Roberto Osuna and the Aging Curve for Young Relievers

Way back in April, the Blue Jays turned some heads when they filled out their bullpen with a couple of 20-year-old A-Ballers: Miguel Castro and Roberto Osuna. Few doubted that these young arms had closer-type stuff, but they also lacked any experience against big league hitting. There wasn’t much of precedent for pitchers making that type of jump, making it darn near impossible to know what to expect.

The two arms went in polar opposite directions. Castro had a brief run as the Jays’ closer, but was sent back to the minors in May after a rough start. Toronto later flipped him to Colorado in the Troy Tulowitzki deal, and he remained in the minors until September.

Osuna, on the other hand, pitched brilliantly from the get-go. He took hold of the closer’s job in June after a strong start, and he never looked back. He finished the year with a 63 ERA- and 73 FIP-, both of which marks ranked in the top 35 among qualified relievers. He struck out 28% of opposing batters while only walking 6%.

The season I just described would be impressive for any reliever. But Osuna’s campaign is especially notable given his age: 20 years old. Twenty-year-old big leaguers are a rarity in modern baseball. Some of the very best prospects don’t debut until they’re 22 or 23. Kris Bryant and Noah Syndergaard are a couple of super-recent examples. Osuna was the youngest player to appear in the majors this year, and is currently the only player born in 1995 (Gosh, I feel old) to appear in a big league game.

It’s widely accepted that players in their early twenties get better as they age. This isn’t necessarily the case for pitchers, at least not on the aggregate, but at the very least, one wouldn’t expect a player to get much worse as he progresses through his mid-20s. This implies that Osuna should remain a quality reliever for the foreseeable future.

However, pitchers have an annoying habit of getting hurt and/or losing their effectiveness with little notice. Relief pitchers can be even more fickle. Good relievers often don’t remain that way very long, while dozens of scrap heap pickups turn into serviceable bullpen arms each year. Osuna might be a ticking time bomb.

These are two conflicting narratives, and their juxtaposition makes it difficult to grasp at what the future may hold for the 20-year-old. To shed some light on Osuna’s unique situation, let’s turn to some historical examples. These should tell us what’s become of Roberto Osuna’s of years past.

First, I took every reliever season since 2002 in which a pitcher 22 or younger faced at least 200 batters as a reliever. Then, I performed a weighted Mahalanobis distance analysis using the metrics used in the SIERA formula that I normalized to league average. I used SIERA’s coefficients to appropriately weigh the variables in my distance calculations. In simpler terms, I compared Osuna’s rookie season to other young pitcher’s seasons by weighing each metric according to its importance in preventing runs. For pitchers with multiple seasons in my sample, I kept only the earliest one.

Due to the paucity of early-20s relievers, a mere 31 players made it through my filters. Since 31 isn’t all that many, I included the full list below, ranked from most to least similar to Osuna’s season. I divided the weighted Mahalanobis distances by the lowest distance to give you a sense of how quickly the dissimilarity escalates as you move beyond the first few. The table below is sortable, if you’re into that sort of thing.

Roberto Osuna’s Mahalanobis Comps
Rank Season Name Age ERA- FIP- Adj. Mah Dist Career WAR to Date
0 2015 Roberto Osuna 20 63 73 0.0 1.3
1 2010 Neftali Feliz 22 61 68 1.0 4.7
2 2006 Matt Capps 22 86 95 7.8 3.9
3 2015 Keone Kela 22 56 63 8.3 1.5
4 2005 Huston Street 21 41 64 10.3 11.0
5 2012 Kelvin Herrera 22 57 66 12.4 3.9
6 2013 Carter Capps 22 143 122 12.8 1.5
7 2007 Boone Logan 22 107 104 14.1 1.7
8 2013 Paco Rodriguez 22 65 82 14.6 1.3
9 2007 Eric O’Flaherty 22 102 82 20.4 2.9
10 2010 Drew Storen 22 89 81 20.7 4.6
11 2010 Alex Burnett 22 127 115 21.1 -0.1
12 2004 Ryan Wagner 21 107 112 21.4 0.2
13 2006 Jonathan Broxton 22 59 69 22.0 10.9
14 2003 Oscar Villarreal 21 55 82 22.3 0.5
15 2011 Chris Sale 22 66 74 22.8 22.8
16 2005 Marcos Carvajal 20 109 103 23.2 0.0
17 2014 Carlos Martinez 22 105 80 26.2 5.0
18 2006 Brandon McCarthy 22 101 111 29.0 13.2
19 2003 Francisco Rodriguez 21 68 89 30.3 16.6
20 2015 Trevor Gott 22 79 96 31.5 0.1
21 2011 Tim Collins 21 87 111 35.6 0.8
22 2006 Joel Zumaya 21 43 74 39.7 2.6
23 2004 Chad Cordero 22 66 86 40.6 2.6
24 2007 Brandon Morrow 22 94 96 46.2 10.8
25 2005 Andy Sisco 22 69 88 46.4 0.1
26 2003 Jung Bong 22 118 115 46.4 -0.2
27 2009 Ryan Perry 22 84 102 48.6 0.2
28 2005 Ambiorix Burgos 21 89 90 51.2 -0.6
29 2014 Dominic Leone 22 59 83 62.0 0.3
30 2003 Wil Ledezma 22 118 109 76.4 1.8
31 2002 Tim Spooneybarger 22 64 104 98.4 0.7

Here’s a look at the career trajectories for some of the pitchers whose seasons were most similar to Osuna’s, based on cumulative WAR. Note that I excluded Keone Kela, whose season took place in 2015.

I purposefully erred on the side of including a lot of relievers in this chart. Although some of their seasons weren’t all that similar to Osuna’s, I think casting a wide net helps to show just how few of these guys remain excellent for very long. Huston Street and Jonathan Broxton pitched well through most of their 20s, but they’re the outliers. Most of these relievers fizzled out relatively young, and never even crossed the five-win mark, including Matt Capps and Neftali Feliz — Osuna’s top two comps.

This jibes with the research done by Bill Petti and Jeff Zimmerman on aging curves, which found that relievers age much more rapidly than starters. Their skills begin to atrophy almost immediately upon their arrival in the big leagues, and the decline shows up across virtually every aspect of pitching. Strikeouts, walks, velocity, you name it.


Osuna’s a very good reliever. Without him, it’s very possible the Blue Jays wouldn’t have made it this deep into October. And his 2015 exploits are made even more impressive by the fact that he’s a mere 20 years old. However, history shows that time is rarely kind to relievers — including good, young relievers like Osuna. As a result, we very well might be seeing the best of him right now. Luckily for the Blue Jays, right now’s version is pretty good.

Chris works in economic development by day, but spends most of his nights thinking about baseball. He writes for Pinstripe Pundits, FanGraphs and The Hardball Times. He's also on the twitter machine: @_chris_mitchell None of the views expressed in his articles reflect those of his daytime employer.

newest oldest most voted

Spooneybarger or Forkeylighter?