Baseball Strategy

The futility of using pitcher-batter matchup history

You might remember last Saturday night, the fateful seventh inning against the Milwaukee Brewers. The Reds were holding on to a 1-0 lead after six brilliant innings by Mat Latos. Manager Bryan Price made the decision to lift Latos and rely on his bullpen to finish the game. Who would Price choose to pitch the high-leverage seventh inning, against Milwaukee’s 3-4-5 hitters?

Answer: Logan Ondrusek.

Bryan Price chose Logan Ondrusek. When asked after the game why he chose Ondrusek instead of Sam Lecure, Price said, “I kind of felt like it was a situation in that exact spot in the order, Ondrusek’s numbers against those guys are off the charts. There were just some really good matchups for him.” (Mark Sheldon)

Price has also used past matchup data to determine playing time. Against Bronson Arroyo, Price started Ramon Santiago because of “past success” Santiago had against Arroyo. Turns out that was three at bats. Price also started Skip Schumaker that game because of his track record against Arroyo. At least in that case, there was a history of 51 plate appearances. Of course, all but six of those were from 2011 or earlier, stretching back a decade.

Price isn’t alone in using data from specific past matchups. Big league managers do it all the time. It makes sense, right? There’s all this specific data, right in one of those thick binders that Mitt Romney isn’t using any more. That match-up data right in front of you, within easy reach. Use it! It sounds perfectly modern, even sabermetric. Also, the media, always looking for something to fill air time, spouts past matchup numbers as gospel truth.

The problem? That data is completely unpredictive. And that’s not abstract theorizing. It’s based on real life major league baseball.

The theory that certain hitters “own” certain pitchers or vice versa has been disproven by numerous studies. For example, Colin Wyers (Baseball Prospectus 2011)┬ástudied hitter-pitcher matchup data over sixty years of baseball history. He found that “ten, fifty or even a hundred plate appearances aren’t enough to tell us whether there’s a special edge or sample-size fluke.”

Tom Tango and Mitchel Lichtman (The Book: Playing Percentages in Baseball, Chapter 3) studied major league hitters and pitcher data from 1999-2002. They used 1999-2001 as the “before” period and 2002 as the “after” period. They stipulated the pitcher-hitter matchup had to take place at least 17 times before and at least nine times after. That let them identify 300 pairs of pitchers and hitters. Their findings:

“We found thirty hitters with fabulous hitting records against thirty pitchers. And yet, given the chance to prove this skill in subsequent confrontations, they fail miserably.” Looking at the pitchers who “owned” certain hitters, “once again, the identity of the opponent was irrelevant. These pitchers didn’t own these hitters.”

That’s not to say all batters and pitchers have the same chance of succeeding. Certainly the overall career numbers for a hitter or pitcher matter. Joey Votto has a better chance of getting a hit than Willy Taveras. Aroldis Chapman has a better chance of getting an out than Jimmy Haynes. The handedness of the batter or pitcher can matter. Handedness has been demonstrated to influence outcome. But that’s based on league-wide data over many years. Not on one hitter versus one pitcher.

Tango and Lichtman’s conclusion: “We’re not saying that it doesn’t matter which pitcher is facing which hitter. …However, you can’t tell by looking at the numbers from twenty-five or sixty plate appearances. There is simply too much noise masking the truth under those numbers.”

In other words, sixty highly specific plate appearances (same pitcher and hitter) is not enough evidence to overwhelm the knowledge from 1,500 random plate appearances by that hitter and/or pitcher over the course of their careers.

Instead of basing his decision on Ondrusek’s matchup data against Lucroy, Gomez and Ramirez, which was comprised of ten or fewer plate appearances per batter, Price should have been looking at Ondrusek’s career numbers or at least those from all of 2014. And by that criteria, of course, the choice was certainly Sam LeCure. If you want to push the #closerrulesstink envelope, bring in Chapman. But really, ABO.

[Interestingly, Lucroy had batted ten times against Ondrusek and only had one hit. But he had walked three other times, giving him an on-base-percentage of .400 against the Reds pitcher. (Please, please tell me that Price doesn't look at batting average.) By the way, Lucroy led off the inning with a walk.]

Not only are the sample sizes too small, but over time, the hitters and pitchers change. I remember a case where Dusty Baker explained that he chose Edgar Renteria to face a certain pitcher based on matchup data that stretched back over ten years, as if Edgar Renteria at age 34 was the same hitter as Edgar Renteria at age 24. It’s an utter waste of time.

If the data is so conclusive about the futility of using specific hitter-pitcher data, why do managers do it? Beats me.

