Starter Durability, Overtaxed Bullpens and the Upcoming Summer of Attrition

By the time you read this, the calendar will have turned over to May, and a full month of baseball will be in the books. Charlie Blackmon is officially a thing, the Milwaukee Brewers have the best record in baseball, and the Arizona Diamondbacks’ season is essentially over – just like we all predicted.

It was a cold April throughout most of the country, and for that reason and others, the recent downward trend in run-scoring has continued. Before long, though, the summer heat will set in, and baseballs will begin flying out of ballparks more often, with pitching staffs bearing the brunt. Some teams will be better positioned to handle this than others. Let’s take a look at starting pitcher durability and relief pitcher usage for all clubs through Tuesday night’s games to get a feel for the clubs who are best and worst prepared for the upcoming summer of attrition.

Below are tables measuring starting pitcher durability and relief pitcher usage in 2013. The key items being measured, from left to right, are: starting pitcher innings per game (AVG SP IP), relief pitchers per game (RP/GM), relief innings per relief game (RP IP/RGM), as well as overall staff ERA for starters and relievers, respectively. The tables are sorted by average starting pitcher innings per game.

2013 AL	GM	SP IP	AVG SP IP	# RP	RP/GM	RP IP	RP IP/RGM	RP ERA	SP ERA
DET	162	1023.00	6.31	428	2.64	439.67	1.03	4.01	3.44
KC	162	986.67	6.09	427	2.64	461.67	1.08	2.55	3.87
BOS	162	984.00	6.07	450	2.78	470.00	1.04	3.70	3.84
CWS	162	982.67	6.07	470	2.90	472.33	1.00	4.00	3.99
OAK	162	976.67	6.03	447	2.76	475.33	1.06	3.22	3.72
NYY	162	969.67	5.99	428	2.64	477.67	1.12	3.66	4.08
TEX	163	971.67	5.96	475	2.91	491.67	1.04	2.91	3.99
LAA	162	964.00	5.95	496	3.06	493.67	1.00	4.12	4.30
SEA	162	960.00	5.93	448	2.77	505.00	1.13	4.58	4.18
TB	163	965.33	5.92	485	2.98	498.67	1.03	3.59	3.81
BAL	162	939.00	5.80	473	2.92	514.00	1.09	3.52	4.57
CLE	162	924.67	5.71	540	3.33	516.67	0.96	3.62	3.92
HOU	162	906.00	5.59	448	2.77	534.00	1.19	4.92	4.72
TOR	162	899.33	5.55	487	3.01	552.67	1.13	3.37	4.81
MIN	162	871.00	5.38	511	3.15	579.33	1.13	3.50	5.26
	2432	14323.68	5.89	7013	2.88	7482.35	1.07	3.69	4.14

2013 NL	GM	SP IP	AVG SP IP	# RP	RP/GM	RP IP	RP IP/RGM	RP ERA	SP ERA
CIN	162	1003.33	6.19	461	2.85	470.33	1.02	3.29	3.43
ATL	162	989.67	6.11	466	2.88	460.67	0.99	2.46	3.51
STL	162	984.33	6.08	483	2.98	475.33	0.98	3.45	3.42
LAD	162	979.00	6.04	504	3.11	471.33	0.94	3.49	3.13
AZ	162	976.33	6.03	527	3.25	518.67	0.98	3.52	4.13
CUB	162	974.00	6.01	489	3.02	474.00	0.97	4.04	3.97
NYM	162	969.67	5.99	535	3.30	507.00	0.95	3.98	3.68
WAS	162	968.33	5.98	440	2.72	477.33	1.08	3.56	3.60
PHL	162	961.67	5.94	466	2.88	474.67	1.02	4.19	4.41
SF	162	946.00	5.84	524	3.23	501.33	0.96	3.30	4.37
MIA	162	944.00	5.83	471	2.91	516.00	1.10	3.42	3.87
SD	162	932.33	5.76	488	3.01	522.67	1.07	3.39	4.31
PIT	162	925.00	5.71	465	2.87	545.67	1.17	2.89	3.50
MIL	162	918.67	5.67	501	3.09	524.00	1.05	3.19	4.20
COL	162	880.33	5.43	503	3.10	555.67	1.10	4.23	4.57
	2430	14352.66	5.91	7323	3.01	7494.67	1.02	3.50	3.86

