Brad Ausmus on Analytics, Closer Mentality, and Pitch-Framing
Brad Ausmus is hard to label. Many see him as an old-school manager — and, based on some of his beliefs and actions, that’s perfectly understandable. On the other hand, he’s Ivy League-educated and well versed on most analytic concepts. From a knowledge standpoint, the manager of the Detroit Tigers is far from a troglodyte.
This interview doesn’t add much clarity to Ausmus’s identity. For one thing, it’s narrow in scope. While other subjects are touched upon, closer usage and pitch-framing comprise the bulk of the conversation.
Of note: this material actually comes from three separate conversations. The first two were in group settings with Detroit beat writers (with my questions eliciting most of these responses) on back-to-back days. I then had a shorter, one-on-one conversation with Ausmus to fill in a few blanks. Because of the manner in which these quotes were obtained, some have been resequenced for continuity.
———
Ausmus on analytics (intro): “Analytics are ubiquitous. I think the dangerous mistake people make — some member of the media make — is believing that they can’t be flawed, because they’re based on numbers. That’s absolutely false. Numbers do not always tell us the whole story. And there are certain things in baseball, because it’s played by humans, that numbers will never be able to put a value on.”
On leverage and closer mentality: “A lot of people in the analytics world think you should bring in your best pitcher in the biggest point of the game. Well, excuse my French, but who the (bleep) knows when the biggest point in the game is until the game is over? You don’t know. It may be the sixth. It may be the ninth. The problem is, if it’s the sixth and you use your closer, and all of a sudden you have a one-run lead in the ninth, who is going to close? You don’t have that guy anymore, because you burned him.
“Anyone who says you have to bring your closer in early, or in the biggest point in the game, has a crystal ball. That argument goes out the window for me. I don’t mind second-guessing, but second-guessing the biggest point of the game after the game? It’s easy to tell then. It’s not easy to tell in the seventh inning.
“If you’ve got another guy who is a closer, you’re fine [bringing your closer into a high-leverage situation early and not having him pitch the ninth inning]. But if you don’t have another guy who is a closer… closing a game in the ninth inning is not the same as pitching the eighth inning, or the seventh inning.
“Could you walk across this table? What about if I put it a thousand feet in the air? Could you walk across it then? That’s the difference between pitching the sixth and pitching the ninth. Anyone who says that pitching the ninth is the same as pitching any other inning is out of their mind. They don’t get baseball. They don’t understand the adrenaline that’s involved; they don’t understand that the 27th out is the hardest one to get. It’s not the same. That’s why proven closers get so much money. It’s a very difficult thing to do on a daily basis.
“[If] you’re making the assumption that whoever is second best, to your best pitcher, is going to finish out the ninth without a problem… that’s a false assumption. You don’t know what’s going to happen. But I do know that my closer has a better shot at closing out the ninth than my second-best pitcher.”
On identifying future closers: “You look at the stuff and try to project, but I think the truth is, you don’t really know if someone can close until they close. Again, it’s different. It’s much more stressful. It’s the most high-leverage situation you can pitch in, the most high-leverage role you can pitch in.
“It’s difficult to predict how people will react to having the entire weight of the game put on their shoulders on an every-two-night basis, or every-three-night basis. The hardest part about being a closer is knowing that if you fail, and your team loses, you have to be able to come back the next day. If you pitch the eighth inning and give up the lead, you still have a chance to get the lead back. The ninth inning, especially on the road, it’s over. There’s a psychological… you need a psychological strength to deal with that.”
On year-to-year reliever volatility and projection: “Again, I think it comes down to the stuff. You think someone should be able to replicate what he did based on stuff. But there are other things that go into it. How does he interact with teammates? What is his work ethic? What’s his body type? Is he prone to injury? What’s his arm action?
“Generally, Joe Average Reliever is tough to predict. The elite are a little easier to predict — their track records [make them] a little easier to predict — or maybe I should say they’re a little less difficult to predict.”
