MASTER NOTES: Planning for the new pitching environment


With pitchers and catchers reporting this week, I have entered Phase Two of my 2018 draft-planning process. Phase 1 started shortly after the end of the regular season, and consisted of me watching the playoffs, and then forgetting about baseball entirely. I know there are many fantasy owners who really dig in during my Phase Two, and it probably helps most of them.

But I think I get a bigger benefit from not doing anything baseball-related. Clears the head, like when I’m doing a particularly tough crossword puzzle—I know that if I just put it aside for a while, the answers will often come popping into my head as soon as I go back to it.

Now I’m looking at my fantasy baseball puzzle. And one thing has popped into my head right away: I need to figure out how to manage a pitching situation that has changed. I play in a league with a 1,000-inning minimum, and I’m not sure how to get those innings, because of the huge change in how major-league teams are using starters.

Where'd All the Innings Go?

My first Roto league, back in the early ’90s, had a 1,000-inning minimum with the standard nine-pitcher roster. I always wanted to just scrape past the minimum, because keeping the innings denominator low meant more second-half flexibility in managing the roster for the pitching decimals. The trick was how to get to the minimum. Just averaging out innings required for pitchers didn’t make sense—nine pitchers to get 1,000 innings meant 111 innings per pitcher, and even then, no pitchers got 111 innings in a season (they still don’t, of course).

So I tried treating pitchers in roster pairs. Instead of nine individual pitchers, I tried to manage four-and-a-half pitcher pairs. 1,000 innings divided by 4.5 pairs was about 220 innings per pair. Unfortunately, this had two drawbacks: So many starters were over 220 innings that second part of the pair needed zero innings. And the .5 of a pitcher, the ninth guy on the staff, had to get 120 innings, and I was right back at square one, since I would either be 60 innings short or sixty innings over, and if the latter, those innings figured to be from an endgame starter, which I suspected would be murder on my decimals.

Then I hit on the idea of pitcher triads. The disadvantage of naming my plan after Hong Kong crime syndicates was more than offset by the workability of the plan: Each three pitchers needed 333 innings. I could triple up two 60-inning LIMA relievers with a 210-inning starter, and I’d make the minimum, probably with great decimals (it was a 4x4, so Ks didn’t matter).

Executing was pretty easy. In 1992, 32 MLB starting pitchers had 210 innings or more (and 15 had 240 innings or more). The sheer availability of long-inning starters presented all kinds of options: You could grab a cheaper high-win guy with so-so decimals like Bruce Hurst (14 W, 3.85/1.26) and offset with two closers and killer LIMA guys? Or you could you pay the premium for Greg Maddux (20, 2.18/1.01) and go with one-dollar relievers.

Fast-forward to 2017, and the situation is entirely different. Just two SPs in the majors last year had 210 innings or more—Chris Sale at 214.1 and Ervin Santana at 211.1. Only another 13 had 200+. So the 2017 pool of 210-inning starters is basically empty.

Now, using the triad strategy means anchoring each triad with a 180-IP starter, which means two 75-inning relievers per triad just to make the innings minimum. Unfortunately, that, too, was easier in 1992. That season saw 47 fulltime relievers get 75+ innings, eight amassing 100+. Last season: 18 relievers with 75+ innings, and only one (Yusmeiro Petit) getting over 90, and he had a start in there somewhere.   

It seems highly unlikely that we will ever see a return to the days when we had dozens of 210-inning starters. If anything, the reverse is true, as more and more teams move to the six-man rotation, the 10-day DL “vacations” pioneered by the Dodgers, and deeper bullpens to keep reducing starter innings.

At the same time, though, rising K/9 rates among relievers are combining with lower innings counts by starters to narrow the K-gap between the two classes of pitchers, such that two decent-K relievers can actually out-fan one more costly starter.

What to do?

When In Doubt, Do the Math

I went and downloaded the Feb. 20 MLB pitcher projections. The news for the triad plan is bad, as I suspected: No pitchers project over 220 IP, with Sale and Corey Kluber topping the list at 218. Just eight more starters are over 200.

After some noodling, I decided to go back to the pitcher-pairs plan, not least because I’m a sucker for alliteration. (“Patrick’s Perfectly Productive Pitcher-Pairs Plan”? Goosebump City.) The difference would be getting four pairs at 240, plus one extra LIMA guy at the end of the staff.

Using a budget split of 180/80 (about 69/31%), I first set up a stars-and-scrubs staff anchored by Kluber and Carlos Carrasco for a combined $54, which is about what I expect them to fetch in my league. It didn’t work—the remaining $26 just wasn’t enough to get eight other pitchers, including some saves.

Next up: the Santana Plan: one ace and then skill pitchers, including closers. This worked pretty well:

SP            /Paired RP    Comb. IP
Corey Kluber  /Hector Rondon     276
J.A. Happ     /Blake Parker      241   
Sean Manaea   /Roberto Osuna     238   
Mike Leake    /Nate Jones        247   
              /Will Harris        58    W   S   ERA/WHIP     K
                               1,060   66  61  3.40/1.14  1040

Based on past years in my league, this line would be worth 42-46 pitching points, which is very competitive. Thanks largely to Kluber’s near-300 strikeouts, even the Ks are in the top half of the league. Wins were mid-pack.

All good, right? Maybe not. The problem with Santana is risk. Losing the ace not only rips Wins and Ks, but endangers the innings minimum. In our league, like most, failing to reach the minimum IP means taking zeroes in both the decimal categories. That’s a lot of fantasy eggs to place in one surgically-repaired basket.

So finally, I tried a Spread The Risk plan, keeping the same relievers but capping individual pitcher spend at $16, while also not dipping down into the $3-$5 range for starters required for the Santana Plan. The results:

SP            /Paired RP    Comb. IP
Marcus Stroman/Hector Rondon     261  
Marco Estrada /Nate Jones        247  
Jose Berrios  /Roberto Osuna     238  
Trevor Bauer  /Blake Parker      256 
              /Will Harris        58    W   S   ERA/WHIP     K   
                               1,060   67  61  3.57/1.18  1028

(Actually, the first run of random selection got the top four in the Blue Jays rotation!) Same innings, slightly worse decimals, slightly fewer Ks. But also reduced risk.


It’s early yet to determine how to approach pitching. Things could change, especially if teams get wise to the idea that they should pitch their better relievers more than their poorer ones. Is there any reason that top relievers can’t pitch, say, four innings a week? If they could, that would be 100+ innings per season instead of 65. Conversely, could teams get even more rigid in allotting their bullpen innings? That might freeze even the top guys in the 60-65 range.

For now, I think I’m leaning toward the STR model, but of course a lot will depend on what happens at the auction table.

Click here to subscribe

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