clock menu more-arrow no yes mobile

Filed under:

Can the Arizona Diamondbacks bullpen recover?

Are recent failures bad luck, regression or attrition ? Is it all three ?

Miami Marlins v Arizona Diamondbacks Photo by Jennifer Stewart/Getty Images

I wasn’t going to write an article, as I’m tired of writing articles about bullpen and then always end up looking stupid. But I still thought this was interesting enough to share:

So first you have some basic bullpen stats broken out by time periods…(Yes the dreaded selected begin and end points). However I think they are useful here in aligning the narrative of the bullpen collapse with performance data:


Season IP NL ERA Rk ERA FIP x FIP K/9 BB/9 HR/9
Season IP NL ERA Rk ERA FIP x FIP K/9 BB/9 HR/9
Thru ASB 338 1st 2.85 3.97 4.09 7.8 3.2 1.0
7/20 - 8/29 95 8th 4.07 4.29 3.95 9.1 4.2 1.1
8/30-9/9 32 13th 5.97 4.36 3.41 9.4 3.7 1.4
YTD 464 1st 3.31 4.07 4.02 8.1 3.4 1.0

So this is pretty straight forward. We can see that in the first half the team had the best Bullpen ERA in the NL (In fact all of MLB), but the FIP/xFIP were over a run higher. These fielding independent metrics indicated that perhaps the DBacks bullpen wasn’t pitching quite as well as their ERA indicated. Nonetheless, watching the games, we saw them coming through and shutting teams down most of the time.

However from the All Star Break up until just before the Dodger series, (7/20-8/29) we saw them have a few bumps in the road which had the effect of pulling their ERA even with their peripherals for that smaller time period. The K’s were up, but so were walks and homers allowed, and not by small amounts. Part of that was due to personnel changes of course. But while less “reliable” or lockdown, the pen was still not yet a glaring weak spot.

Then the Dodgers series, and everything since happened, and the numbers have blown up. The ERA is almost 6, and they’ve blown game after game. Interestingly however, note that the xFIP is actually lower in that sample size than the previous ones, and overall ERA has been much higher than the basic peripherals. I’ll come back to the xFIP in a little bit.

Bullpen Batting Against

Thru ASB 338 .220 .295 .357 .653 .259 1.6 40.6% 81.3%
7/20 - 8/29 95 .233 .324 .391 .715 .286 1.9 42.8% 72.9%
8/30-9/9 32 .298 .364 .508 .872 .366 2.6 49.0% 68.9%
YTD 464 .229 .306 .376 .682 .272 1.7 41.7% 78.3%

Obviously the batting against numbers, up slightly in the 6 weeks post ASB, jumped in the last ten games. However note the GB/FB rates. Those increased a lot in the 6 weeks after the ASB, and then really jumped the last 10 days. We usually associate high GB rates with more success. And in this case, the increase in GB rate is why the xFIP is lower. Fewer fly balls should translate into fewer homeruns. xFIP regresses the Homerun rate on fly balls to neutral or average. So thats why xFIP is going down. But in fact, HR per FB has jumped:


Time HR/FB
Time HR/FB
Thru ASB 11.80%
7/20 - 8/29 16.20%
8/30-9/9 23.80%
YTD 13.20%

So here you see quantified what our eyes have been assaulted with on a nightly basis recently. They are actually giving up far fewer fly balls, and inducing a ton more ground balls. But the Hard Hit rates have been up, and a lot of those ground balls have been hit sharply - but also many are just finding holes. Now look at the BABIP rates in the second table (.366 in most recent small sample). BUT it’s the HR/FB rate that is really killing them. Despite giving up far fewer fly balls, almost 1 in every 4 balls hit in the air is leaving the yard lately.

Of course this is all on a team-wide basis, and it does us no good trying to understand this better without looking at the individual performances. Here is a larger table showing relievers since the all star break. (Minimum 10 IP)

I’m not going to say they’ve been unlucky, but it should be noted that it is both a relatively short period of time, (i.e. Small Sample Size) which is magnified by the importance of the games now, and they might have been just a little unlucky lately.

Clearly Archie Bradley and Brad Boxberger need a break from the 8th and 9th inning. (And we need a break from watching them pitch those innings!). Just don’t overlook the workload that Yoshihisa Hirano has been asked to shoulder. He has the most relief IP on the team since the ASB.

I think the team should use Yoshi as the closer, and play the match ups in bridging to Yoshi. Many think the team should go with Ziegler as closer, but if they need him 2 days in a row and 3 out of 4, then he probably has issues. His workload must be regulated due to age and wear and tear. He’s already pitched in A LOT of game, in fact the most in the NL.

As has been pointed out numerous times, the D Backs have 4 relievers in the top 5 in games pitched. I raised concerns about too much concentration of the workload in too few hands as early as May 1st in this fan post. So this all may just be the workload coming home to roost.

It doesn’t get any easier either, as the team is in Colorado tonight for the start of a 4 game series. Not exactly the place that a weary bullpen goes to get healthy and back on track. Feeling discouraged and pessimistic about their prospects this week is fully justified. I am encouraged slightly, however, by the fact that in the big picture, this is still a relatively small sample size of the season where they’ve been this bad, AND the fact that it seems almost certain Torey will make some changes to the 8th and 9th innings now... Right? Right??