On pitch-framing: “When pitch-framing came out, I was a special assistant in San Diego. I said, ‘Look into the pitchers that are involved.’ It turned out there was no correlation in terms of reputation. When they looked at it, empirically, there didn’t seem to be a connection between the pitcher’s longevity in the game, or his reputation in the game, and strikes being called. I would have thought there would be. Maybe it’s changed. I don’t know. But I thought a guy like [Roger] Clemens would get calls over a rookie.
“I think people got excited about pitch-framing numbers [too] quickly. They wanted it to be… I think it may have been prematurely overvalued. I think if we can actually use the numbers — know they’re not flawed — they are valuable, because pitching is such a big part of the game, and one pitch can change a game. But I don’t think we knew what we really had.
“Framing is something where you have to have thousands of pitches for it to have any value. James McCann has improved over his first three years. Alex Avila has been kind of up and down with his framing. So I don’t know if there is any application to our pitching staff, and our catchers, with those numbers.”
On pitcher quality and framing: “If you have good pitchers, you’re probably going to have good results, regardless of the catchers’ framing. It can have an impact, for sure, but Chris Sale is going to be good whether there’s a catcher back there or a piece of wood.
“Numbers are very valuable, but sometimes they get way overinflated in today’s sports-media market. Most teams use a lot of numbers to put their team together, but once you have them, there’s not much you can do with them.
“You have catchers who are ‘good receivers’ and get strikes called, or maybe you have ones who don’t. But what are you going to do about it? You can maybe work on things and try to make them better at it, but that’s about it. If you have outfielders who can’t run balls down in the outfield, the only way you can change that is to get guys who can run the ball down in the outfield. You can try positioning differently, but I don’t think that would make a vast difference.”
On analytics (coda): “I was texting back and forth with [director of baseball operations] Sam Menzin earlier today. I was checking a bunch of different things. I’ve always told him, ‘Do some free-thinking; think outside the box.’
“We’re open to anything if it has some basis, some merit. You might as well use the information you have. What’s changed so much in the game is the amount of information. You have to be able to decipher it and make it usable.”
“When you’re making decisions, and you’re using analytics — you’re using numbers — you’re playing the odds. I did it as a catcher. I played odds when I was calling pitches for pitchers. You’re playing the odds and trying to get the best results for that particular pitch, for that particular game, and over a six-month season. As a manager, a decision in a given game may not work, but if you’re playing the odds, just like you would in blackjack, you’re going to win in the long haul. In that sense, a manager is like a gambler. That’s how I look at decision-making and analytics.”
David Laurila grew up in Michigan's Upper Peninsula and now writes about baseball from his home in Cambridge, Mass. He authored the Prospectus Q&A series at Baseball Prospectus from December 2006-May 2011 before being claimed off waivers by FanGraphs. He can be followed on Twitter @DavidLaurilaQA.
Great analogy about walking across the table
I’m not so sure that one worked for me. Yes the 9th inning must be very stressful. However pitching to protect a small lead in the 7th or 8th inning of a major league game against the best hitters in the world when you have millions of dollars in future money making potential in front of you sounds awfully stressful to me too. Maybe the 9th is like raising that table 1000 feet. But if we’re going with that analogy, I only think it would be fair to liken a high leverage relief situation to raising that table to 800 feet. The prospect of failing either one sounds stressful to me.
You can still fall off the table and die (blow the game) in the 6th inning. The 6th inning is not like walking across a table on the ground.
A better analogy might be that the ninth inning is like doing it naked: it’s the same task but you’ll look a lot more ridiculous if you fail.
In all seriousness, how many recent examples are there of lights out setup men who for some reason, couldn’t hack it in the 9th inning?
Ausmus doesn’t come across as being too bright. Analytics are of limited value because all they do is tell you who’s good and who’s not? Yeah, why would that be of any import…
Hunter Strickland, for one. 7 blown saves in 10 opps. from 15-16. If you really think every quality set-up man can be a closer, you must not follow baseball very closely. Tony Watson, Bryan Shaw, David Phelps….all recent examples of quality pitchers who have struggled in the 9th.
