While this may seem like second-guessing in the wake of the consecutive blown saves by Putz in the Diamondbacks' last games, it's not really the case. It's something first floated in the GDT on Sunday by txzona, and evidence to that end was provided by yours truly, a little later. Certainly, I was very surprised yesterday afternoon, to check the score on Yahoo! and see us going with the usual pairing of Hernandez for the eighth and Putz for the ninth - each man appearing for the fourth day in five. The subsequent failures of both didn't come as much of a surprise. But let's do some more analysis, including a repeat of some of the information posted Sunday.
Overall bullpen usage
Firstly, let's look at the bullpen usage patterns on this road trip. The chart below shows the pitchers used out of the bullpen in each game, along with the number of pitches thrown.
[Takashi Saito and Josh Collmenter were allegedly activated on September 1, but haven't been seen in a game. Collmenter did warm up during the early stages yesterday, when Corbin looked like he wouldn't get through five, but sat down after Chris Johnson's two-run homer got the Diamondbacks on the board.] The uneven pattern of usage is obvious, though things are definitely influenced by all five games being close: four one-run games, and Saturday's in Los Angeles was a margin of two. To some extent, Gibby is damned if he does overuse his best pitchers in close games; and certainly pilloried if he uses the back end relievers.
However, particularly after September 1, with an effectively unlimited number of relievers available, there should be no reason to keep thrashing the same arms until they fall off. Today, after Corbin got through five, I'd probably have sent Collmenter out there for two, and pieced it together thereafter, trying to avoid Hernandez and Putz, and perhaps even Ziegler, unless absolutely necessary. But I know that Kirk Gibson is relentlessly old-school in these things: if there is an eighth-inning lead,.he'll got to his set-up and closer, almost regardless of the circumstances. Putz has 44 of the last 49 saves recorded for us (and one of the other five was a three-inning save by Corbin!).
Putz under pressure
Here's Putz's appearances, batters faced and pitches thrown by month this year:
As you can see, August was his heaviest workload, both in terms of appearances and batters faced. Now, there wasn't an immediately obvious impact in terms of performance - heck, he posted a zero ERA. But he allowed a hit in six of the last seven games, and this month has started badly. It's interesting to compare results with the last time Putz made 13 appearances in a calendar month: May last year, with almost identical numbers of batters faced (48) and pitches thrown (178). As here, that month was perfect, with a zero ERA. But the next one? Putz was awful in June: four blown saves in nine attempts, and a 6.17 ERA, resulting in him being shutdown for almost all July.
Even more damning striking is the time before that in his career Putz reached 13 games in a calendar month, back in May 2009, when the Mets flogged his arm for 16 appearances. Over the first 15, he was excellent, with a 2.65 ERA. But in his next three, Putz faced 15 batters, retired only four, and that was him done for the year, his elbow needing surgery to remove a bone spur. The evidence seems difficult to refute: Putz's appearances need to be managed carefully, in order to keep him both healthy and effective.
Putz and rest
That said, he doesn't immediately appear to have a major problem working back-to-back. games Here are his stats for his career, based on the number of day's rest:
However, not all back-to-back appearances are the same, and if we look at the cases where Putz has done so earlier in the year, he has generally been given longer to recover. Nine times this season prior to this road-trip, he had worked on consecutive days (I'm including the double-header. where he threw in both ends, but that only took a total of 23 pitches). They were: April 6-7 (2 days rest till next appearance); May 17-18 (1), June 18-19 (3), June 23-24 (3), July 6-7 (7), July 15-16 (5), July 22-23 (4), Aug. 16-17 (3), and Aug. 21-22 (3). Discounting the one over the All-Star break, that's an average of 2.6 days rest after going back-to-back.
There was only a single case of Putz being used with one day's rest after back-to-back starts. Even there, he wasn't used back-to-back following the day off, and the initial appearances were less stressful, as he threw only 34 pitches, compared to 45 on Thursday and Friday. This was the first time he had reached that number over consecutive days since coming to the Diamondbacks, and would seem to merit particular rest. For instance, the only times Putz threw 25 or more pitches this season, on May 9 and 22, he was given five and three days off respectively. He threw 26 on Saturday, but was back on the mound after one day's rest.
It's possible to use Putz with a fair degree of frequency, but his arm needs time to rebound, particularly from extended outings on consecutive days. The evidence suggests you also need to monitor carefully his rate of use. A dozen appearances in a month is about the maximum he's good for; if you go beyond that, you appear to be asking for trouble - if not necessarily immediately, then not very far down the road. Should the team bring back Putz for 2013, and talk seems to be that they will (admittedly, before these consecutive blown saves), then they may need to be more cautious with his usage, and that of likely stand-in, David Hernandez.