Andrew Miller Is the Perfect Relief Pitcher
As the Blue Jays take the field in Toronto tonight, they’ll find themselves down two games to none in the ALCS, with Game 3 representing as close to a must-win game as you can get without actually facing elimination. Teams have come back from down 3-0 before, of course, so the Blue Jays aren’t definitively done if they can’t figure out how to win on Monday night, but having to win four straight games against any good team is a massive challenge. And the idea of winning four straight against Andrew Miller’s team seems downright impossible, because right now, Andrew Miller is basically the walking embodiment of the perfect relief pitcher.
Miller’s numbers in the ALCS are just silliness. In the first two games in Cleveland, the Indians asked Miller to face 12 Blue Jays hitters, and two of them put the ball in play. Ezequiel Carrera had the second best outcome against Miller when he grounded out to second base on Saturday. Josh Donaldson is the only Toronto hitter to reach base, which he did on a single up the middle on Friday night. And even that seems like kind of a miracle.
In the first two games, Miller has thrown 55 pitches, 31 of which have been sliders. The Jays have swung at 16 of his sliders; they have swung and missed on 12 times. Three of the four swings at Miller’s slider that haven’t been whiffs? Those were fouls. On 16 occasions now, the Blue Jays have chased Miller’s slider, and only once (Donaldson’s single) did it lead to anything besides a strike on the hitter. Of the 15 at which they didn’t swing, only five were called balls, so taking the pitch wasn’t exactly a winning strategy either. The Blue Jays haven’t been able to even come close to hitting Andrew Miller’s slider, and like a wise man, he just keeps throwing it.
Two dominant games do not an elite player make, but Miller has quietly been the best reliever in baseball for a while now. While Aroldis Chapman throws the hardest, Zach Britton will get the most Cy Young votes, and Kenley Jansen has been consistently incredible for the Dodgers, no reliever in baseball has been as dominant as Miller over the last three years.
Going back to the start of the 2014 season, here are what batters have done against Miller in the regular season.
| Innings | BA | OBP | SLG | wOBA |
| 198.1 | 0.153 | 0.219 | 0.256 | 0.212 |
Those are hilarious numbers, but they actually pale in comparison to what Miller has done to opposing hitters in the postseason. He’s pitched in the playoffs each of the last three years — granted, only one inning in the Wild Card game in 2015 — and his postseason numbers make his regular-season numbers look like a slump.
| Innings | BA | OBP | SLG | wOBA |
| 16 | 0.080 | 0.145 | 0.120 | 0.131 |
The 55 batters who have faced Miller in the postseason have a combined wOBA 16 points lower than what pitchers have put up, while batting, during the same timeframe. And keep in mind that Miller is being used against the best hitters on the opposing teams, and those are the guys he’s turning into the equivalent of a bad-hitting pitcher at the plate. That is nothing short of remarkable.
When you fold Miller’s postseason numbers into his 2014-2016 performance line, you find that hitters have put up just a .207 wOBA against him over the last three years, over 818 batters faced. That’s the kind of wOBA you see from guys like Erik Kratz, the veteran backup catcher who bounced around the league and got 230 PAs over the 2014-2016 seasons. That’s the average performance of the guys who have faced Miller over the last three years, now spanning a total approaching 1,000 plate appearances. While being selectively used to only face the other team’s best hitters.
A couple of years ago, Wade Davis was at this level. He held batters to a .196 wOBA in 2014 and a .202 wOBA in 2015, though he did it more with weak contact than Miller’s no-contact approach. But remember back to 2014, when Davis was free to pitch the middle innings while Greg Holland racked up the saves in the ninth inning, and that seemed like the absolutely perfect October bullpen? In 12 games pitched during that playoff run, Davis faced six batters twice and seven batters once, in the seventh game of the World Series. He never faced eight batters in a game during that run.
During the first four game the Indians have played in the postseason, Miller has already faced eight batters twice, and then faced six more in his other two outings. Miller is matching peak-Davis’ dominance while extending the workload to levels that other pitchers who have gotten to this level just haven’t matched.
The most recent example of this kind of dominant relief weapon in the postseason? Mariano Rivera, of course. In 2003, he faced 55 batters in eight games, and they hit .127/.127/.182 against him. In that run, he faced eight batters in the fifth game of the ALCS, then 11 batters in the seventh and deciding contest, then faced seven batters in both of his World Series appearances. That Miller is even invoking Rivera tells you how dominant a force he’s been for the Indians thus far. That Terry Francona feels free to use him in any spot, to squelch any rally, makes him the kind of bullpen weapon not seen in a very long time.