Playing time, pinch-hitting and bullpen decisions can be complicated and therefore hard. Maybe that itty bit of data seems like an oasis in a sea of indeterminacy. It’s ultimately a mirage. When you hear Bryan Price explain that he used specific history for a decision, it should make you cringe a little. Because while he’s using data, it’s useless. And it’s important for managers to know which data matters and which doesn’t.

The matchup data does give managers a ready excuse when the beat reporters come around asking about a specific decision.

Last Saturday, it also gave us Logan Ondrusek in the seventh.

51 thoughts on “The futility of using pitcher-batter matchup history

  1. As you said, big league managers use these ‘stats’ all the time. What do you think it will take to change this practice?

    • Probably a confident, successful manager who says explicitly that he doesn’t use matchup history data and why. And before that manager will have to come a smart, confident, visionary general manager who hires him. Even then, this kind of stuff takes forever to change.

    • I think it will just take time. All the “old school” guys just need to get so old they are out of baseball. Perhaps 40 years in the future will be long enough to allow a majority of decision makers to understand the difference between data and useless data.

  2. God, yes. So much this. Announcers, of course, will also quote these numbers like crazy. Utterly meaningless crap.

    • “And you won’t believe this… Andy Stankiewicz is a .568 hitter against Randy Johnson. You really gotta favor Stanky in this situation.”

      (All stats made up, or your money back)

  3. I suppose it possible if someone knew the actual video of the at bats, they might know that the hitter had a problem with a certain pitch in the few plate appearances. Pitcher A throw more of a sinking fast ball that hitter A doesn’t like where pitcher B challenges up high and hitter A kills those kind of pitches. I doubt this is the case, but I say possible, especially in a division where they might break down the same hitters year after year and know them well.

    Most of these quotes are just grist for the mill and probably not wanting to air out the real situation as that isn’t public business.

    It’s probably more that Sam LeCure is probably dealing with some arm issues had pitched the day before and had gotten hit pretty good his last three appearances to Saturday. Hoover would have been the guy last year, but this year he is a mess, so it was Ondru’s shot Saturday.

    If it was September and going down to the wire, he might have gone to Broxton there and then tried to stretch between him and Chapman to cover.,.but it’s early June, not quite to panic button time.

    • One of the things that can work (and that, surprisingly Dusty sometimes did), is what’s called complex platooning. This is where you don’t worry so much about the handedness of the pitcher but about the kind of pitcher he is and how his repertoire matches up with your hitters.

      • For example, Donald Lutz can’t hit a breaking ball, so he shouldn’t be started against Clayton Kershaw! ;)

    • Good points. Might it also be possible that the managers talk to the pitcher (or hitter) to get hints about how they feel facing particular guys? There is probably no real science involved, in any case; it would boil down to executing. Sometimes you do, and sometimes you don’t.

    • I was standing in my mom’s kitchen, watching that game across the living room when it happened. Did you also experience an imminent feeling of dread when the pitch was on its way?

      • Yes. I seem to recall Weathers just missing on a two strike pitch and then I thought “That was his chance right there.” Albert then proceeded to hammer that point home.

        In the spirit of the article, I could say that it wasn’t just David Weathers that Albert owned, but all mediocre pitching.

  4. What do the advanced stats say about Justin Verlander’s decline the last couple of years? The risk of those long contract extensions and the Reds have 3 guys to consider.
    Personally, I think Verlander problems stem from a condition I wish I was afflicted with, TooMuchKateUpton-itis.

  5. I’ll bet the teams have cards they ain’t showing or talking about such as data if players would seem to be vulnerable to certain pitches and their hot and cold zones.

  6. It was kind of sad today to see Bronson Arroyo’s elbow is a mess and him have to go on the DL. Hopefully a few weeks off it will bounce back, he had pitched a few good games for Arizona this year.

    • Not wishing Bronson any ill-will, but I’m glad it’s not the Reds who are on the hook for his $23.5 million two year deal he signed in February. Maybe he had a premonition last September when he told MLB.com, “My value in the game could go down, so you’ve got to take opportunities when you can. If there are other teams out there that are willing to give me a three-year deal, and I couldn’t get that here (Cincinnati), it’s going to be very tough to stay.”

      More details regarding Bronson’s elbow: http://sports.yahoo.com/news/end-arroyos-career-sight-045941358–mlb.html

      • Yeah, that’s tough for Bronson and the DBacks. The guy is a ball-player and straight-shooter. One of my favorite all-time Reds. If he does decide that retirement is his best option, it is likely that he and the DBacks will negotiate a buyout of the remainder of his contract. If he decides to go that route, it’s a pretty classy move on his part. He could spend 2015 mostly on the DL and get paid for doing rehab. I’m hoping it doesn’t come to that and with rest, his arm will bounce back.

