Tom Eshelman is quietly having an awesome season

The Orioles have played anemic ball this week, but try telling that to the bullpen. With a 4.20 ERA and a 4.05 FIP, 13^th in the big leagues (note: if that doesn’t blow you away, recall last season’s MLB-worst mark of 5.79) the relief corps is doing their darnedest to carry the team.

The offense has dropped off the table in the last couple of weeks. The rotation has an ERA of 5.37 and the ninth-fewest innings pitched in all of baseball (a stat that looks way better than it is, considering that the Yankees, Cardinals, Braves, and Marlins have all missed a bunch of games). Length and consistency have haunted Orioles starters this season, and that was even before this week, when Wade LeBlanc went down for the season with an elbow injury.

But the bullpen remains a bright spot. Guys like Miguel Castro and Tanner Scott are starting to attract notice for their 2020 bouncebacks. Cole Sulser has done a serviceable job since being thrust into the closer role, and Mychal Givens, effective as a set-up man this season, is drawing attention as a trade chip from a bunch of teams.

One member of the relief crew whose services have gone somewhat unheralded is Tom Eshelman. When the Orioles’ shortened season started on July 24^th, Eshelman was on the taxi squad, somewhere in limbo between the big-league roster and the alternate site in Bowie. You know it’s not a good sign when a team as pitching-starved as this one still picks Kohl Stewart and Cody Carroll over you.

So far in 2020, though, the redheaded righty has been surprisingly solid. Eshelman’s ERA is an excellent 2.75 and he’s been stingy with the walks and hits: at 0.7, his WHIP is considered “ace material,” if you trust Reddit. (You do trust Reddit, right?)

It’s worth remembering that for the 26-year-old Eshelman, this week’s appearance, in a narrow 5-4 win over Boston, was only his sixteenth career big-league game. Prior to this season, his numbers (i.e. his 2019) were not exactly the stuff of legend.

Last year, Brandon Hyde initially tried stretching Eshelman out as a starter, but after four outings in which he gave up a combined 19 runs, he was stashed in the bullpen, where he was not a ton more effective. Opponents hit .315 off Eshelman in the second half of the 2019 season, and his K/BB ratio actually went down as a reliever, from 2.50 to 1.40. No wonder that no one was exactly begging him to join the rotation this year.

But, like the Orioles themselves, Eshelman has been a pleasant surprise this year. He’s provided the Orioles with length at a time when the team desperately needed it: in six appearances, Eshelman has pitched 19.2 innings, an average of 3.2 innings per start. He’s come in as a last-minute starter to replace John Means, he’s entered in the sixth and closed out a game, he’s eaten innings in the seventh and eighth. And with a WAR of 0.6, he’s tied with Chance Sisco, Mychal Givens, and Alex Cobb as the third-most valuable Oriole.

Compared to 2019, Eshelman’s stats look better across the board. His ERA+ is a shiny 167, while his home run rate is way down from 3.0 per nine innings last year to 1.4. The gap between Eshelman’s ERA and his FIP (4.93) is worryingly high, meaning his ERA will climb eventually. On the other hand, it’s also a natural consequence of the fact that he is not, and will not be, a strikeout pitcher. In fact, Eshelman is striking out even fewer people this season. But that’s OK, because he’s also cut his walk rate in half (2.8 to 1.4), and he’s not giving up a ton of hard contact. Opponents are hitting just .182 against him, and their average exit velocity of 88.4 mph is low, as well.

What gives? We know the Orioles have tweaked a ton of pitchers’ mechanics behind the scenes, and guys like John Means, Tanner Scott, and Miguel Castro have credited the coaching staff for improvements. Data from BaseballSavant leads to the conclusion that Eshelman has made changes to (1) his five-pitch mix and (2) his actual stuff. On the mix, Eshelman has played down his not-so-mighty four-seam fastball (which clocks in, a little comically, in the bottom 3% of the league in velocity) and is relying more and more on his sinker and his curveball. As for stuff, Eshelman has also clearly tweaked the slider: he’s averaging five more inches of vertical break, and six of horizontal break on it. The effects are clear: whiff rates for the slider went from 21.5% in 2019 to 30.3% in 2020, and on the sinker, now his primary pitch, from 4.9% in 2019 to 20% in 2020. Eshelman now seems to be using the sinker to draw early swings-and-misses, which helps to set up his slider for the putaway.

Now is an important juncture in the season for Eshelman because, as CC’s Harrison Jozwiak wrote this week, with LeBlanc out for the rest of 2020, a rotation spot has opened up. The two leading candidates seem to be in-house: new pickup Jorge López and Eshelman himself.

While Eshelman seems an improved pitcher in every way, it’s worth noting that he remains more effective out of the bullpen than as a starter. In two games started this season, Eshelman has a 4.00 ERA with 9 hits allowed and an opponent batting average allowed of .273. In approximately the same number of innings in relief, he has a 1.69 ERA with just 3 hits and an opponent average of .091.

Could it be that Eshelman’s slow-moving stuff benefits from being set up by a flamethrower beforehand? Could be. He may also be reliever material by virtue of his success in high-leverage opportunities: with 2 outs and runners in scoring position, for instance, he hasn’t allowed a hit yet this season. On the other hand, he’s faced lineups several times through in several appearances, and batters don’t seem to be evolving and hitting him any better later in the game, attesting to his ability to go multiple innings.

When it comes to how best to use Eshelman to help this team, Brandon Hyde has a tough decision to make. At least, when it comes to whether Eshelman should get the ball, the slow-tossing righty has proven himself deserving of his manager’s trust.