Andrew Miller, right now, is something not that different from a left-handed Mariano Rivera. And given the way the Indians are using him, he’s about as valuable as any relief pitcher can be. We haven’t seen him make it through a full playoff workload yet, and perhaps the heavy pitch counts will catch up to him at some point, but right now, Miller looks like an unhittable machine bent on destruction of the Blue Jays lineup. If Toronto is going to battle back and take this series, they’re going to have to figure out how to hit the most unhittable pitcher in baseball right now.
Good luck.
Dave is the Managing Editor of FanGraphs.
Dave,
How good would Miller have to be, or how many innings would he have to pitch for him to eclipse Kluber in value delivered?
Is there an argument to be made for Miller as series MVP?
Depends how you want to measure “value”. He’s got a terrific chance of beating Kulber in WPA simply by virtue of the leverage of the innings he appears in. That’s not terribly uncommon for relievers generally. He won’t throw nearly as many total innings, however, so something like pitcher WAR will always favor the starter.
Kluber will likely affect two games. Miller could be a factor in four or five. If he’s seen as the deciding factor in those four or five appearances, Miller certainly could win a narrative-driven award like the MVP.
Using FIP and IP so far: we have Kluber with 13.1 IP and a 2.55 FIP and Miller with 7.2 IP and a -0.51 FIP, so Kluber has an extra 5.2 IP over Miller but a FIP 3.06 higher. If you use RA/9 then Kluber just has an extra 5.2 scoreless IP over Miller, and if you use xFIP then the difference is 3.52 in Miller’s favor. Obviously they both have very small sample sizes but if you think FIP or xFIP is a better measure of value than RA/9 I would argue that Miller has been more valuable in his 7.2 IP than Kluber in his 13.1 IP even without factoring leverage into the equation. For the record though, Kluber has a .57 WPA and Miller has a .56 WPA so far.
Did not know it was possible to have negative FIP. Damn.
Miller is basically unfair right now. I’d be happy to do that in a video game.
A player with these stats in a video game would be a hack.
He’s lost a couple of miles per hour on the fastball in his outings against Toronto. It is the two different sliders that are driving hitters nuts, but it’s mostly about recognition. One has a nasty break and is unhittable but is almost always out of the strike zone; the other is somewhat lazier but he’s got good control of it and can hit a spot. He is throwing more sliders than fastballs, and so job 1 has to be recognition of the sliders- and driving the lazy one on the outside corner the other way.
I am pretty sure that he’s not going to be as good for the remainder of the playoffs. He’s absolutely fabulous when he has Randy Johnson’s fastball to go with the sliders (as he has in previous years), but he doesn’t now and it looks to me like the hitters will be adapting very shortly. And then he will be very good but mortal.
Between today and the end of October, I’ll happily take the other side of that bet.
Where’s this velo data? Having trouble finding it here as it looks like the Pitch FX data only captures regular season.
It isn’t just miller’s ability which makes him valuable. He has zero ego in terms of how he is used, which is not at all the case with many top relievers, and this means that his manager can use him so as to be most helpful to the team. This makes a big difference too.
I wouldn’t say it’s a lack of ego, per se, but rather a different interpretation of the prestige of his role. For instance, if you see “closer” as the guy who finishes the game and is the #1 relief guy, the hero, the best, then Miller’s role seems less prestigious. However, the way that Miller’s role is presented, especially by Francona to the media, is as the best reliever who is the only one he trusts with the most important and most difficult outs. Miller is the fireman, and he has the most dangerous job because he is the one who can handle it. He gets to be the hero. Not that Miller has excessive pride, but I bet he is proud of his role on the team and doesn’t think he is being undervalued or kept from the limelight or slighted at all by not being the traditional closer.
TL;DR: I would say that Francona has successfully presented Miller’s role as more prestigious and important than the traditional closer role, which keeps Miller and everyone happy.
It doesn’t hurt that he’s on a FA contract and no longer has to rack up saves to make money in arbitration.
And if he’s healthy, in 13 months he’ll get another big FA payday.
He’s just taking care of business like a pro instead of being a diva.
25 months
The Indians control him through 2018.
And to his credit, Miller has been forthright about this since the moment he signed with the Yankees.
December 2015:
For what it’s worth, Miller has insisted his role doesn’t matter since signing with New York. He was open to setting up or closing last year. In fact, after manager Joe Girardi officially declared him the closer, Miller told reporters, “For what they’re paying me, I’ll do anything.”
http://www.cbssports.com/mlb/news/andrew-miller-on-aroldis-chapman-im-not-stupid-hes-a-heck-of-a-pitcher/
I agree that francona is helping with his presentation but miller showed the same no ego behavior when he was on the yankees. Girardi didn’t clearly define miller as the closer for a long time and miller was happy to let him improvise, then after he did clearly make miller the closer (mainly to stabilize betances’ role), when the yankees later got chapman, miller was moved out of the closer role without one word of protest even though it was obvious that he’s more effective than chapman. The guy just has a great attitude.
