A starting pitcher has a larger impact on a baseball team than does a reliever. That claim is as much a matter of opinion as the operation of addition. Full-time starters throw 180-220 innings in a single season. Healthy relievers typically log 60-75 innings. 200 is more than 70. That’s why organizations tend to put their best pitchers, at least the ones with a full pitch portfolio, in a starter role.

But what if the standard practice for a reliever was adjusted so he pitched 120 innings? Suppose the team used him for two innings per appearance and in close ballgames, when the team was a run ahead, tied or a run behind. Under those parameters, the value of a hybrid-role reliever could come close to that of a starting pitcher.

But is the mental picture of a hybrid, 120-inning reliever in today’s game purely stuff of historical fantasy?

In 2015, Dellin Betances of the Yankees pitched the most in relief of any major league pitcher; that was 84 innings. In the last ten years, only a handful of relievers have pitched more than 90 innings. Scott Proctor (yes, that Scott Proctor) is the only pitcher to throw more than 100 innings in relief since 2005. Rob Dibble threw 98 innings (4.3 WAR) for the Reds in 1990. Scott Sullivan (113 IP) and Danny Graves (111 IP) were the most-used relievers in 1999 and the latest pitchers to throw more than 110 innings in a season.

It was not always so. In 1976, 17 pitchers threw more than 100 innings in relief. The Reds’ Pedro Borbon (117.2 IP) and Rawly Eastwick (107.2) were among them. Rollie Fingers pitched 134.2 innings (4.1 WAR) out of the bullpen. In 1975, Rich Gossage threw 141.2 innings (4.0 WAR).

Back then starters also tossed more innings. In 1976, Vida Blue had 298 IP (7.6 WAR) and in 1975, Catfish Hunter (328 IP) and Jim Palmer (322 IP) threw the most. Tom Seaver led the league in 1975 with 7.8 WAR in 280 innings.

Starters and relievers throw fewer innings than they did a few decades ago. Practices have changed.

Why bring this up?

On July 29, I wrote about the resurgence of the Reds bullpen. The main explanation was the return of Raisel Iglesias (June 21) and Michael Lorenzen (June 24). During the previous month, the Reds had used both in an unconventional way, with spaced-out, two-inning appearances. More math: If the Reds were to use Iglesias and Lorenzen for an entire season the way they had through the end of July, Iglesias would pitch 117 innings and Lorenzen 98. It recalled a warm, hazy vision of Scott Sullivan and presented a template for compromise between a starter’s full load and traditional reliever.

A little more than a month later, my ears perked up at the Q&A session with the Reds front office staff when Nick Krall floated the notion that a reliever like Raisel Iglesais, if he pitched 120 innings, might be nearly as valuable as a starting pitcher. It’s one thing for an uncredentialed writer (me) to spitball a new idea. It’s another level of seriousness when a major league assistant general manager does it.

Let’s take a closer look, starting with reaffirming the transitive property of inequality.

Starting pitchers and relievers have been worth different amounts. The top 10 starters in 2015 earned an average of 6.1 WAR (using FanGraphs WAR). For example, Clayton Kershaw pitched 232 innings and earned 8.5 WAR. Anthony DeSclafani, who was the #30 ranked starting pitcher, threw 184 innings and earned 3.1 WAR.

In contrast, the top ten relievers in 2015 earned an average of 2.2 WAR. Cody Allen of Cleveland was tops, with 69.1 innings and 2.6 WAR. The #30 reliever was A.J. Ramos of the Marlins, who threw 70.5 innings with 1.2 WAR. Mariano Rivera was a reliever for 18 years. He pitched 80 innings twice. Rivera, a Hall of Fame lock, averaged 2.2 WAR/year. Another example closer to home is Aroldis Chapman. He has been used an average of 60 innings each season, generating 2.6 WAR/year.

Behold, math. Starters are worth more than relievers because (a) clubs tend to use their best pitchers in that role, and (b) other things equal, starters throw about three times the number of innings as relievers. But in theory, a reliever who was assigned 120 innings instead of 60 could close that value gap about half the way.

