MASTER NOTES: The End of the (Pitching) World As We Know It?

You might not have heard, but major league baseball recently passed its first in-season trading deadline. By the time the bell rang at 4 PM on Tuesday, more than 40 trades had been completed and more than 100 players and prospects were on new teams.

Most of the analysis of who did what will center, as it probably should, on the immediate effects—various takes on winners and losers, expert opinions, team grades, our own rankings of the prospects dealt, and even predictions about players who weren’t traded who could still be traded after the deadline.

Because of this clear shortage of commentary about the trading deadline, Master Notes will step into the breach. But what is interesting to me about this year’s trading action is that this year's deadline trading foretells a potentially major shift in pitcher usage.

I did a quick check of the players traded, and 19 of them were relief pitchers, at least six of whom were established closers going to teams where they wouldn’t be closing games. That number could be a little higher if you give Brad Brach credit for being the closer in BAL after they traded Zach Britton, and Jake Diekman credit for being the closer after TEX dealt Keone Kela. The number goes higher still if you count Roberto Osuna and Ken Giles as closers, although they weren’t officially closing games. And I don’t know how to count Darren O’Day, who would have been next in line to close in BAL had he not already been declared out of the year after surgery (oddly on his leg, not his arm).

The point is that 19 relievers is a lot—about half of the major-league players who got dealt this trading season were bullpenners. And almost without exception, they went to teams in the playoff hunt:

  • ARI got LHP Jake Diekman, RHP Brad Ziegler and RHP Matt Andriese
  • ATL got LHP Jonny Venters and RHP Brad Brach
  • CLE got LHP Brad Hand and RHP Adam Cimber
  • HOU got RHP Roberto Osuna and RHP Ryan Pressly
  • LA got RHP John Axford
  • MIL got RHP Joakim Soria
  • NYY got LHP Zach Britton
  • OAK got RHP Jeurys Familia
  • PIT got RHP Keone Kela
  • SEA got RHP Sam Tuivailala, RHP Zach Duke and RHP Adam Warren

Three teams added both a lefty reliever and a righty, and a fourth, the Yankees, added LHer Britton to a bullpen already loaded with RH arms like Dellin Betances, David Robertson and Chad Green, in addition to lefty closer Aroldis Chapman.

The contenders have been mobilizing all of these bullpen reinforcements because deep and powerful bullpens are now seen as the safest path through the playoffs to a championship.

Experts argue about the evolution of the bullpen-centric model. Quite a few put it in 2014, when the Royals came within a whisker of a World Series by relying on their bullpen arms. KC starters averaged barely over 5 innings per start in those playoffs, got 4 wins and posted a 3.91 ERA/1.30 WHIP with a paltry Dom of 5.4 K/9. The relievers got 6 wins, and posted a 2.54/1.18 aggregate line with a Dom of 10.3 K/9. The Royals model had the last three innings locked down by Greg Holland, Wade Davis and Kelvin Herrera, whose aggregate line was 1.01/0.84, 11.6 K/9.

In recent years, managers and teams have shown themselves increasingly willing to experiment with pitching-staff optimization that veers away from the traditional model of the starter going six, the relievers an inning apiece, and the closer always and only in “save situations.” COL tried a “tandem starter” approach a few years ago, looking at third-time-through falloff and trying to prevent otherwise good starters from facing a third time through. That didn’t work because the starters weren’t good enough the first two times through. This year, TAM made headlines with their “openers” or “bullpen days,” when they’d start a reliever with the full intention of putting a new guy every couple of innings. TOR also tried that approach late in the year.

And looking at the trades, it seems highly likely that for the stretch run and playoffs, starters, especially starters at the ends of rotations, will be on very short leashes. Teams rely on deep bullpens to carry the load. We will probably see several teams go with pitching staffs of three starters and nine or 10 relievers, including fourth and fifth starters who will become “openers” or long men. Any game that gets to the fifth or maybe even the fourth will be delegated to a parade of lights-out relievers.

It doesn’t take a whole lot of imagination to see this model continuing to evolve, until it becomes the norm use throughout the regular season as well as the playoffs. It just makes baseball—and financial—sense.

The average MLB game requires pitchers to face about 38 hitters. The average 2018 MLB starter faces 23 batters per game (BF/G) and gets 16-17 outs (5.1-5.2 innings).

That leaves the relievers to get 10-11 outs. But here’s the thing: They get their outs while allowing a lot fewer baserunners. The average 2018 fulltime starter has allowed about .41 baserunners for every out (BR/Out). The average reliever (min 20 GP, no starts) allows .33 BR/Out. Substitute the reliever rate into the SP rate for their 16-17 outs, and the team would allow about 1.33 fewer baserunners per game by using short-stint pitchers. That’s 215 fewer BR per season. And the advantage is even greater if the short-stint pitchers replaced only starters with .40 BR/Out or more (39 SPs, mostly stars like Sale, Verlander, Nola, etc. this season are under .40 BR/Out).

So instead of a 13-pitcher staff of five or six starters, some of them quite awful, plus seven or eight relievers, a staff might have an ace-level starter or two, and 11 or 12 relievers. You might argue that only about 80 relievers are under .40 BR/Out in 2018, which is far short of the 300 or so that would be needed.

To which I say, “Hah! You’re forgetting that many of the subpar starters could be effective relievers if they had shorter stints! You’re forgetting that there will be relievers in the minors who could be effective but can’t get to big-league rosters because the pitching staffs are clogged with awful starters! And, most importantly, you’re forgetting that the existing solid relievers could easily be stretched out from their current workload of three outs, twice a week, to maybe eight to 10 outs, two or three times a week!” Or even more. It’s even likely that teams would schedule all their pitchers for all their games.

It can be done. It would take some work for teams to optimize throwing schedules, warmup routines and other aspects of preparation and training. There wouldn’t be much worry about “buy-in,” considering how many pitchers, currently barred from the big leagues because they are one-pitch guys, would see a path to big-league paychecks opening up before them.

But it will happen. The money talks. Teams with lower revenues (and greedy owners in general) would see opportunities to cut their pitching costs hugely, while improving outcomes and reducing the risk inherent in paying ace starters. Max Scherzer makes $30 million a year to get 20 outs per game. That $30 million could pay an entire staff of eight-out guys, especially if the owners agreed to expand rosters to 27 players (the union couldn’t object to all those extra jobs, right?).

I know sim players who are already using their pitching staffs along these lines, and they are finding that it works. Of course sims are not reality, but the Rays and Jays have been working with the idea, with mixed results. Other teams are working to figure out better ways to manage reliever workloads, to get their pitchers into games rather than wasting pitches warming up.

Add it all up, and it looks like a decent bet that we might not recognize pitching patterns and staff usage in a few years. The traditionalists won't like it, but I can’t wait.


Click here to subscribe

  For more information about the terms used in this article, see our Glossary Primer.