View from the Porch: Bullpen in a China Shop

(Mike DiNovo--USA Today Sports)
The tumultuous start to the 2015 season for Cody Allen has somewhat subsided. His struggles have been a microcosm of what has been a bad month-and-a-half of relief pitching for the Indians. Things are turning around for Allen in the month of May, however. Diving into the numbers for Allen is interesting, because they speak to the volatility of bullpen arms. More often than not, bullpens are inconsistent from year-to-year for any number of reasons – regression, workload, turnover – to name a few.

It’s tough to pinpoint an exact cause for Allen’s slow start to the season. A lack of control and a slight drop in vertical release point may have gone hand-in-hand. The lack of control was certainly an issue. Allen walked seven of the first 42 batters he faced this season. Of course, he also struck out 11 of them, but batters posted a .565 BABIP against the right-hander, which was stunningly high because only 23.8 percent of balls in play were hit on the ground.

What was most frustrating is that Allen was throwing strikes. The problem was that they weren’t very good strikes. Hitters made contact with nearly 90 percent of the pitches in the zone in the month of April and only chased 23 percent of the time. Allen started hitters off with a first-pitch strike two-thirds of the time, but he swing-and-miss rate dropped and the Indians’ awful outfield defense did the rest.

Hitters batted .500 on four-seam fastballs and .308 on curveballs in April. To put that into perspective, the best hitters did against Allen’s fastball in 2014 was .333 in September. They hit .303 in May and .300 in April. In June, they batted .118. In July, .192, and in August, .147. Now all of these are small sample size stats, so they are to be taken with a grain of salt, but the fastball didn’t have a lot of life and the hitters were not swinging at the curveball unless it was elevated.

As mentioned, hitters batted .308 on the curveball. Last season, by month, hitters batted .063 in April, .083, .056, .154, .200, .111 on the deuce. The lack of swings outside the strike zone was largely comprised of curveballs that bounced. Allen’s chase rate outside the zone was just 23.3 percent in April. For the 2014 season, it was up over 36 percent.

Two-pitch pitchers need swings and misses. There’s no third pitch to put doubt in a hitter’s mind and he can often eliminate one of the two pitches based on the count or the situation. That’s exactly what hitters were doing to Allen because he couldn’t spot the curveball, rendering it useless in most counts. That shortcoming, if we want to call it that, is exacerbated by not commanding one, or both, of the pitches. That was the main issue for Allen.

Moving forward, things have already improved in May. He’s relying more on the four-seamer with a 72/28 split and he has had better batted ball luck. The BABIP against has dropped from .565 to .211. Now, once again, everything with a reliever, especially on a month-by-month basis is filled with sample size bias. But, Allen has seen an uptick in ground balls and hitters are chasing 28 percent of the time. They’re also swinging and missing more in the zone. He has only thrown first-pitch strikes to 19 of the 38 batters he has faced in May, so he’s having better results while not working ahead in the count as much as he should.

As observers, we often adopt a recency bias based on what we just saw. It’s hard to say that Allen has improved while keeping in mind his last outing against Chicago, in which he nearly blew the save against the White Sox in Shaun Marcum’s start on Wednesday. It was Allen’s second back-to-back appearance in less than a week and fourth in sixth days. That hasn’t happened much because the Indians have not had many save opportunities. He also had his first four-out save of the season the game before and it was his most dominant outing of the year.

Allen should be the least of your worries with the bullpen. He probably was, but everything is trending in a positive direction. The bigger concerns are Scott Atchison and Bryan Shaw. In my brief podcast with colleagueMichael Hattery, we discussed the future of these two and wondered aloud if either could be a DFA candidate. Atchison could be, though a look at his advanced metrics suggest that he has simply gotten unlucky with fly balls. He gave up four home runs over 280 batters last season and has allowed three home runs to 54 batters this season.

He’s still not issuing walks, but the velocity is down substantially from last season. Atchison worked in a career-high 70 ballgames last year. His velocity is not breaking 90 on average. The plate discipline numbers are about the same in terms of swings and misses and contact in the zone. The chase rate is down and the amount of pitches in the zone is down, but a positive regression in Atchison’s home run rate could at least make him serviceable enough to use.

As for Bryan Shaw, A combination of bad command bad control has been too much to overcome. Ignore the 3.29 ERA because Shaw hasn’t been as good as that antiquated metric would suggest. The right-hander has a 5.54 FIP and a 4.41 xFIP. A 4.41 xFIP doesn’t play in any bullpen. He’s fortunate to have stranded over 84 percent of his baserunners, a number that will regress in a negative way.

The concern here is that Terry Francona will continue to trust Shaw in key spots. The high-leverage setup role should only belong to Zach McAllister right now. Shaw is giving up way too many line drives and a velocity drop may be to blame. He has gone from sitting 92-93 to 90-91 and his cutter is essentially the third-worst cutter among relievers in all of baseball so far this season.

Overuse is probably to blame in this instance. Shaw made 150 appearances over the last two seasons covering 150.1 innings of work. I’m hardly an expert on predicting pitcher injuries, but the velocity drop plus a major drop in control and command are symptoms of a pitcher injury. The worst part for Shaw is that the 49 righties he has faced have slashed .296/.367/.500 off of him. In 2013, righties hit .178/.269/.236 and in 2014 they batted .164/.233/.260. Perhaps this is just small sample size variance wreaking havoc, but Shaw doesn’t look right either.

In Shaw’s defense, average cutter velocity in April last year was 92.45, the same as it was in April of this season per It is down slightly in May. It went up throughout the 2014 season. Shaw may just be a slow starter in terms of building up the arm. He may be just fine once the velocity returns, if it does. For now, I’m very skeptical. He cannot be relied on in important roles for this team right now. It’s a bit of a tough spot for Terry Francona to be in if he doesn’t get seven full from a starter because the bridges to McAllister and Allen are burning. They aren’t in five-alarm fire territory, but they are certainly starting to burn a little bit brighter and more intense.

We’ve seen a hesitance (read: stubbornness) from Francona during his tenure to make changes, especially those that involve veteran players. It’s unlikely that Atchison or Shaw will be removed from the roster for younger talent with upside. We may have to live with this as fans unless one or both of them end up on the disabled list.

If the Indians miss the playoffs this season, it’s possible that we can look back at the stubbornness of Terry Francona for not fixing the bullpen and defensive issues sooner.
Share on Google Plus

About Adam Burke

Under Construction