The Reds used Raisel Iglesias and Michael Lorenzen in unconventional ways, at least for certain parts of the season.

In 16 of his 25 appearances, Iglesias pitched for at least two innings. Twice he went three. Not only have the Reds not used Iglesias on back-to-back days, he’s enjoyed at least two days between 22 of his 25 trips to the mound. However, the club has cut back his innings of late. Bryan Price has used Iglesias in 42 innings since his return on June 21. That extrapolates to 90.2 IP/season, down from the projected 117 IP/season when we calculated it at the end of July. In August and September, Iglesias has pitched at the standard modern bullpen rate of 71 IP/season.

The Reds have given Michael Lorenzen 41 innings over 29 appearances since he came back on June 24. That works out to 93.2 IP/season. Through August 9, the club used Lorenzen in similar fashion to early Iglesias. In eight of Lorenzen’s first 16 trips to the mound, he threw at least two innings. But that pattern has changed. Price has assigned Lorenzen more than one inning in only two of his last 13 appearances. In September, for the first time this season, Lorenzen has twice been used in back-to-back games. Over the last month, Michael Lorenzen has been used like a regular high-leverage reliever.

What caused the Reds to slow down their use of the two pitchers in August and September? Maybe concern for the health of their arms. Or maybe the early usage practices had to do more with medical protocols than paradigm busting.

What conclusions can we draw about the now-fictional 120-inning reliever?

The hybrid use of a relief pitcher as conceived here would be a sharp break from modern practices. It has been a quarter-century since any reliever has thrown 120 innings in a single season. The Reds seem an unlikely organization to challenge this (or any) conventional wisdom. But maybe that’s selling the new front office short.

We know they aren’t experimenting with it in 2016. The hybrid would average 20 innings per month. The Reds used Raisel Iglesias 18 innings in July, but only 12.1 in August. He’s on pace for 12 in September/October. Michael Lorenzen hasn’t been used even 15 innings in any month. To reiterate, the Reds are no longer doing anything provocative with their high-end relief pitchers.

That said, forty years ago a 120-inning reliever wasn’t a freakish rarity. Major league teams play an average of 27 games per month. Pitching 20 innings in 2-inning stints would mean appearing in 10 out of 27 games. That doesn’t seem crazy impossible.

Maybe the hybrid strategy works if it’s only one pitcher, not two, at a time. There are only so many appropriate situations. Lorenzen and Iglesias may be taking innings from each other. One solution would be expanding the parameters for when the hybrid pitcher is used. Include games when the team is two runs ahead or behind.

At 120 innings (or even 90), relievers will still be worth less than full-time starters because of math. But confining their innings to high-leverage situations would bring the values closer together. A certain percentage of the starter’s innings are with leads or deficits of three runs or greater.

If Raisel Iglesias’ balky shoulder or Michael Lorenzen’s pitch portfolio mean one or both are destined for the bullpen, then any number of innings beyond the typical modern reliever role would be beneficial to the club. 90 innings from above-average pitchers have more value than 65. While you’re thinking about it, throw this on the mounting pile of arguments against the restrictive, Dusty Baker-style closer role for the best arm in the bullpen.

Finally, if the Reds front office staff genuinely wants to explore a hybrid reliever strategy, they’ll need to make sure the manager agrees. He’s the one who determines the use of pitchers in games, not the general manager. Flexibility on this issue is yet another reason to make sure the next manager is wide open to new ideas. Dusty Baker isn’t the only manager to graduate from the Old School. If the next guy believes the best relievers only play when the team is in the lead, and then only for one inning, then the Reds will never come close to maximizing the use of their bullpen.