While having your starters go relatively deep into games doesn’t guarantee a winning season (2013 White Sox are a case in point), it certainly doesn’t hurt. With the exception of the Indians and Rays in the AL and the Pirates in the NL, the 2013 playoff clubs all finished high on their respective league lists. In fact, the other four NL playoff teams were the top four NL finishers, and the other three AL playoff teams finished first, third and fifth. It’s pretty straightforward – the deeper your starter lasts into games, on average, the better your chance of winning.

Beyond that, however, a club’s relative starter durability doesn’t necessarily correspond with the stress applied to that club’s bullpen. One might expect a club whose starters last deeper into games to use fewer relief pitchers over the course of a season. While this is usually the case, there are notable exceptions. In the NL last season, for example, the Pirates’ starters were the 13th most durable in the league, but they used the third fewest relief pitchers in the league. The reason? They didn’t constantly chase the platoon advantage, averaging an NL-high 1.17 innings per relief appearance. Multiple members of their pen were fresh and rested almost every day, and they went on to finish second in the NL in relief ERA. The Astros took a similar approach in the AL, but under much different circumstances, as they were often far behind in games, and logged more long relief innings than most.

On the other hand, we have the 2013 Dodgers, who ranked 4th in the NL in starter innings per game – but also used the 4th most relief pitchers in the league. They averaged an NL-low 0.94 innings per relief game, as they consistently chased the platoon advantage with their two lefty specialists, J.P. Howell and Paco Rodriguez. What did this accomplish for them? Not much, as their overall pen ERA ranked 9th in the NL, in a pitchers’ park. The Mets managed their pen similarly to the Dodgers, and their AL counterpart was the Indians, who led all of baseball in relief outings and ranked last in the league in innings per relief appearance.

Now let’s take a look at the exact same information for 2014, through Tuesday night’s games, and attempt to identify some emerging trends that could cause some repercussions, both positive and negative, as the weather warms.

2014 AL	GM	SP IP	AVG SP IP	# RP	RP/GM	RP IP	RP IP/RGM	RP ERA	SP ERA
KC	25	157.67	6.31	60	2.40	63.67	1.06	3.39	3.20
LAA	26	161.00	6.19	86	3.31	77.33	0.90	4.42	3.58
OAK	27	166.67	6.17	78	2.89	83.00	1.06	2.60	2.97
BOS	27	161.67	5.99	74	2.74	83.00	1.12	3.25	4.23
DET	22	130.00	5.91	69	3.14	67.33	0.98	5.48	3.12
TEX	27	158.00	5.85	77	2.85	83.00	1.08	4.45	3.99
NYY	26	151.33	5.82	75	2.88	79.67	1.06	4.07	4.34
BAL	24	138.67	5.78	70	2.92	76.00	1.09	4.03	4.74
CLE	27	154.33	5.72	90	3.33	79.67	0.89	3.16	4.72
CWS	28	158.67	5.67	79	2.82	92.33	1.17	4.58	4.99
HOU	27	153.00	5.67	72	2.67	89.33	1.24	6.04	4.06
SEA	25	139.00	5.56	81	3.24	81.67	1.01	3.42	3.82
TB	27	149.67	5.54	84	3.11	91.00	1.08	4.25	4.51
MIN	23	126.67	5.51	71	3.09	82.00	1.15	3.40	6.04
TOR	26	140.00	5.38	82	3.15	90.33	1.10	5.08	4.37
	387	2246.35	5.80	1148	2.97	1219.33	1.06	4.13	4.15