But you would expect some number of those just due to random variation. Relievers are known for being good one moment and bad the next. In fact, we should expect this to happen on a semi-regular basis due to selection bias, when this happens:
1. Good reliever starts pitching really well, maybe partially because he has improved but also partially because of small-sample cluster luck
2. Reliever gets promoted to closer because of #1
3. Reliever starts pitching worse, maybe partially because he actually got worse for whatever reason that good relievers sometimes just get worse, but also due to bad or regressing cluster luck.
4. Reliever feeds narrative that “some guys can’t handle the ninth inning.”
As always, even if there was no “ninth inning effect,” it would still be incredibly unlikely that every good 8th inning reliever went on to be good in the ninth.
These are guys with repeated opportunities, not one season of bad luck.
Do you also believe that it is impossible to prove that a football kicker is an under-performer in high-pressure situations?
In general, do you believe that athletes all have the same psychological strength?
You would expect even guys with repeated opportunities to sometimes happens to fail in the ninth more often just due to chance. What you would *not* expect is for every reliever to have a stable true talent, or to always have performance consistent with that true talent, in every situation, and so for no reason other than chance you would expect relievers to perform differently in the eighth and ninth. The trick is to discern the real effect through the effect of chance.
My beliefs are not relevant here because I am not claiming that psychology does or does not have an effect on ninth inning performance, I am merely pointing out the pitfalls of trying to draw conclusions from purposely selected examples. What you’re doing is basically just confirmation/selection bias.
If psychology explains how a closer does, Brad Lidge must have been bipolar.
They’re solid bullpen pieces, but none of them are elite or would be the best RP on a good team. Besides, one could easily point to any number of guys who got anointed as “proven closers” despite not being that good, eg John Axford, Jeanmar Gomez. Or conversely, guys who were “just setup guys” who became “proven closers”.
It seems more likely that while certain exceptions may exist, the label of “proven closer” is irrelevant, and it mostly just comes down to a combo of sample sizes, luck and obviously, how dominant you are.
Put it another way, if clutch hitting doesn’t really exist for hitters, why would one think that clutch pitching is a thing? Remember when people used to think Barry Bonds couldn’t hit in the playoffs?
Jeanmar Gomez is a proven closer in which timeline?
Kelvin Herrera.
Well, he’s handsome
Yes, he is!
On high leverage situations, you don’t know for certain when the highest leverage will be in a game until it is over. However, it isn’t hard to figure out that when in-game leverage is higher than a typical closer save situation. Deploying a closer in a situation more critical than closing and ending up with a lead in the 9th doesn’t sound like a failure to me. Astros have two guys that they are deploying in higher leverage situations than their closer.
That said, GM has to give manager at least two guys that can handle leverage as identifying leverage may be easy, but winning high leverage battles early in the game taxes pen and typically leads to high leverage situations later in the game.
I think his point about identifying the highest leverage situation after the fact is a good one. It’s not just recognizing that hey, I have a high leverage situation on my hands! You have to have your best guy warmed up for that purpose, in that moment. It’s not so simple. You do have to have some sort of an innings based plan so that your relievers have some idea of when to begin getting loose and mentally prepared before they even start throwing. There aren’t too many guys who can jump up, throw for a minute and be prepared to come in because what looked like a low leverage situation quickly became the potential turning point in a game. I’d be curious to see what would happen if Ausmus were actually given a legitimate major league bullpen to work with.
I disagree, it’s a poorly thought out argument. When faced with the certainty of an important inning, why save your relief ace for the mere possibility of a more important inning later, which may never arrive if a lesser pitcher blows it now? How many games are lost in the 6th, 7th, 8th innings because the manager saves his “closer” for a 9th inning save opportunity that either never arrives or is not high leverage?
I’d hope my manager is prepared enough to anticipate when high leverage situations may arise based on our SP’s stats 3rd time thru the order, the opposing batting order, and our bullpen availability.
