Is there any way to bolster the bullpen in August?
Since July 1st, the Twins' bullpen has tossed exactly 81 frames, or about 29.9% of Minnesota's innings. In that time, it's been a crew that's been ineffective on the whole: 5.33 ERA, 1.41 WHIP. Because we're not dealing with a great number of innings, these last few days haven't helped anyone's case, but that doesn't make concerns about relief any less valid. In the final two months of the season it's a necessity that the bullpen is not only fresh but useful; particularly when there are still 24 days until rosters expand, which is a long time. Especially in the month of August when the Twins have just one homestand.
So, with 14 away games before September, I wanted to examine the 'pen. Not just to see who's been used, but when they've been used. A primary part of this study is something I've called the leverage appearance. A leverage appearance is any time during (or after) the seventh inning where a relief pitcher came in when the Twins were A) up by three or less, or B) down by two or less. This includes situations where the pitcher has either put himself into that scenario (giving up a 3-run homer with a four-run lead, for example), and also situations where the team may have expanded a lead in between a relier's innings (Crain gets one out in the seventh, Twins then extend lead from one run to five, Crain comes back out to pitch the eighth). We'll also use another completely made up term, which I'll call the Leverage Index, which is just the ratio of leverage appearances to total appearances.
While we'll exaine some statistics here, performance and results are equally as important as how the guys were utilized. This table is cumulative from July 1 - August 6.
Joe Nathan: 16.3% of bullpen innings, 16.7 % of bullpen appearances, 23.9% of leverage appearances, 78.6% leverage index, 2.6 days between appearances
Nathan's numbers the last six weeks have been stellar, but he's also been used appropriately; at least in relation to how the rest of the bullpen is being utlized. He's tied for the highest percentage of leverage appearances, meaning Gardenhire is turning to him when times get rough. Naturally, as the closer, Nathan also has a high leverage index, meaning he rarely gets used when the game isn't tight.
Matt Guerrier: 18.3 % of bullpen innings, 19.0 % of bullpen appearances, 23.9% of leverage appearances, 68.8 % leverage index, 2.3 days between appearances
While he's had the most work of any bullpen arm since July 1, it's still been less than three innings per week. So what, then, is the problem? It certainly appears that it's his performance, and his performance alone, that's been the cause of his recent struggles. Because in spite of getting nearly 1/5th of the relief innings he's not been over-worked. In the last week, perhaps, but not prior. He has the second highest leverage index on the team, which could mean he's not been dealing well with the stress of the situation, but it's far morely likely that A) it's just been a rough week for Guerrier, or B) opposing hitters have found a weakness in either his pitching patters or delivery. At any rate, no matter how disappointing his stints on the mound have been the last few days, focusing on the mistakes of one guy from this past week sees like a waste of time considering there are arms who have struggled all season.
Jesse Crain: 14.7% of innings pitched, 16.7% of bullpen appearances, 15.2% of leverage appearances, 50% leverage index, 2.6 days between appearances
It seems somewhat of a shame that Crain's appearances and innings, as well as what kind of situations he's been used in, mirror those of Brian Bass. Crain since July 1 hasn't been the most reliable either, but at least his numbers insist he's pitching better than his results. Earlier in the week I suggested that the innings given to Crain and Guerrier had made them over-worked, and by consequence, made two or three other pitchers under-worked. I was wrong. If Guerrier hasn't been run out too many times, Crain certainly hasn't, either. The difference is that Crain's role has been somewhat undefined, in spite of Gardenhire's penchant for having a defined bullpen heirarchy. In the last six weeks, at least, it's seemed that while Guerrier was the "go-to" fire fighter, Crain was Mr. Versatile, coming in to record the final out of an inning (and then sitting down for the night) just as often as he was asked to pitch the whole of the seventh inning. But if there's an issue with his use, it isn't how often he was used, it was when. If Crain and Bass are both available, Bass should probably be the one to stay on the bench.