2014 NL	GM	SP IP	AVG SP IP	# RP	RP/GM	RP IP	RP IP/RGM	RP ERA	SP ERA
ATL	25	165.67	6.63	70	2.80	63.00	0.90	3.43	1.90
CIN	26	170.67	6.56	59	2.27	56.67	0.96	4.61	2.85
MIL	27	173.67	6.43	86	3.19	83.33	0.97	2.16	2.80
NYM	26	163.00	6.27	81	3.12	81.33	1.00	3.98	3.37
STL	28	172.00	6.14	78	2.79	79.00	1.01	3.87	2.41
CUB	25	152.67	6.11	76	3.04	73.67	0.97	3.79	3.83
PIT	26	157.67	6.06	66	2.54	81.33	1.23	2.66	4.17
PHL	26	155.33	5.97	78	3.00	74.33	0.95	4.84	4.06
SD	28	164.67	5.88	79	2.82	84.33	1.07	1.81	3.99
MIA	26	152.33	5.86	68	2.62	76.00	1.12	3.55	3.37
SF	27	156.67	5.80	83	3.07	87.33	1.05	2.06	3.96
LAD	26	150.33	5.78	99	3.81	92.33	0.93	3.61	2.81
WAS	27	155.00	5.74	80	2.96	89.67	1.12	2.21	3.89
COL	28	157.67	5.63	91	3.25	91.33	1.00	3.74	4.40
AZ	30	160.33	5.34	101	3.37	106.67	1.06	3.63	6.34
	401	2407.68	6.00	1195	2.98	1220.32	1.02	3.28	3.59

The first team of concern is the Detroit Tigers. In 2013, they had by far the most durable group of starting pitchers in baseball, with their 6.31 inning average start length over a full standard deviation higher than the AL average of 5.89. This year, they rank 5th in the AL at 5.91, just above the league average of 5.80. This has something to do with the departure of Doug Fister, but he has yet to pitch a single inning for the Nationals, so shouldn’t be used as an excuse. All of their starters save Rick Porcello averaged well over six innings per start in 2013. While Justin Verlander, Max Scherzer and Porcello have averaged over six this season, recently disabled Anibal Sanchez and fifth starter Drew Smyly have averaged only 4.57 IP per start in their seven outings. Robbie Ray is set to soon make his Tiger debut, and needs to at least be respectable in this department to prevent this from becoming an even more pressing issue.

Compounding their starter durability issue is a fairly major change in bullpen management style from Jim Leyland in 2013 to Brad Ausmus in 2014. Not only do the Tigers need more pen innings this season, they are also generating far more pitching changes to generate them, chasing the platoon advantage more so than in the past. The Tigers tied for the second fewest relief pitchers used in the AL in 2013 (2.64 per game), and rank 5th in the AL in this department in 2014 at 3.14. This increased reliance on Phil Coke and Ian Krol from the left side has not worked out for them so far, as their pen ERA of 5.48 is second worst in the AL. Krol, Joe Nathan, Al Alburquerque and Joba Chamberlain are all on pace to pitch between 81 and 88 games, and haven’t dealt well with the heavy workload as a unit thus far. Starting pitcher durability has been one of the stealth drivers of Detroit’s success in recent years – they need to improve in this area to maintain their winning ways.

Next up are the Seattle Mariners. They currently rank 12th in the AL in starting pitcher innings per game at 5.56 – despite the presence of Felix Hernandez. Take him away, and the average drops to 5.14, by far the lowest in baseball. Take Hernandez and Hisashi Iwakuma away from their 2013 total, and their average drops from 5.93 to 5.48, 28th in the majors. Yes, taking away a team’s ace or aces would be damaging to any club, but it doesn’t detract from the fact that the Mariners have done a very poor job of filling out their starting rotation in both 2013 and 2014. Some combo of Iwakuma, James Paxton and Taijuan Walker should be added to their rotation within the next few weeks, but even a best-case scenario would see them rising only into the middle of the pack in starter durability, especially given Paxton and Walker’s status as relatively untested rookies, neither of whom has a history of averaging six innings per start, even in the minors.

Also concerning is the manner in which the M’s have deployed their bullpen thus far in 2014. The Mariners have used far more relievers per game (3.24) in 2014 compared to last season (2.77), chasing the platoon advantage far more often. Their relievers have averaged 1.01 IP per outing this season compared to 1.13 in 2013. Though their pen performance has been better so far this season, they have three relievers on pace for 78 outings (Danny Farquhar, Yoervis Medina and Charlie Furbush), with Farquhar on pace for 85 innings and rookie Dominic Leone, who didn’t even start the season with the club, on pace for 78.