So basically you think the manager should have a crystal ball.
It sounded like he was saying the manager should know his team, the other team, and the situation at hand.
Sorry, but I don’t get this argument. He said ‘faced with the certainty of an important inning’, which is a bit silly. There are a lot of important innings. If your starter only goes 5 and leaves with a 1 run lead and the heart of their order is due up in the 6th, do you go to your closer then? Have him pitch an inning and keep the lead and then use 3 lesser relievers for the rest of the game with progressively fewer chances for your offense to get back any runs given up? If the heart of their order is due up in the 6th, it’ll be due up again before the game is over. You’ll need a crystal ball to know which of those instances will be higher leverage, but if you’re playing the odds then you’ll save your closer for the later chance and count on your offense to get some runs if necessary.
If it’s the 8th inning and the heart of their order is due up, sure, use your closer if he’s the best pitcher available, but it’s not ridiculous for a manager to be reluctant to have some scrub to pitch the 9th with zero chance of a comeback. What’s the point of holding a lead in the 8th and blowing it in the 9th versus blowing the lead in the 8th. At least you’ll have some tiny chance for your offense to do something.
This is of course all assuming your closer is your best pitcher, which is not necessarily the case. This is the Tigers bullpen, after all, which is generally composed of exactly one good reliever and a cast of scrubs.
No, but a reasonable ability to look at odds and assess probabilities doesn’t seem too much to ask.
“You’re down by one run and have the bases loaded with two out in the 8th inning with a .190 hitter at the plate. Do you want to use your ace pinch-hitter, or save him for later?”
“How can I possibly decide? I don’t have a crystal ball!”
A lot of so called “fans” think that. SMH….
I’m glad to hear someone finally come out say that there was a pitch framing “bubble” at one point, where every blogger and his brother were claiming to have a precise value for each catcher’s future run-saving skills.
In general, the metrics that gain a team an edge are not the ones being beaten to death all over the internet.
Disagree slightly. Calling PF a bubble implies there was nothing there and that’s just not true. Ryan Doumit was considered capable enough until it could be reliably qualified how much he cost his team behind the plate. Granted no one thought he was Pudge back there but he was previously considered passable by a MLB front office.
Good receivers are good receivers in the way good OF are good OF. KK & Pillar are going to be good year in and year out but some years avg OFs are going to convert more 50/50 chances and look better by the numbers than they might otherwise. It doesn’t necessarily mean they unlocked something in their abilities long-term. Pitch framing deals in the same realm, some years average receivers will look good but you can’t overreact and say “Iannetta is a Molina now!” If you want to say small samples of PF data are beaten to death on the internet I won’t argue but don’t throw the good out with the bad. Accept that it’s as noisy as all other fielding data and draw conclusions from there.
framing is huge. It is real. We have exact precise locations of where the balls are crossing the plate due to our techonology. The teams that utilize it will continue to steal wins from people in denial.
It’s true that framing correlates year-to-year, but there’s also some evidence that the year-to-year correlation has declined in recent years as framing has been discussed more – http://www.fangraphs.com/blogs/the-beginning-of-the-end-for-pitch-framing/
So there is a difference between pitch framing and CF defense, in that there’s a risk that once it’s discussed a lot that a certain catcher is good at framing, umpires may react (perhaps subconsciously) to lessen, but not eliminate, the impact of good framing by that catcher.
Similarly, my understanding is that the home plate umpire gets a report card evaluating balls and strikes call after each game, so it’s easy to think that umpires could pick up on what catchers are “stealing” a lot of strikes and adjust accordingly. The overall trend through at least 2015 has also been for umpires to be more accurate on pitch calls – https://fivethirtyeight.com/features/umpires-are-less-blind-than-they-used-to-be/ – which logically suggests less impact from good framing over time.
Relief pitchers all basically say what Ausmus said, that pitching the 9th is not the same thing as pitching earlier in the game. The psychological factor is too important not to consider.
