sat in the mid-to-high threes.

This got me looking at their splits. A friend told me I could find splits based on batted ball type at b-ref. What I found there was a stat b-ref calls sOPS+. This is just a stat that tells us what a pitcher’s OPS allowed was relative to league average when surrendering each of the batted ball types. The nice thing here is that walks aren’t factoring into the on-base side of this OPS calculation because the split is only looking at balls put into play. So essentially this is just slugging plus the on-base bump for singles. This jives with the idea of focusing on hit quality, but this doesn’t completely eliminate singles from the equation.

I decided to look at how Cueto and Cain performed in 2011 since this was a year in which they both had a sub-3.00 ERA despite having a SIERA over 3.75. Unsurprisingly, each was better than league average according to sOPS+ on all three batted ball types. Cueto had the best looking set of numbers. He was well above average on ground balls and fly balls and 10% better than average on line drives. But as Blengino pointed out in his presentation, it’s harder to separate too far from the pack on line drives given that line drives so often result in favorable outcomes for hitters. Cain was just as good as Cueto was on line drives and was better than almost everyone in the league on fly balls. But he was only somewhat above average on ground balls.

If you calculate their sOPS+ on all batted ball types based on their performance on each batted ball type and the frequency with which they generated each batted ball type, you see that Cueto was downright dominant in 2011. His sOPS+ for all batted ball types was 59. Just to be clear, that’s 41% better than league average. To put that into perspective, just four pitchers this season have an ERA- that is more than 41% better than league average. Cain clocked in with a total sOPS+ of 71. I’m speculating, but I think that would have been a borderline top ten number.

If I’m trying to make a long story short here, I’d say that I think sOPS+ can be a good indicator of a pitcher’s ability to induce weak contact. And thus it may be something we can turn to when a pitcher’s ERA is vastly outperforming what the ERA estimators say. I’d dig into this further, but I wrote most of this after landing back in Dallas around midnight. And we’re over 800 words anyway. Look for more on this topic on Wednesday.

Print This Post

You can find more of Brett's work on or follow him on Twitter @TheRealTAL.

10 Responses to “Trying to Measure Contact Management”

You can follow any responses to this entry through the RSS 2.0 feed.
  1. TheBillsFly says:

    Why isn’t the hit-type/SLG%/ISO allowed available on the pitcher pages on this site? It seems like something that would be quite easy to add.

    Vote -1 Vote +1

  2. DrBGiantsfan says:

    BA recently posted an article from the Area Code Games showing data from a military grade radar system dubbed Trackman that not only measures velocity of pitches but spin rates in RPM. They then posted data correlating higher RPM’s with swinging strikes and in the case of fastballs, a higher flyball rate. Just a guess, but I bet higher spin rates also correlate with weaker contact.

    What I found surprising is the large differences in measured spin rates and spin rates do not necessarily correlate with velocity, even in fastballs.

    Vote -1 Vote +1

  3. 92.456789% of all stats are false! says:

    Interesting. Could be a cool tool for fantasy analysis. Are you arguing that contact management is a skill independent of K, BB and pitch type abilities? I’d love more evidence before I’m with you on that point. Also, are you arguing that contact management is descriptive or predictive? If the latter, I’d love to see data showing that past contact management is predictive of future contact management. I’m guessing there’s a lot of random error in it, which probably makes it a bad predictor.

    Vote -1 Vote +1

    • Brett Talley says:

      I don’t know about the independent part. I think it could be, but I didn’t look into that at all, and it wasn’t the question I was asking myself. But it is an idea. All I’m really wondering is if this might be an explanation for why pitchers outperform their peripherals. I’m going to dig a little deeper into that for Wednesday.

      Vote -1 Vote +1

      • 92.456789% of all stats are false! says:

        Thanks. Hmmm sounds like there are two sources of random error. One is that the apparent FIP “exceptions” like Cain could be random error. After all, Era predictors aren’t very accurate to begin with. A second source of random error could be that contact “management” is really just largely random noise. If it’s a real skill, how does a pitcher get good at it?

        Good luck!

        Vote -1 Vote +1

  4. emrldwpn says:

    This is really interesting. Eagerly awaiting the follow-up article. Thank you!

    Vote -1 Vote +1

  5. Rob says:

    I would like to see the results of this analysis for Mark Buehrle. He has almost always outperformed his FIP and xFIP.

    Vote -1 Vote +1

  6. Milad says:

    Great article. Would also be interesting to see if the players the ERA predictors are wrong on (Ricky Nolasco, etc.) are the opposite in this respect. That would jive with the fact that high HR/FB rates tend to happen to those types of pitchers a lot, possibly indicating harder contact as compared to league average.

    Vote -1 Vote +1

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>