It’s June now and, unlike in my previous version of this article, we have tons of data at our disposal to pour through that has made holes in the market generally smaller and more difficult to spot. Still, there are a few sneaky options that are widely unrostered that I think could provide saves in the second half of the season.
Joe Kelly, CHW
At this point, it looks like Joe Kelly is third in line with Liam Hendriks getting back into the 9th inning and recording his first save. As amazing as the Hendriks story is and as much as I’m rooting for him to succeed, for betting and fantasy purposes, I understand the need to hedge. Hendriks has been hovering a bit under two ticks under last year’s average fastball velocity and has yet to really harness his curveball. His Stuff+, one of the few stats that is even close to stabilizing for him in the amount of work he has so far, has dropped from 133 last year to 115 this year. 115 is still a good number, but Hendriks is doing something unprecedented. It’s probably prudent to wonder about his ability to come back day after day, month after month with similar intensity when his body just went through such a difficult treatment. I’m not trying to say I know something that others don’t, but rather I’m looking at options behind Hendriks because I think we’re in uncharted territory and there’s a lot we don’t know.
Enter Joe Kelly, who is throwing his fastball and sinker about a mile an hour faster, but is now throwing a sharper slider that sits 91/92 and has generated a 23.7% swinging strike rate and 42.9% o-swing, both in the top 10% in the league. He also has a curve that he now mainly throws to lefties, but has served as a primary pitch for him and that Stuff+ likes best among his five listed pitches. The slider is something I want to dive into a bit more as it’s now his most thrown pitch overall and is a key option to both righties and lefties. Last year, Kelly threw his slider at an average of 88.8 mph and generated 13.3 inches of run and a 24.3% swinging strike rate. This year, he’s throwing it more and harder at 91.4 mph, so naturally with less run (5 inches of horizontal break) and a similar swinging strike rate of 23.7 %. How, then, has this faster version of the slider been able to generate better results (.363 wOBA on the slider last year vs .202 this year)? It’s been primarily driven by a large increase in swings generated outside the strike zone.
Kelly’s o-swing has climbed to a career-high 41.5% which ranks sixth among relievers with at least 10 IPs this year. A big reason for this seems to be this new slider and Kelly’s ability to command it around the zone. In this May 5th at-bat against Tyler Stephenson, Kelly front doors him with a slider on the first pitch of the at-bat, then comes back with this on the second.
That looks like it’s just a bad swing from Stephenson, but that slider was 92 with three feet of drop and fifteen inches of glove-side run. After being hit on the inside part of the plate with a similar pitch, he attempts to protect the outside part as well and isn’t able to keep up with the combination of speed and movement. The vertical drop of this slider is even more apparent in this May 30th matchup against Maikel Garcia. This one completely disappears under his bat.
The extra o-swing this slider is generating is the major improvement that Kelly has made over the offseason and is a big reason why he’s now making it one of his primary pitches. As I mentioned, though, he has several pitches that rate as pretty good that he can pull out when the matchup calls on it including the aforementioned curveball and a fastball that he apparently taught himself in a dream. With a lot of uncertainty around Hendriks and Kendall Graveman sitting with a 4.30 FIP right now and having been unable to hold on to closing duties when given a chance in the past, I think Kelly is a cheap, sneaky stash that could pay off in August and September.
Carlos Hernández, KCR
The Royals are one of the teams who should clearly be sellers at the deadline. While it’s impossible to know exactly what they’re going to do, a stash of Carlos Hernández could pay off in more ways than one. Scott Barlow and Aroldis Chapman are both in contract years and performing well enough to fetch some kind of price at the deadline from a team looking to bolster their bullpen depth. If they’re both moved, the best remaining member of the bullpen will be Hernandez, and by a considerable margin. Models love him. His four-seamer is in the 94th percentile of PLV and 8th in Stuff+ tied with Félix Bautista. His overall Stuff+ of 129 is 13th in the league among relievers with at least 10 IPs and his Pitching+ is 15th sandwiched by current closers Evan Phillips and Kenley Jansen. Now that he’s in the bullpen full-time, he’s increased his fastball velocity by more than two ticks and increased his swinging strike rate from 10.7% to 15.5%. The root cause of this breakout begins with the fastball.