I guess we can consider it, but then the question becomes what should we do with it? Should we hold back a guy from being a closer if we believe he isn’t fit to handle the ninth psychologically? If we promote a guy and he struggles, should we have a short leash because he showed he can’t handle it? Should we have a long leash because we believe he needs to adjust to the pressure? There aren’t a lot of clear prescriptions.
What Brad Ausmus said was the way you find a closer is to have a pitcher close. That is the opposite of holding someone back.
Huh, the guy who doesn’t really believe in framing numbers is currently running out the number 71 and 78 (out of 79) ranked catchers by Catching Runs (a combined -11.2 on the season).
And wouldn’t you know, the Tigers are 24th in team K%-BB%, in between such luminaries as the devastated Mariners and the (high walk) Marlins.
Huh, a guy who thinks that Baseball Prospectus’ numbers are gospel. Just because they’re the only numbers available to the public doesn’t mean they’re accurate or that teams don’t have different numbers. Prospectus’ catching numbers beyond framing are not good and teams certainly value more than just framing.
Plus I don’t think the Tigers’ pitching staff sucks because of their catchers, like you seem to be insinuating. Might be the other way around. According to BP, Avila had good numbers with a good pitching staff but as soon at the Tigers’ pitching staff sank into the gutter his numbers suddenly don’t look so great, which makes it seem like he’s a neutral framer and his ability to get calls is based on the quality of his staff. As far as deciding who gets the blame for a called strike between the catcher, the umpire, and the pitcher, I think that’s something Prospectus can’t accurately measure because all they have is the pitchf/x database. They’ll always lag behind guys like Baseball Info Solutions, which has a way to track pitcher command and use that in their framing numbers. And wouldn’t ya know it – last year BIS had Avila as a neutral framer, but BP had him as terrible. BIS actually has Avila as a fairly neutral framer his entire career. The only season where he wasn’t between -1 and +1 was 2015 (-5), but right back to 0 in 2016. So you’ll get different framing numbers depending on what information you have and how you use it. I agree with Ausmus when he says fans are jumping the gun a bit. Yeah, we have numbers. But how useful they are when their accurately is in question is debatable.
If you accept that the -11.2 runs is fully legitimate, it still gets wiped out by Alex Avila’s batting runs.
Kudos to Ausmus for giving the best argument I’ve heard for hanging onto your closer until the 9th. Although one could correctly argue back that you actually need to survive that critical 6th inning jam to even GET to your closer, his rationale makes the case that closer usage is more of a chicken-and-egg dilemma than a cut-and-dry decision.
My problem with his argument regarding closer usage is that he’s a major league manager who has played/seen thousands upon thousands of games. He doesn’t have a crystal ball but he SHOULD have a strong intuition as to how different games unfold and when the game tends to be determined most often.
Also, what does it do to a reliever when you tell him, “You’ve got good stuff but we don’t think you have what it takes psychologically”? When a manager doesn’t trust his other relievers to go out in the ninth, wouldn’t those same relievers lose confidence in themselves as a result? And then when they do get the rare chance to pitch the ninth, they pitch like crap because they aren’t as confident as in other innings.
Great interview.
If those comments aren’t purposefully obtuse to keep from saying anything meaningful, then I think we know pretty clearly where he stands on analytics. This is probably his last job as a manager; there just aren’t a lot of organizations that will hire him because he’s pretty.
This won’t be Brad’s last mgr job unless he wants it to be. When it looked like the Tigers might let him go in 2016, several teams called about his availability for mgr interviews. No they won’t hire him bc he’s handsome but bc he knows his stuff and players like and respect him. Jealous much??
Ausmus makes a slightly weird argument. He alludes to “playing the odds” as a catcher, and explains situations where he did this. Yet seems to gloss over the “playing the odds” aspect of using a reliever in a higher leverage situation. Yes, bringing in your best guy in the 6th or 7th isn’t guaranteed to work out, but by “playing the odds”, you’re giving yourself the best opportunity to ensure you get the most of the situation.