15 Responses

  1. cfd3000

    Price’s use of Iglesias and Lorenzen (at times) and Nick Krall’s acknowledgement that the front office is at least evaluating this type of approach does give me hope for this bullpen. When Iggy or Lorenzen pitches two innings they don’t (or shouldn’t) see any batter twice. And Lorenzen seems to be growing more pitch efficient with time. If this approach allows the Reds to have a shut down bullpen (though they need a Cingrani upgrade), to go with what should be a dominant starting staff, this usage makes too much sense not to explore. I like Iglesias, Lorenzen, Wood, Cingrani (mostly), one more lefty (Peralta or Lamb?), Josh Smith or an equivalent, and a couple of long men / low leverage guys. Combine that with a starting staff selected from a strong list of locks and contenders and the Reds could have good to very good pitching next year and beyond. But if the old school philosophy kicks in things could stay mired around just average. If the starter goes five innings and four or five relievers are needed to finish the game, that’s not a winning formula. I’m not so optimistic as to expect 120 innings from Iglesias and Lorenzen. But if they both stay in the bullpen next year I am hoping they both top 90, mostly high leverage, innings. It could happen. Right? Go Reds.

  2. David

    I would place a modest wager that Cingrani is traded in the off season.

    I think there is still hope for Lamb as a starter, as he was not at full strength this season after back surgery last winter, but he could make a good reliever, if groomed for it.

    Peralta was a starter until this year, and now looks pretty impressive as a reliever.

    And I still think the Reds would like Iglesias to be a starter.

  3. Nick Carrington

    Great article, Steve. It seems likely that barring trades, at least three of the following will be starters long term:


    I think the Reds need to give each of them a starting chance and take the best ones. Why can’t two or three of them pitch major innings in the bullpen? Having guys like this in the bullpen doesn’t mean they will never start either.

    They could theoretically have most of these guys on the staff and still transition guys pitching 2-3 innings to bullpen roles.For instance, if Cody Reed started next season pitching 2-3 innings out of the bullpen, the Reds could quickly and easily stretch him to make starts if/when an injury occurs.

    If Stephenson and Finnegan don’t improve their command/control, this role might be perfect for them. It also may be a good opportunity for Reed to learn to get big league hitters out and build confidence.

    Iglesias and Lorenzen can obviously do it or it appears they can. I think they are both best used as starters, but I understand the shoulder situation with Iglesias. As far as Lorenzen’s pitch profile, I found this interesting in research for something else: Lorenzen and DeSclafani throw almost the exact same pitches at extremely similar rates, at least in 2016.

    Four-seam fastball: 30.8%
    Two-seam fastball: 23.6% (Lorenzen calls this a cutter)
    Slider: 38%
    Curve ball: 6.4%

    Fastball: 31%
    Two-seam fastball: 26.2%
    Slider: 30.8%
    Knuckle Curve ball: 9.9%

    Disco throws a different, curve ball, but they aren’t that dissimilar. I found this fascinating because I hadn’t thought of them as similar pitchers, but in this sense, they are.

    The biggest question for Lorenzen does seem related to his pitch profile as Steve notes. Can the curve ball be effective enough to be a viable off speed pitch? In limited usage this season, the curve ball has been effective, but the sample isn’t big enough to say much of anything.

    Anyway, the Reds should consider these 100-120 inning roles for a number of guys in they don’t make the rotation.

    Sorry for the long post. I got excited.

    • old-school

      Reds Memo
      From: Dick Williams
      To: Bryan Price
      Re: Pitching Plans 2017 and offseason planning
      CC: Tim Kremcheck

      Iglesias: bullpen
      Straily: SP
      Homer SP plan A- Meet with Tim monthly
      Lorenzen tentative SP Plan A-Meet with Tim monthly
      Bob Steve SP Plan B
      Garrett SP Plan B
      Cody Reed SP Louisville

  4. WVRedlegs

    Mr. Grossman and Mr. Krall are no dummies. They came across as very intelligent men. I still can’t get over what a great Q&A meeting that was. And the fact that Mr. Krall is over player development makes this even more intriguing and noteworthy. There is no visible evidence yet though that this is an organizational philosophy change. We have only seen this type of usage from Iglesias and Lorenzen. Even if Iglesias’s shoulder can withstand being a starter, with as many starters that the Reds will have to turn into relievers, the possibility is there that they can convert and use a selected few former starters in this manner. And if they can hit like Lorenzen, they wouldn’t necessarily have to come out of a game for a pinch hitter, and thus go an extra inning if need be.
    I’m still holding out for Iglesias to be a starter. With that, maybe John Lamb would be a good candidate for this, too.

