MASTER NOTES: Changing pitcher categories

Master Notes has been wondering about how player values might change if we changed the fantasy categories. Two weeks ago, we looked at hitters, changing HRs to Total Bases and BA to OBP. The result was that players like Bryce Harper, Freddie Freeman, and Trea Turner moved well up the value charts, while JD Martinez, Ozzie Albies and Starling Marte fell down.

This week (a little later than first promised, I know), we’ll look at pitchers, again changing some categories to see how player values change with them. We’ll make three different category changes, then use the Custom Draft Guide (CDG) to revalue all players using the different category combinations. The settings were a 70/30 hit/pitch split, balanced valuation model, no scarcity.

What I was interested in knowing was whether the value distributions between starters or relievers change. As well, I scanned through the new value lists to see which individual pitchers gained or lost significant value.

Then I went and sat on my deck, tuned in the radio broadcast of one of my pitchers’ starts, and enjoyed a cold Ball’s Falls Session IPA from Bench Breweries of Beamsville, Ontario (no, they didn't pay for the plug). There’s more to life than Master Notes.

* * *

The first task was to set a baseline for the standard 5x5 categories. Through Tuesday’s games, the top 135 pitchers (15 teams x 9 pitchers) were 56% starters, 44% relievers. The pitcher value split was a little more heavily tilted towards starters: 62%/38%. The average starter value was $9, the average reliever just under $7. Starters dominated the top of the value list, with 15 of the top-25 pitcher seasons to date, including all of the top-10.

The question was how those values and value distributions might change if the categories were changed:

  • Using Quality Starts (QS) instead of Wins. Everybody loves to hate wins because they’re fickle, luck-based, unpredictable, unprojectable, and frustrating. And those are their good points. I would actually prefer to have used the Ryan Quality Start (7 IP, 3 or fewer ER), but while better commish services can accommodate RQS, the CDG doesn’t include it among the pitcher category options. Yet.
  • Using “Net Relief” (NetR) instead of saves. I have to admit that when I first saw “Net Relief,” I thought of the tremendous throughput speeds my cable company promised when I signed up for their high-speed internet. Then I thought of the antacid I chew on when I’m watching my fantasy team’s nightly stats on my high-speed Internet. But actually, NetR is saves-plus-holds-minus-blown saves. This seems to be an improvement on saves alone, which are right up there with wins as aggravating scoring events. Still, NetR is not perfect, because the Hold is not perfect as a reflection of a reliever's positive contribution: A reliever can get a hold only if he enters the game in a save situation. If a reliever enters a tie game, or a game with his team down, he cannot get a hold (or a save), even if he pitches perfectly. That means he can literally “hold” the game in its current state for his team, and still not get any scoring reward. (And then if he induces a grounder that goes right through the third baseman’s legs, it will be scored a “hit.” Baseball scoring stinks.)
  • And I also checked what the value effects would be if I changed both of the above categories.

The short answer: The differences weren’t that dramatic. The top starters—Verlander, Scherzer, Cole, Sale, Kluber, deGrom—stayed atop the list, although in all the changed environments, their value rose a buck or so apiece. Lower-valued pitchers gained more significant value here and there, and lesser pitchers dropped some value, occasionally enough to fall out of the top 135 and therefore to no longer be rosterable. The one good thing was that some valuable setup guys added value, and a few poor starters lost value.

QS for Wins

Overall, the value change in swapping QS for wins was pretty negligible. SP value and reliever value stayed at that 62%/38% split. I had thought that SP value would rise and RP value would fall, because of all the vulture wins lost by bullpenners and QS gained by the SPs.

While that rise in SP value overall didn't happen, there was a change was in the value of individual SPs. Again, not too dramatic, but the new values seemed to better reflect good pitching rather than good luck, which is a good thing. For instance, Garrett Richards dropped out of the top 135 with one fewer QS than win, and Chris Archer got in with 6 QS but only 3 wins. As noted, there wasn’t a lot of such movement at the very top or very bottom of the SP-value list, but four pitchers made very big jumps into the top 35:

  • Kyle Gibson, barely rosterable at the #128 slot in regular scoring, jumped 48 spots and added $2.91 in value when QS were used—Gibson has just one win but 6 QS.
  • Reynaldo Lopez topped all starters with a $3.47 value gain, jumping up 44 overall places and into the top 40 of starters. He has only one win despite 7 QS (and 2.93/1.17 decimals).
  • Trevor Cahill, with 5 QS but only one win, also gained almost $3 in value, and climbed into the Top-40 among starters and top-70 pitchers overall.
  • And with 8 QS but only 6 wins (QS count more because there are fewer of them), Tanner Roark picked up more than $3 and jumped 27 overall slots into the Top-30 among all pitchers and the top-20 among SP.

Big fallers included J.A. Happ, Eduardo Rodriguez and David Price.

The relievers who gained value are those with significant saves totals, as we’d expect. Which brings us to the next change.

Net Relief for Saves

This was not a surprise: Giving more relievers paths to value resulted in more relievers offering value and reaching the top-135 rosterability threshold. Although I’m not actually sure “rosterability” is a word. The starter-reliever number split rose a couple of points in relievers’ favor, although the value split actually moved slightly towards starters. The two big gainers were hybrid RPs—lefty fireballer Josh Hader became the top RP on the board, adding more than a dollar because his 6 saves have been enhanced by 8 holds. Similarly, Adam Ottavino saw a $4+ jump thanks to his 12 holds. Jeremy Jeffress and Joe Kelly also climbed, both into double-digit value.

The big value drops were all by closers, led by Edwin Diaz and Wade Davis, who lost more than $6 each when their regular values, driven by their league-leading saves totals, suddenly had reduced impact. The idea of including holds in the category was to add value to non-closer relievers, who lose out in the saves-only scoring. From that angle, it seemed to work, as the overall category values were more compressed.


Finally, I ran the CDG with both the QS change and the NetR change. There were big gains for setup guys like Archie Bradley, Matt Barnes, Ottavino and Chris Devenski, largely offsetting the loss of value from having to forego vulture wins with the change to QS. The biggest drops included saves-only guys with pedestrian (or worse) decimals like Shane Greene, Fernando Rodney, and Alexander Colome, as well as starters like Brandon McCarthy, Brent Suter and Chris Stratton.


Overall, the pitcher value proposition still weighed significantly towards starters. Feeding some value into the Haders and Devenskis of the game seemed like a positive outcome, given how valuable they are to their real-world teams. The major gain was de-emphasizing wins and saves, which was kinda the point at the outset.

Once again, I don’t believe for a minute that any of this generate excitement for getting rid of wins and de-emphasizing saves from our scoring methods. I still think they should be replaced, to get a broader and fairer way to assess actual pitchers. And it was more than a few years ago when we started hearing fantasy writers saying OBP should replace BA, a change that has taken hold is many leagues. We can hope.

Click here to subscribe

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