Bullpenning: Will it be Baseball’s Next Big Trend?

Bullpenning is the idea that a team’s pitching staff would be most effective if there were no starters, and all pitchers threw a max of a few innings per game.

After reading Brian Kenny’s book, Ahead of the Curve, I was left with many thoughts. First and foremost, I have a lot to learn to get to Kenny’s level of sabermetrics knowledge. Secondly though, was my thoughts on bullpenning.

Kenny, along with other sabermetricians, believe that pitchers are most effective in short stretches, when they are able to truly give it their all every batter instead of pacing themselves. As he points out, bullpenning could be the latest way for a team to gain a competitive advantage. Kenny retweeted the tweet below from one of his fans:

The biggest stat to support his claim is that bullpens around the majors continue to have better ERA’s than starters. In the MLB, starters have had a 4.37 ERA this year while relievers have had a 3.95 ERA. A pretty large difference. This trend has been occurring for years now and has shown no signs of ending.

His idea is to have a “starter” throw two-three innings, followed by a barrage of pitchers throwing anywhere from ⅓ on an inning to maybe two or three complete innings. One of the better pitchers would start, since he found MLB teams score the most runs in the first inning. Then, there would be two-three firemen relievers ready to come in whenever the game was on the line.

After hearing Kenny’s thoughts on the idea and then doing my own research on the topic, it seems the idea can truly work. Now in the book, Kenny proposes a full fledged bullpenning attack as the most effective way for a team to deploy its pitchers. But he also mentions an adjusted strategy, where a team uses two good starters and has the rest as “bullpenners.”

If you wanted to learn more about Kenny’s ideas, you should definitely check out his book, it was a great read. The bullpenning chapter is chapter 10. He goes into much further detail about why MLB teams use starters and how it is a trend from the early days of the sport. I will not go into that part but it is interesting to say the least.

My Example

For the purposes of my example, I decided to look at the Baltimore Orioles, a playoff contender who cannot seem to figure out their starting pitching. I have provided a two week example from August and shown how I would distribute the innings between pitchers.

First, the basic 12-man pitching roster:

Chris Tillman, Kevin Gausman, Dylan Bundy, Brad Brach, Tommy Hunter, Yovani Gallardo, Wade Miley(L), Ubaldo Jimenez, Darren O’Day, Zach Britton(L), Mychal Givens, Donnie Hart(L)

In Minors: Vance Worley, Chaz Roe, Oliver Drake

Starters: Tillman and Gausman

Firemen: Britton and O’Day

Gallardo 1

O’Day 1

Britton 1Bundy 2

Hunter 2

Brach 1

Givens 2

Hart ⅔

O’Day 1⅓ Gausman 6

Britton 2

Jimenez 1Brach 2

Miley 2

Hart 1

Givens 1 ⅓

O’Day 1 ⅔

Hunter 1Gallardo 3

Bundy 3

Britton 2

Jimenez 1Tillman 6

Hart 1

O’Day 1

Hunter 1Brach 2

Gallardo 2

Miley 3

Givens 2@SFGOFFVs BOSVs BOSVs HOUVs HOUVs HOUGausman 6

Hunter 1

Britton 2Hunter 2

Hart 1

Miley 2

Jimenez 2

Bundy 2Tillman 6

Britton 1

O’Day 2Gallardo 2

Brach 2

Miley 2

Givens 1

Hunter 1

Britton 1Gausman 6

Hart 1

Hunter 1

O’Day 1Brach 2

Bundy 3 ⅓

Givens 1 ⅓

Britton 1 ⅓

Gallardo 1

Note: I did not take into account injuries but did note the next three guys I would put in the pen if someone were to go down.

Also, the matter in which the relievers go in does NOT matter. Kenny goes into this in detail in his book, but basically it is up to the coaches discretion when to use his best guy. Bases loaded in the fifth and the teams up 4-1: Put in Zach Britton!

I had starters Gausman and Tillman throw an average of six innings a start as an estimate.