They lack a true long reliever at present, and actually sent a starter capable of getting them 7 IP in a start — Erasmo Ramirez — down to the minors rather than putting him in that role. Their pen simply cannot hold up with this level of churn and overall usage. There is a path to success here, one that involves seamless returns to health and peak durability levels for Iwakuma, Paxton and Walker. History would suggest that this is unlikely to happen, and attrition could be a very real foe to the Mariners’ staff as the weather warms.

Why don’t the Toronto Blue Jays win? They were a trendy AL East pick in 2013 after “winning” the offseason with major trades that netted them R.A. Dickey, Mark Buehrle and Jose Reyes, among others. A prime reason for their ongoing struggles is their lack of starting pitcher durability. They finished next to last in the AL in 2013 with 5.55 starting pitcher innings per game, and are dead last at 5.38 thus far in 2014. One player who carries a great deal of responsibility for this shortcoming is Brandon Morrow. In 2013, injuries limited him to 10 starts and 54 innings, and he has managed only 22 1/3 innings over five starts thus far this season.

At any given moment, Morrow can be truly dominant, but his mere presence in a starting rotation all but guarantees bullpen overuse. His presence in a rotation beside fellow injury reclamation projects Dustin McGowan and Drew Hutchison is a recipe for disaster – the three have combined for an average of 4.76 IP per start in their 15 outings. As a result, the Jays have two relievers, Steve Delabar and Brett Cecil, who are on track for 81 appearances, and another, Esmil Rogers, on track for 89 relief innings. Unsurprisingly, the Jays’ pen ERA of 5.08 ranks 13th in the AL thus far.

The Cleveland Indians continue to use more relief pitchers than any other AL club (3.33 per game), and use them for the shortest average outing (0.89 IP per relief outing). This is a difficult, if not impossible way to win. The Indians clawed their way into the postseason last year with Cody Allen, Bryan Shaw and Joe Smith logging 70 or more appearances, with Rich Hill (63), Matt Albers (56) and Chris Perez (54) also used liberally.

Except for Allen and Shaw, this entire group has turned over, with John Axford and Scott Atchison added on the right side and Marc Rzepczynski and Josh Outman added on the left. They’ve been solid to date, 2nd in the AL in relief ERA, but Shaw is on pace for 84 games and 82 innings, and Allen (78), Rzepczynski (72) and Outman (72) are also on pace for very heavy appearance workloads. The Indians are beginning to drift away from the AL Central leaders already, and things could get really dicey once the weather warms.

The Dodgers are a very interesting case thus far. Even without Clayton Kershaw, their starters have been exceptional, with a cumulative 2.81 ERA. Despite this, they rank just 12th in the NL in average starter innings at 5.78, way down from their 2013 average of 6.04. Only Dan Haren has averaged six innings per start – despite pitching very well, Zack Greinke and Hyun-Jin Ryu have averaged only 5.69 IP per start between them. This early-hook approach has put a great deal of additional strain on the bullpen, which as noted earlier, logs a ton of appearances even when they are getting many more innings from their starters, as in 2013.

This year they have used 3.81 relievers per game, over two standard deviations more than the NL league average of 2.98. They continue to chase the platoon advantage excessively, averaging 0.93 relief IP per outing, 14th in the NL, after finishing last in this category in 2013. Kenley Jansen is on track for 100 appearances – beyond brutal for a closer – and 87 innings. J.P. Howell (87), Jamey Wright (81) and Chris Perez (81) are also on pace for huge appearance totals. The innings the Dodgers took from some of their better starters and gave to their pen in April could come back to bite them later.

There is also some positive info in the starter durability/bullpen utilization data. In bite-size form, here it is:

– The Angels are up from 5.95 IP/start in 2013 to 6.19 in 2014, 2nd in the AL. Hey there, Tyler Skaggs. The D-Backs could sure use him right now. The Angel pen has been leaky at best, and despite a lack of lefty options, they are second in the AL in relievers used, with the second shortest average relief outing. They aren’t chasing the platoon advantage – they’re spinning the Wheel of Mediocrity.