I agree that francona is helping with his presentation but miller showed the same no ego behavior when he was on the yankees. Girardi didn’t clearly define miller as the closer for a long time and miller was happy to let him improvise, then after he did clearly make miller the closer (mainly to stabilize betances’ role), when the yankees later got chapman, miller was moved out of the closer role without one word of protest even though it was obvious that he’s more effective than chapman. The guy just has a great attitude.
He took what, if he was a capital-C closer, would be a below-market deal with the Yankees, and he even agreed not to grow the mountain man beard he is rocking now. I don’t want to go overboard speculating into the personalities of baseball players, but that suggests above-average humility.
Clubhouse leader for ALCS MVP?
I think clearly, if only based on the media coverage of the series so far. Not saying I disagree with that assessment.
Can Miller pitch in every game like this going forward? I know that the postseason has more off days, but that’s still 30+ pitches two out of every three days.
With a lot of failed starters becoming dominant relievers (i.e. Miller, Britton, Wade Davis), would it make sense for sure-fire playoff teams to try converting erratic but hard-throwing 3rd/4th starters into multi-inning relievers for the postseason a few months before the end of the regular season? (Someone like Samardzija?) Or is this starter-to-reliever transition much more difficult? I guess it’s also rare for a team to have high enough playoff odds early enough in the season to sacrifice the regular-season value
If your thesis is true, which it may be, I would remove the “erratic but” part. I think the winning formula might be a starter with less than stellar secondary pitches; a good fastball, wipeout slider, and not much else would seem to be enough, if there’s reasonable command.
I too would remove the erratic butt part.
I’m afraid to pronounce “erratic butt part”
I would be cautious of taking those examples a little too far. They are extremely convincing, but there are other examples of starter conversions being anywhere from a push to a failure as well.
Miller and Britton, they were barely worth the roster spot as starters, and so that conversion is relatively low risk because if it fails, nothing is lost. The Giants needed to avoid starting Cain or Peavy in a playoff game, so Samardzija is clearly way more valuable as a starter. I mean Shark pitched over 200 innings with an average FIP, for 2.6 WAR. He would have to be guaranteed to be the best reliever in baseball for that to be even worth considering.
A perfect relief pitcher would be infalliball, like the pope. I’m sure there are a lot of mistakes Andrew Miller has made which is why he was a failed #1 pick. Good on him to get his career back on track though, maybe next year he can even try starting again.
Also, how much regret is Lucroy harboring right now? Guy clearly made the wrong choice.
I would say Lucroy has zero regret. He didn’t veto the trade because he didn’t want to play in Cleveland. He vetoed the trade because Cleveland wouldn’t commit to playing him at catcher since they have Yan Gomes signed to a long term deal, and they understandably wouldn’t agree to waive his team option for next season. If Lucroy spends 2017 playing 1B or DH then his value entering free agency after next season will take an enormous hit.
More precisely: cleveland would not commit to him being the #1 catcher next year and did not rule out trading him again.
Still a bad call, results-wise.
At his age he may never sniff the playoffs again.
As is, Cleveland kept a pair of high ceiling prospects that will come in handy.
I stand corrected, you’re right about the part that Cleveland could trade him again.
However hindsight is 20:20 so of course you could say this was a bad call results wise. And I don’t think a 30 year old star player is worried about never getting to the playoffs again. He’s not 35 with perhaps only another year or two of usefulness. I don’t think there’s a 30 year old star player in the league who would have picked Cleveland given his situation.
Probably not. Odds are he would have been traded in the off-season. He still might be traded before the end of 2017. The rangers are no surefire contenders.
I am not a big fan of people mentioning their fantasy baseball teams out of context, but I look forward to running an “Infalli-Ball” roster out next April.
Wait, what? It’s not like Lucroy went to a non-playoff team. What’s the lesson Lucroy should learn from this? Pick the correct playoff team?
Pick the best pitching staff?
Miller is only pitching 2 innings a game. They can win this series without getting another hit off of Miller. But it starts with scoring off the starters, making sure Miller doesn’t come into the game with a lead.
Miller can still pitch when Cleveland is behind. No matter which team is leading, one team gets 27 outs while the other essentially only gets 21. If Toronto takes a 3-2 lead into the 7th inning Cleveland gets 2 innings to make sure they’re at worst still behind by only 1 run heading into the 9th.
I’m not saying he can’t come in for 2 innings when Cleveland is losing, just that if the Jays maintain their lead then it doesn’t matter if he shuts the Jays down.
> Teams have come back from down 3-0 before, of course
Wrong. Team (singular) has come back from down 3-0 before (2004 Boston Red Sox).