  7. Price also started Bernadina at first against Arroyo, which worked out but still seems like a terrible idea.

  8. My favorite stat type is the one like this (obviously made up numbers aside):

    Since 1990, the Reds are 33-6 when leading by 2 in the 9th inning.

    With the implication that the game is all but over since the 1993 Reds didn’t lose any games under that regime and that makes the stat look good. The “historical team results” are the best, because this team clearly has Barry Larkin playing shortstop.

    • Have you seen Little Big League? The announcer in the movie has a stat guy always handing him things like that to say on air, but much more strange, like “Player X is batting .400 this year on the road in a dome against teams north of the Mason-Dixon line!”

  9. While everything written here is 110% true, the reality behind using Ondrusek was probably more complex. Perhaps LeCure was not feeling right that night. Perhaps Price, recognizing that as recently as a week ago, Ondrusek had faced 35 batters without relinquishing a run and thought a a higher leverage situation was in order. Managers have to measure who is best in any given situation and weigh that against relationships. They have to let players know that they haven’t given up on them. In short, they have to manage for tomorrow, not just tonight.

    Fans don’t have to do that.

    I hated seeing Ondrusek coming in. I was hoping the Reds could sweep. That was a big series. But the reality is, we seem to be facing BIG series routinely. If it’s not the league-leading Giants, it’s the Dodgers and their vaunted pitching staff. Or it’s the division-leading Brewers. Shortly, it will be the Cardinals again.

    I keep reading in the comments what a lousy job Price is doing. It makes me laugh. Price has done myriad things differently (and better) than Baker. He’s been handed a rubik’s cube for a baseball team and he’s tried different combinations, which is a tribute to his inquisitive and open mind. I’m convinced he’s mitigated a horrible string of injuries that would find a lesser manager ten or more games under .500 by now.

    But if you want Walt Jocketty and Bryan Price to be Mark Friedman and Joe Maddon, you are going to be sorely disappointed. The regime at the top is conservative and old-school in their baseball thinking for the most part. With a younger, more progressive GM and organization behind him, Price might be making even more sound data-driven decisions.

    Appreciate who you have, Redleg Nation. Bryan Price is a vast improvement on the past.

    • Leadership and change must come from the top and WJ is at the top of the Reds pyramid. This is the final year of WJ’s current contract, so BC doesn’t have to fire or dismiss WJ after this season, just not renew his contract. I would really like to see a younger, more progressive GM at the top of the Reds pyramid, but the unknown can be scarier and riskier than the known and WJ has done a lot of things right.

      Richard, I agree that Bryan has done an admirable job under very difficult circumstances (injuries, lineup holes & a totally unreliable middle relief staff). A reasonable evaluation of Bryan’s performance must coincide with his chance to manage the team under more reasonable circumstances.

      • I just don’t get all the grief that Price catches. Richard and Cossack are spot on and right on point that the Reds are lucky to have Price as Manager. The learning curve is vast, and Price is ahead of that curve.
        As for Jocketty as GM, he has done things well and he has done things poorly. Its about 50/50 with Jocketty. But a GM for a contending team, a team that wants to get to the World Series, needs much better than 50/50 from the GM to reach those goals. Since 2010, Jocketty has failed to get this team to a World Series. Failed BIG TIME. Time for Big Bob to get someone else to try their hand at it.

      • I think it’s going to be very interesting after the season when WJ’s contract is up (assuming this team does not make the post season). Will Big Bob stay with the guy he knows and likes or let him walk and go to someone else that’s more modern thinking? We know he’s not afraid of change or Dusty Baker would still be the Reds manager. I don’t believe that, baring a second half collapse, that Price has anything to worry about.

    • I want to join Richard’s comments about Price in general. Price has had a difficult situation with all the injuries. He’s been open minded about plenty of decisions compared to the previous manager. I support the job he’s done so far and even hold out hope that he’ll become more progressive as he develops as a manager. The talk of firing him now or at the end of the season is ridiculous, in my mind.

      That said, I hope we can handle the nuance of generally liking the manager (or a player) but still finding ground for improvement. I don’t think Richard is suggesting that just because he likes the job Price has done overall (as do I) that Price should be beyond criticism.

      I’m sure the Ondrusek decision was based on a number of factors. But the one factor Price did mention to the press was the match-up history. And that’s simply data that he shouldn’t even be looking at, let alone factoring it into a decision.

  10. It’s great to think that a managers and GM’s will be more from the “new school”. But it’s also hard to believe it will ever happen because of the fact that coaches now that will eventually become mangers learned under the same old regime. The same practices keep getting passed on from generation to generation. Every now and then someone will think differently and only then does a little bit a “newness” get introduced to managing.