This fastball, in addition to being dotted on the outside corner, featured just seven inches of drop at 99 mph. This rise creates a fastball that is easy to pop up or swing under which is why it has an average launch angle of 25 degrees. His slider is also a dangerous weapon with a 44.6% o-swing that ranks in the top 6% of sliders in the league. Everything is pointing to him having dominant stuff as a reliever on the back of his fastball/slider combination, however, it has only produced a 4.45 ERA in 30.1 IPs so far. It isn’t fair to evaluate a reliever based on one or two outings, but it also isn’t really fair to completely dismiss them, so I’ll just note that nine of his fifteen earned runs were allowed in two outings in which he recorded only one out. Perhaps this means that he’s an extremely volatile reliever, but, based on all his other stats, I’m thinking those are anomalies that won’t happen in one out of every twelve appearances like they have so far. Regression to the mean, if Hernandez keeps up this kind of performance, is telling me he’s capable of a 30% K-rate and an ERA in the high-2s for a whole season.
Kansas City’s management is one of the biggest blockers of this potential as it relies on them moving Barlow and Chapman, but also it relies on them not trying to stretch him out to start or use him as an opener anymore. His huge boost in stuff tells me that his best life is in the bullpen and, without Barlow or Chapman in the bullpen anymore, I think Hernandez at least would have a shot to earn some higher leverage work and save situations late in the season. For now, though, he’s a great arm that’s nearly universally unrostered that will give you good innings in whatever role he’s in.
Grant Anderson, TEX
The most speculative name on this list, Grant Anderson has immediately taken on a high-leverage role in the Rangers bullpen after his electric debut on May 30th where he struck out seven in relief earning a win against the Tigers. In his very next appearance, he was sent out in the 8th inning of a 1-0 game against a key division rival and came away with a hold. Anderson has some gaudy strikeout numbers in the minors over the past couple years including a 43.2% clip on 88 batters faced in AAA this year. His velocity, stuff+, and movement numbers aren’t very impressive, so where are all these strikeouts coming from? Well, look at this delivery
That was actually his changeup which he only occasionally throws but actually looks like a really good pitch. That funky delivery, though, combined with excellent extension on all his pitches, can really play up his stuff and velocity. That kind of combination makes for a pretty unique release point and attack angle that could be flying under the radar of the current stuff models. Looking at the numbers of what Anderson has done this year compared to what models say he’s earned, there certainly seems to be some kind of disconnect. We also should take into account how much faith the team is putting in him right out of the gate. Manager Bruce Bochy says he’s “loaded with confidence” and teammate Jon Gray says “I love everything I see”. Talk is cheap, but talk that is backed up by immediate high-leverage usage could mean something.
So, let’s assume that the models are missing something here and that this delivery and his extension give him excellent strikeout stuff. There still needs to be an opportunity. Will Smith is currently the closer in Texas and, while his numbers look really good right now, I don’t see anything he’s doing that’s tangibly different from last year when he finished up with an ERA near four. We have a couple of years of data saying that this version of Smith is a high-3s guy at best and I don’t think that flies for a team like Texas who’s trying to make a deep playoff run. The most likely situation is that they go outside the organization for their playoff closer, but there’s a small chance that they like Anderson enough and that he shows enough over the next month and a half that he ascends to a dominant 9th-inning option. This is definitely an extremely sneaky option given that we have little major league data on Anderson and the team has an established closer already, but the Rangers are a good team who will rack up a lot of wins, so even a small stretch in the 9th could be very valuable.
Bonus Names
Here are a few extra names that I don’t have enough thoughts to write on, but could still fit the bill here.
Nick Anderson – Raisel Iglesias is fine, but Anderson has quietly found his 2019 dominance and would be a great closer if Iglesias can’t work the 9th.
Chris Martin – Kenley Jansen has been OK for now, but early data indicates a higher injury rate for pitchers who were slower in the pre-pitch clock era. Given that Jansen was one of the slowest pitchers in the league last year, having a share of Martin, whose numbers indicate would be solid if he got the job, could end up being a good handcuff to have.
Justin Lawrence – As I was writing this article, Lawrence started to get 9th-inning opportunities, so it’s not that sneaky anymore, but he’s still being used a bit in non-save situations. Colorado is still a tough place to be a closer and Daniel Bard’s season last year is very much the exception to the general trend.
WikiCommons / Flickr Adapted by Kurt Wasemiller (@KUWasemiller on Twitter / @kurt_player02 on Instagram)
GREAT ARTICLE