Andrew Miller, Fire-Fanner


Thursday night, with one out, with the eventual tying run on third base and the potential winning run on first, Mike Shildt called upon Andrew Miller. Over his career, Miller has thrived as the "fireman," most famously in the Cleveland Indians almost-World-Series run. And this time, like so many other times before, Miller entered and retired the only two batters he faced.

Inning over.


Inning over – except now the game was tied. You see, Miller induced a groundout to the first batter he faced. With the runner on third, the Cardinals needed Miller’s first out to be a strikeout. Miller got a groundout, then a flyout, and the game was tied.


Frankly, that outcome shouldn’t have come as a surprise. Jason Heyward, the groundout “victim,” is a tough strikeout. He’s struck out in only 19% of his plate appearances this year and puts nearly half the balls he hits on the ground. Heyward was the perfect hitter in the situation for the Cubs. That didn’t used to matter against Andrew Miller, because he used to strike everyone out – used to:





Yikes. Miller’s strikeout rate over his last 20 games sits at 19.4% - the lowest of any such stretch since he converted from starter to reliever. He had a blip where he sorta-maybe looked like his old self this year, but other than that - yuck. And you know what makes a strikeout rate that low even worse, especially for a fireman reliever? Walks.





Yeah, that’s a big WOOF. Over those last 20 games, Miller’s walking one out of every six hitters he faces. Compounded by his struggles to get outs lately, that means he’s walking almost a batter per inning (~7 per 9IP). His K-BB ratio in that stretch sits at a lowly 3.0% - Miller had never run a K-BB ratio lower than 8% in his career as a reliever and hadn’t dropped below 13% since 2013. He's not putting out mid-inning flames - he's starting the fire.


This isn’t really a new development for Miller, either. While it’s never been to this extreme, the warning signs have been everywhere. His 2018 ERA of 4.24 was more than 2 runs worse than his bullpen ERA since 2012 and almost a run higher than his worst season as a full-time reliever. His FIP, while still clocking in at a respectable 3.51, was more than a run and a half worse than the prior year.


The building blocks of his recent performance might be even more troubling. Fastball velocity? Down from 95.4mph in 2016 to 92.7mph this year with successive drops every season. Slider velocity? 84.6 to 82.6. Batters used to chase 40% of his pitches outside the zone and only swing at 50% of the ones inside it - not exactly a recipe for hitter success. This season, those rates are at 30% and 60%, respectively. He’s allowing 1.79 HR/9 which would be the highest of his entire career (and would be higher if not for this play by Fowler). The last time he ended a season above even one homer per nine was 2011.


In other words, Miller doesn’t look like has the stuff anymore to be an effective fireman. And yet, that’s exactly how he’s been used out of the Cardinals bullpen:





Back to K-BB% again, but this time with gmLI. gmLI is a measure of leverage index at the time a pitcher enters the game; higher number, higher pressure. For the Cardinals, eleven pitchers have tossed at least five innings of relief since the break (plus Mike Mayers which, well, you know). Eight of them have a higher K-BB% than Miller. None of them have entered into higher pressure situations, on average, than Miller. Being in the top-left of the chart is bad. Miller is the top-left of the chart. He’s not getting the results, he doesn’t have the stuff, but he’s still facing the most important batters.


Frankly, the way Shildt uses Miller at this point in the season is unlikely to change. After all, the organization paid him to take these innings and face it – if you were in Shildt’s shoes, would you rather answer questions after a game you lose with an unheralded bullpen arm from the organizational depths, or one you lost with Miller? You go with Miller and lose – shit happens. You go with Junior Fernandez and you’ll be second-guessed.


Thus far in the second half, everyone in the Cardinals bullpen has a better FIP than Andrew Miller. Most of them have a better ERA. It probably won’t change how the Cardinals use him, and maybe it shouldn’t – the team needs Andrew Miller to be at least Andrew-Miller-ish if they want to make a deep postseason run. We fans better hope Miller figures it out.


I don’t know if I’d put this in the “encouraging” bucket, but when Miller gets his fastball up above 94mph and his slider above 84mph, his pitches get whiffs and suppress wOBA just like they did during his peak. The obvious problem is that he hasn’t hit 94 with a heater since July 19th or 84 with a slider since August 7th.


(Of course, as I wrote this, Miller pitched against the Cubs in the 8th inning of another key one run game. On his second pitch of the outing, he eclipsed 94 with a fastball to Schwarber in the midst of a clean inning.)


Maybe this stretch run will give him an extra shot of adrenaline.