And none other than Mariano Rivera was instrumental in that comeback. So if Miller is the new Rivera, and the Jays do go down 3-0…..
And the other team had to lose its first baseman in the middle of the series for that to happen. Maybe Bauer can substitute for Olerud, but he’s a pitcher . . .
Good stuff, one correction: Cleveland has played 5 postseason games.
Also, it’s worth noting the obvious factor in most of the dominant PS relief efforts — extra days off. Miller had 3 days of rest before his first, second and third outings, thanks to the slow PS schedule, plus the rainout and clinching early. His fourth came on no rest (and kudos to him for sustaining the brilliance), but now he’s had another day off.
It will be interesting to see how much and how well he pitches in the upcoming 3 games/3 days. Then again, he’ll get at least 2 days off before the World Series, no matter what.
FWIW, we’ve seen 52 relief games of at least 4 outs so far. Only 8 of those came on zero days’ rest. For outings of 2+ innings, 3 out of 22 on no rest.
I don’t mean to detract from Miller’s performance. But he’s logged 17% of their innings so far. The same workload in a season would be 246 innings — more than any pitcher in the last 2 seasons, and more than his total in the last 4 years.
This isn’t some plea to fix the PS schedule, because that would mean MLB leaving money on the table. It’s just a reminder of how different postseason play is.
Here’s a strange little baseball fact…we all here know that pitcher wins are not that indicative of a stat, even more so for a relief pitcher, but something very odd happened this year. Andrew Miller only walked 9 batters, and won 10 games. In the history of baseball only three pitchers have ever finished a season with 120+ strikeouts and more games won than batters walked, and two of those seasons occurred this year
-Brett Saberhagen in 1994 had 14 wins and 13 walks..
-Clayton Kershaw had 12 wins and 11 walks
-Andrew Miller had 10 wins and 9 walks
At this point the real life fan in me, and tbh fantasy owner, hopes Andrew Miller starts a new trend of the modern 100 inning RP. There are plenty of pitchers with the Andrew Miller story, drafted in the first round, premier SP prospect, and failed to develop into a MLB starter. Usually due to injuries, or inability to not walk batters, or failure to deliver a third pitch. The type of things that hamper a guy his 2nd or 3rd time thru a lineup. There’s no reason though why these young players who can’t develop into starters, automatically need to be delegated to a one inning role. Let them go through the lineup once, get between 6 and 9 outs; pitch the 7th and 8th innings.
Other potential 100 inning relief pitchers : Dylan Bundy, Alex Reyes, Edwin Diaz, Rasiel Iglesias just to name a few..
“hopes Andrew Miller starts a new trend of the modern 100 inning RP.” This.
And I’d settle for 80-90 (a la recent vintage Betances.)
The irony here is that the “new trend” advocated by contemporary SABR types is really just a return to the “old trend” of (notable anti-stat loudmouth) Goose Gossage and others of the early 80s and prior.
Getting those guys to now argue against thoughtful reliever usage because
“geeks” advocate for it is my favorite development since watching all of the guys who hated MoneyBall because it was ‘ignoring defense and baserunning and everything besides walks and homeruns” now ignore defense and baserunning and the like to find reasons to deny Trout an MVP.
And this is as good a place as any to mention that Mark Eichorn’s 1986 season is a thing of beauty to look back it.
What are the odds the Indians trade Miller? Chapman, Jansen and Melancon being on the market might reduce their haul in the offseason, so I guess maybe a trade deadline move works better. Of course, Miller will be a rental at that point, but the Yankees got a good haul for Chapman, perhaps even better than what they got from the Indians.
I think the cat is out of the bag with regard to how valuable a guy like Miller is in the playoffs. If the Indians don’t win it all this year, they would be crazy to trade him. They are going to try again. If they do win it all, they may feel a little better about making some more long-term focused moves, and so if they start next year off poorly they might try to cash him in. If they’re competitive, then trading him would be like the Angels waiting until they are good to trade Trout.
I couldn’t help but think “maybe he has learned something, how bad a starter could he possibly be now?” I look at him and I see Randy Johnson, and think “if it worked for him.” Mainly I’m sad that we will likely never see another Randy Johnson again.
However, watching him, it looks to me like, strangely, the thing holding Miller back is fastball command. When he throws the slider, he looks like he is doing what he was born to do. He looks like Rich Hill looks when he throws his curveball.
When he throws the fastball, it looks like he just wishes he was throwing the slider. It’s very odd, but there is little doubt in my mind that as long as that is the case, he could never start. However, I can’t help but wonder, what if he learns to command the fastball as well as he does the slider? If that happens, then I think he could legitimately be an ace. Rich Hill has basically pitched like one by having unreal command of an unreal curveball and having good enough fastball command to keep hitters from sitting on it.