– The Mets are up from 5.99 IP/start in 2013 to 6.27, 4th in the NL, in 2014 – without Matt Harvey. Dillon Gee has stepped up, and the addition of veteran stabilizer Bartolo Colon along with the return of Jon Niese from injury has allowed them to bring along youngsters Zack Wheeler and Jenrry Mejia relatively slowly.

– Good and bad news for the Nationals. On one hand, they continue to possess one of the game’s more reliable bullpens, featuring a core of Rafael Soriano, Tyler Clippard and Drew Storen. On the other hand, new manager Matt Williams appears to have a much quicker hook with his starters than did predecessor Davey Johnson. They rank 13th in the NL in starter IP per game (5.74), well down from 5.98 in 2013.

While the Washington debut of Doug Fister should help, it doesn’t make a lot of sense that the trio of Gio Gonzalez, Stephen Strasburg and Jordan Zimmerman has been limited to an average of 5.69 IP per start to date.

– There are some positives to be found in the Pirates’ dismal April. They are getting far more innings out of their starters, up from 5.71 per game in 2013 to 6.06 this season. Only Wandy Rodriguez appears to be on shaky ground in their rotation – the Edinson Volquez experiment unsurprisingly appears ready to yield dividends, and Charlie Morton is at least giving them some innings bulk. The Pirates are last in the NL in relief outings, and are again leading the league in average relief outing length (1.23 IP per outing). They have been hurt by sequencing in the season’s first month – their run prevention fundamentals are still strong, and as the weather warms and their offense awakens, they should have a strong, fresh bullpen at their disposal, especially if Jason Grilli returns fairly soon.

– Don’t sleep on the Cincinnati Reds, either. Among their chief strengths is the durability of their starters – they led the NL with an average of 6.19 IP per start in 2013, and are second to the Braves at 6.56 thus far this season. Their pen has not performed well, but it is certainly fresh, ranking last in the majors in appearances and innings. Their performance should positively regress somewhat, and that’s before taking the return of Aroldis Chapman into account. The Reds will be heard from this season.

– Have to mention those Milwaukee Brewers. The biggest changes to this club from their disappointing 2013 campaign are the return of Ryan Braun and the added length received from their starting pitchers. (Of course, as I typed this, Matt Garza left their Wednesday game early due to injury, and they needed to use a position player to close it out.) They ranked 14th in the NL at 5.67 IP per start in 2013, and are 3rd at 6.43 in 2014. This enabled them to handle a spate of extra-inning games without overly taxing their pen, even though they’re carrying a Rule 5 guy, and essentially going with six relievers. Each of their five starters is averaging over six IP per start, with the addition of Garza to replace their 2013 fifth starter revolving door and the development of Wily Peralta the key factors.

– The Braves’ starters’ performance has been off-the-charts great, and top-of-the-charts durable. Their pen hasn’t been nearly as effective as in 2013, but they have handled the second-lightest workload in baseball to date, and should be ready when needed later on. The Royals’ staff has quietly emerged as a lesser, AL version of the Braves’. They are leading the AL in average starter length, and have used the second fewest relievers in the AL, over two standard deviations lower than the AL average. Like the Braves, they have an elite pen that hasn’t been at its best thus far, but it will be ready when needed, giving the Royals a shot to contend despite their significant offensive woes.

A final note – only 14 relief pitchers in all of baseball made 75 or more appearances last season. Of the 14, 10 had an ERA better than the MLB average for relievers. Of these, two accumulated 75 or more relief innings. Of the entire group of 14, only four had made 75 appearances in a previous season, one time apiece. Relievers turn over, fast, and it is very rare and difficult for a reliever to carry such a heavy workload over the course of even one full season, let alone multiple seasons. Mowing through relievers at a rapid pace forces clubs to continuously identify entire new crops of relievers. This clogs 40-man rosters and can create all sorts of unintended consequences. The baseball season is an unforgiving grind, and proper utilization of complementary resources is an underrated trait of winning organizations. Organizational inefficiency in April often comes home to roost later in the season.