    • You pretty much have it spot on. Players for the most part are more ‘traditional’ in their thinking as well. Players become coaches, who become managers/GMs as well. It’s kind of a slow go, especially for managers. As more ‘progressive’ GMs win pennants, I could see front-offices continue to ‘evolve’ at a quicker rate than on the field leaders.

    • The next manager for the Cincinnati Reds…Ryan Hanigan. There may not be a major league manager more ‘old school’ Clint Hurdle, and just look what a little progressive thinking (and mandates) from the GM to an old school manager accomplishes.

        • I think WJ is a bit of a victim of his success. I recall reading comments from other GMs stating that you “rarely get the best of Walt” or whatever in trades. If that perception is true, it helps explain some of the inactivity on the trading front.

  11. I agree starting people because of 3 AB or something is ridiculous, but I think there’s still some times that some people just own other people. Maybe they pick up the ball really well off your delivery for some reason, maybe they pick it up really badly. Maybe they throw a really nasty pitch that happens to be your weakness.

    Part of it may be psychological, the player knows he doesn’t have success against someone, and it fuels them not having success. Like Jay Bruce. I don’t care if he has 200 AB versus him, I don’t think he’ll ever be great against Liriano. 0-14, 6K. Or Wandy Rodriguez, 3-37, 1HR, 16K…. but he owns Lance Lynn 9-19, 2BB:6K.

    Aramis Ramirez has shown he consistently owns Cueto. 14-42, 5 HR, 6BB/7K.

    Homer Bailey owns Andre Ethier with 0-21, 3BB/7K

    …I think that’s the major problem with Sabermetrics. If it can’t be proven by their numbers, then it’s silly and doesn’t exist. Matchups are totally a thing, just not 3 AB matchups. When you watch someone play though and they look completely outgunned (see: Bruce v Liriano) then you just know it’s not a good idea, regardless of how many AB in that sample size.

    Human element gets discounted too much by Sabermetrics. You might not be able to explain exactly why, but some guys just are other guy’s kryptonite.

    • Trot Nixon, a career .274/.364/.464 hitter had a 1.450 OPS in 40 AB against Roger Clemens, a 7 time Cy Young winner. Sometimes people just have your number. Happens in all things, not just baseball.

    • This isn’t really about sabermetrics or “their numbers” it’s about science vs. anecdote. The study I cited looked at all the actual real hitter-pitcher match-ups for 60 years. That’s studying the human element. How is that sabermetric? It’s just being awesomely comprehensive. Yes, there are definitely handedness splits, as I mentioned in the article.

      The “just have your number” theory has been disproven by those studies. There will always be outcomes like Nixon/Clemens because in any large number of events, you’ll have outliers in the tails of the probability distribution. What these studies found is if you take all the matchups similar to Nixon/Clemens – 300 of them – where one person seems to own the other, and you look at a subsequent period, there is zero ownership. That’s not based on a statistical theory, that’s based on actual data of real major league baseball players. All of them, in fact.

      • Ok, Steve, playing devil’s advocate here. You are the manager of a team. Bottom of the 9th, runners on first and second, down by a run. Your weakest batter (probably the pitcher’s spot) is up. The opposition brings in a LHRP. You have two guys on the bench a RH who is 1 for 23 with a HBP against this guy. You have a LH who is 3 for 8 with 2 HR against the guy. This is all the info you have (they have cleaned house and just brought you in). Who you going to bat, the LH, the RH, or the pitcher, and why?

        • You mean I don’t know the career batting average of my two hitters, or what their career splits are against LHP? That seems pretty unrealistic. Those would be the stats I’d look at to decide. The LH hitter might be Joey Votto and the RH hitter might be Willy Taveras. I’m sending Joey Votto up because he’s a better hitter than Taveras over 1,500 at bats, not because of the 8 he’s had against this one pitcher.

        • Ha ha ,good answer, I should have given you at least season averages. So if it was Joey and Willie you would bat Joey, even if you had watched Joey look lost against the guy 23 times and Willie look like Willie Mays? How about Jay Bruce and Ludwick?

        • Likewise, if the RH hitter was Pete Rose in the prime of his career and the LH hitter was Mike Costanzo, I’d send the Hit King up there.

        • I could see matchup history being used as a sort of tie breaker (i.e. having the choice of two similar right handed hitters against a LHP and going with the one with slightly worse numbers against LHP because he has hit this pitcher well over a reasonable sample size, say 35 ABs.) Also do you think it would be more significant against pitchers whose stuff fall well outside the norm, like against an R.A. Dickey or a sidearm pitcher?

        • There is data to support “stuff” match-ups. Like certain hitters can’t hit a curve ball. But you don’t need history data for that.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s