Let’s look at the total innings pitched and how that would very roughly project over a 162 game season:

Keep in mind: these projections are only if the Orioles keep the same 12 pitchers on the staff all year. True bullpenning, as stated by Brian Kenny, involves using your minor league team to supply fresh arms throughout the year.

Why It Would Work

You may be asking, why is Zach Britton throwing 130+ innings?!? He’s a closer! Well as Kenny stated in his book, isn’t it strange that your best pitcher either throws 200-220 innings (ex: Kershaw) or 65-70 innings (ex: Mariano Rivera). That margin should be lessened by allowing your best pitchers to pitch as much as possible regardless of their “role.”

Now again, I did a modified version. I had Chris Tillman and Kevin Gausman remain starters because when they start games, the Orioles have a high chance of winning.

The O’s have averaged 4.76 runs per game this year. In 15 of Tillman’s 26 starts, he has allowed two runs or less (58%). In 21 of his 26 starts, he has given up four runs or less (81%). Theoretically, Tillman has put the Orioles in position to win the game 81% of the time, as long as the offense scores their average amount of runs.

Similarly, Gausman has given up two or less runs in 15 of his 26 starts (58%) and four or less in 23 of his 26 starts (88%).

When Gausman or Tillman starts, Baltimore is 32-20. With anyone else starting they are 44-44. Engaging in a full bullpen attack on the days the top two guys aren’t starting allows the coach, in this case Buck Showalter, to play the matchups and situations. If Dylan Bundy is starting and pitching magnificently, sure, let him pitch five or six. But if he gives up a run or two in the first and second, why take the time to let him “find his groove?”

Instead, just move on to the next pitcher. The Orioles do not have much talent in the pen behind Britton, O’Day, and Brach. Allowing these three to each pitch over 100 innings would maximize the O’s chances of winning. This year, Zach Britton is on pace for only 64 innings. Wouldn’t a team want their best guy and possible Cy Young candidate to pitch much more?

Think this wouldn’t work? Well, as Kenny mentions, it already has. From 1972-1980, “closer” Rollie Fingers pitched 1044 ⅔ innings. Or, an average of 116. He started two games through that entire span. But surely he wasn’t as effective, right!? Yep he was, pitching to a 2.76 ERA that looks even better when you get rid of his rough year in 1979.

Moving down the ladder, fringe starters like Miley, Bundy, Gallardo, and Hunter could pitch the middle innings and give it their all in short spurts. Instead of going into a game knowing they are being counted on to go five innings plus, they could focus on getting three-six outs.

Traditional relievers like Givens could still pitch under 100 innings using this model. Struggling pitchers like Ubaldo Jimenez could also be limited to 50-60 innings and specialists like Donnie Hart don’t have to throw 100 innings either.

Next: Rays Starting Knuckleball Revolution?

Right now, the Orioles have had Ubaldo Jimenez (6.19 ERA) and Yovani Gallardo (5.44) throw over 100 innings, with Tyler Wilson (5.38 ERA)  on pace to join them.

All the while, Brad Brach (1.88) has only thrown 69 innings and the previously mentioned Britton (0.65 ERA) has only thrown 55.

The team continues to throw out guys like Jimenez, Gallardo, Miley, Wilson, and Mike Wright to start games. They then proceed to allow them to give up ugly run totals though the first three-four innings, possibly cementing the game result in the early going. Wouldn’t it be better to give Brad Brach the first two innings? Then, maybe give Gallardo the ball for two more strong innings. If he gets in trouble, throw Darren O’Day in. Facing a tough lefty? Throw Donnie Hart (or hopefully a more proven lefty specialist) in the game.

As Brian Kenny said, most smart baseball people, including Bill James and himself, understand that bullpenning is the most effective way to pitch. At the same time, no one is doing it. Kenny believes it is because of the fear of doing something different and failing. While I agree, I think it is mostly that baseball people just don’t like to change.

I can guarantee that a team will try this in the future. Hopefully it is sooner rather than later.