Understanding leverages isn’t rocket science. And while I completely get that numbers do not override the human element, it seems Ausmus cherry-picked when the argument for playing the numbers game worked best.
Um, turn my comment into a +1 for this one.
It’s funny that he knows managing’s about playing the odds for the long run, but he doesn’t go for that idea when it comes to trying to use your best pitchers in the most important spots — just goes all ¯_(?)_/¯.
“As a manager, a decision in a given game may not work, but if you’re playing the odds, just like you would in blackjack, you’re going to win in the long haul.”
vs.
“who the (bleep) knows when the biggest point in the game is until the game is over? You don’t know. It may be the sixth. It may be the ninth. The problem is, if it’s the sixth and you use your closer, and all of a sudden you have a one-run lead in the ninth, who is going to close?”
I pretty much agree with Brad Ausmus’ point about not knowing when the crucial point of a game is beforehand. It would be interesting to know which out is statistically the most difficult to get. Something like OBP broken down by number of outs.
Other managers don’t have any trouble figuring out when to bring in Andrew Miller, but it sounds like Ausmus would have no clue.
Ausmus’ point about crystal balls and closer mentality are well taken, but I don’t think he articulated the point very well.
Identifying a particularly critical situation in the early to mid innings is relatively easy. The issue is that “leverage” in the most general sense is also affected by the consequences of decisions made earlier in the game (or even series, if speaking about the playoffs).
Bringing in a relief ace to bail out the struggling starter in the middle innings to prevent an early blowout may be the technically “correct” move at the time, but not having that reliable bullpen arm available will magnify the impact of each subsequent critical at-bat, to the point where in hindsight, the 9th inning prototypical save opportunity could become the highest leverage situation in the game, and the manager is out of bullets.
Of course, refusing to use the best reliever in any inning but the ninth because you can’t see the future is an absurd mentality, but reducing the hypothetical scenario to a sixth or seventh inning situation to face the heart of the order with a slim lead on the line doesn’t do the issue justice, either.
At a minimum, the manager needs to have faith in two relievers: a “stopper” who is ready and willing to enter the game at any time, in any situation, and a “closer” who is able and willing to occasionally get the last 4-5 outs.
Fair point.
I think another big point over the course of a season is that having a “closer” or even “stopper” ready to enter almost every high leverage situation in the middle of the 6th or 7th inning logically also implies that this pitcher will be up and throwing a lot in fairly low leverage situations – such as a runner on 1st with a 1 or 2 run lead in those innings – that don’t actually develop into high leverage situations.
That mindset can and should change in the playoffs due to the number of off-days and the importance of each game, but bullpen use in the regular season needs to consider more than just the current game.
Speaking as a Tiger fan… There’s a chicken-and-egg argument with Ausmus and bullpen usage. Does it seem like he is a great bullpen manager? No. Has he ever had a reliable, quality bullpen? No. So who knows how good he actually is.
He seems kind of ignorant on it. Well all i’ll tell him is he can watch the Cleveland Indians and the Houston astros leave everyone in the dust. Its no secret they are the 2 most analytical, Indians won world series last year, and the astros should win it this year. While doing it with some of the smallest payroll .
You really don’t want to have to explain the Kansas City Royals.
Um, Guru you might want to re-educate yourself. Indians were runners up last year to Cubs in WS.
In Ausmus’ defense, he HAS been using the Tigers’ best reliever in high-leverage situations (Shane Greene). Greene has pitched as early as the fifth inning multiple times and almost always comes in with a lead or tie game.
Well, he’s too sexy to disagree with so I’m all in with what he’s saying! Just kidding-that everyone always tries to devaluate his knowledge by mentioning his looks or ivy league education sounds like jealousy to me. Whether with the Tigers or another team, he will be a manager. He knows his stuff. Already has improved in last few yrs. Remember, he had no mgr experience to speak of. People seem to forget that except in a condescending way. “Excuses my French”… I hadn’t heard that in so long! LOL!
Say a prayer for the Tigers.