Dennys Reyes: 7.9% of bullpen innings, 14.3% of bullpen appearances, 17.4% of leverage appearances, 67% leverage index, 3.1 days between appearances
While part of me argues that it's best to limit Reyes' exposure in order to get the most bang for your buck, part of me also argues that by using him as sparingly as they do that they're simply wasting that buck anyway. Of course, since he's around, it makes sense to use him to the benefit of the team, and that's definitely what's been done. Seven of Reyes' 12 appearances have been to record one out (or less), but in spite of that he has been put into leverage situations. He's the LOOGY to Crain's ROOGY. Or at least, that's how I see the rationale. As for results, it's hard to argue with how Dennys has performed this season. He's effective in his role, and in spite of what some might say, when players understand their role it's easier for them to understand what's expected, and therefore be more effective in their job.
Boof Bonser: 15.5% of bullpen innings, 9.5% of bullpen appearances, 4.3% of leverage appearances, 25% leverage index, 4.6 days between appearances
Clearly, Boof is getting mop-up duty and long relief. He isn't trusted in situations that could decide the outcome of a game, and his use reflects that strategy. Boof's been striking people out in his relief role, which is great, but he's still allowing way too many hits.
Brian Bass: 15.1% of bullpen innings, 14.3% of bullpen appearances, 15.2% of leverage appearances, 58.3% leverage index, 3.1 days between appearances
Gardenhire's use of Bass is one of the more confounding issues confronting the use of the bullpen. While his fastball routinely touches 95 or 96, that's far from untouchable for most major league hitters. It seems he may be overmatched for the role that's been assigned to him.
Craig Breslow: 8.7% of bullpen innings, 9.5% of bullpen appearances, 4.3% of leverage appearances, 25% leverage index, 4.6 days between appearances
The final arm to examine is one that's raising some questions. Overall Breslow has been more effective than both Bass and Bonser, yet he's used so rarely that when he takes the mound you're like "Oh yeah, that guy". If the Twins are looking to improve the performance of their relief corps, this would be a great place to start.
We'll go over specific options for improvements over the weekend. This is already my longer than I thought it would be.
Since July 1st, the Twins are 18-13 (.581), which is still pretty good. But considering the pen's overall performance, I have to believe that even an average core could have prevented a pair of the breakdowns we've been witness to over the last six-plus weeks. When we see how close this division is right now, and when we understand that this club hasn't been the best when it's been on the road, each game matters.
If the organization doesn't want to work a waiver-wire transaction, and if they refuse to bring up help from Rochester before September, there are a couple of suggestions I have to improve bullpen reliability.
1. This is something that Gardenhire has already vowed to do, and that's to use Joe Nathan in more situations. If a game is close in the eighth inning, Nathan's your guy. As discussed by PhoenixV, he usuallly does pretty well in one-plus inning scenarios.
2. Define the role of Jesse Crain in order to alleviate the pressure on Matt Guerrier. He's one of few solid options in relief, so keeping them both as set-up men is a good option, but they both need to be ready to go when a game is tight in the sixth or seventh innings as well. Using Crain as a one-out guy is a waste of resources.
3. Promote Craig Breslow to the fourth rung in the bullpen heirarchy. With a LOOGY on board, a pair of set-up men and a pair of mop-up men, Breslow's sucesses this season have earned him the right to own middle relief. So far this season he's been effective while rusty from disuse; give him a shot to show what he can do when he's given a defined role. Two point five innings a week doesn't sound like much, but if he can be as effective in that role as he's been in the one he currently holds, this move on its own could solve a lot of problems.
4. Consider the consolidation of the roles of Dennys Reyes, Boof Bonser and Brian Bass. With 38 appearances and 32 innings between them over the last six weeks, there can only be so much disuse and redundancy.
These moves would mean more innings by both Nathan and Breslow, therefore relieving pressure from arms who haven't been as effective. Guerrier and Crain would continue in their current roles, with a slight shift in priorities that would allow Crain to shoulder part of Guerrier's load in leverage situations. Shifting the garbage innings that Bass had been receiving to Bonser, allowing Boof more appearances and more innings, because he's equipped to handle that load. Finally, instead of saving Reyes for late-inning appearances versus left-handed hitters, try to use him earlier. If it's the sixth inning and there are a pair of left-handers coming up, use Reyes to get a cheap two outs. Not only does this increase how any outs he could get (if only from one to two, but every out counts), but lessens the liklihood that the opposing manager will send out a pinch hitter. There's less pressure to come through before the seventh and eighth innings, meaning the match-up may not be quite as crucial to the opposing team.
Beyond those suggestions...well...I'll rant on those this weekend. Leave your own crackpot ideas below, and enjoy your Friday, everyone!