    • ohiojimw

      I was a little surprised the other night that Lorenzen did not pinch hit for the pitcher in the bottom of the inning before he entered the game then I remembered that Iribarren isn’t quite the lost cause most of the Reds PH guys this year have been 🙂

  5. Scott Carter

    Great article! I have been pondering pretty much the same thing. You have several guys who could pitch 2+ innings every third day say. Then if you had a few of the starters, say Desclafani and Bailey, that could give you 7 innings plus on their starts. You could have a pretty good staff. I like the thinking.

  6. ohiojimw

    The tendency seems to be to look beyond Lorenzen’s elbow issue; and, I believe that is an error. I’ve noticed that lately he is wearing a compression tube on his throwing arm which extends from several inches below the elbow to well above it. This would certainly indicate his elbow area issues are not a thing of the past but rather something they are trying to work through.

    A gimpy elbow and high slider rate would not seem to be comfortable bedfellows in a starting pitcher. I think this puts him in the bullpen where he will throw fewer sliders over all simply as a function of pitch count even if he does not change his mix.

    In regard to this post, the question becomes was he experiencing a flare up of elbow issues when he was working two inning stints? Is that a reason why he is working one inning stints now? And what is the prognosis for his elbow after several months rest?

  7. lost11found

    Its possible that they are kicking the tires on the idea with both pitcher seeing which one does the best in the role and who’s arm handles the shift in workload and preparation the best.

  8. The_Next_Janish

    I like this idea. Go with 3 two inning pitchers as a bullpen rotation. Every third day you pitch two innings. If the starter goes seven, well you’ll be closing that day. We count innings but throwing 20 pitches to warm up to may or may not pitch an inning of 18 pitches should be taken into account. Some of these pitchers probably are throwing 100 innings unfortunately 40 innings of its wasted in poor bullpen management.

  9. cupofcoffee1955

    The Reds on the cutting edge…, has Uncle Walt been woke up & told about this?

  10. Tom Mitsoff

    The L.A. Dodgers used this approach with relief pitcher Mike Marshall in the 1970s. In 1974, he appeared in 106 games, pitched 208 innings, FINISHED 83 games, and had a 15-12 record, 21 saves and a 2.42 ERA. This was just a year after Montreal had used him in a very similar manner. After this two-year stretch, he got hurt and was no better than average until he recovered for a couple of good years for the Twins to finalize his career.

    At the time, people marveled at how often he was used and how he maintained his effectiveness. Part of the explanation was that he was a professor of kinesiology (the study of the mechanics of body movements), and therefore had very unique insight into how much and in what ways he could push his body. His out pitch was a screwball, which might have had something to do with his arm giving out.

    At that time, Mike Marshall was THE relief pitcher in the bigs, even though Rollie Fingers was pitching for Oakland then. Very few teams, if any, have used relievers that way since then, and it may be because that usage resulted in an injury to a player who at the time was considered the best at his position in the game.

    I personally think 120 innings each for Lorenzen and Iglesias in relief roles would be an outstanding idea. You wouldn’t need as much depth in the pen for crucial situations. I would really rather see Iglesias start, but his arm is telling us that does not agree with him.

  11. redsfan06

    Here are a couple of examples of relievers throwing 90+ innings a year from the Reds past:
    Clay Caroll – pitched over 90 innings every year from 1968 – 1975 and occasionally started (15 times) during those years
    Pedro Borbon – pitched over 120 innings every year from 1972 – 1977

    If Lorenzen and Iglesias can contribute at the level those two did, the Reds bullpen would be solid for years to come.

    • Old-school

      So guys from the BRM to Scott Sullivan to multipurpose high workload relievers from a different era are referenced….yet modern Renaissance laid back Dusty throws Chapman with limited innings in very narrow roles in only 60 innings and old school is used to define him….I’m confused.

  12. kmartin

    Great article Steve! I could not agree more with your post. You can also add Bruce Sutter to your list. He pitched 100 or more innings five times.