Yesterday, I got into a bit of a Twitter tiff about the relative values of Jean Segura and Carlos Santana. My position was that they were secretly much closer in value than people realized. And to make that argument, it was necessary to unveil one of my secrets.
In short, the major league replacement level for shortstops is either 1 WAR (Freddy Galvis, Miguel Rojas, and Jordy Mercer) or 0.5 WAR over half a season (Ehire Adrianza, J.T. Riddle, and Adeiny Hechavarria). So basically, scrubby shortstops produce at a one win pace.
The same analysis for second baseman reveals a comparable one win pace as replacement level. This is further reinforced by the release of Cory Spangenberg by the Padres. He produced 0.6 WAR over 329 plate appearances, yet San Diego clearly believed he had neither a place on their roster nor enough trade value to waste time on the phones. Sounds like he got the replacement level treatment.
Surely this has fantasy implications too.
There are a few methods for calculating replacement level in a fantasy setting. For the sake of ease, I’m going to take a massive shortcut and skip the step of creating auction values using any of the usual methods. Instead, I’ll use offensive WAR. This is imperfect for many reasons – most notably because it’s a context neutral stat and our fantasy games are very much context aware. However, I like the idea of looking at this from a neutral perspective. It only fudges the math on the margins. This exercise isn’t intended to name exactly who is fantasy replacement level. We’re here to identify what replacement level looks likes relative to the available talent pool.
In 12 team leagues with no middle infielder, the replacement level is ridiculously high. At shortstop, we’re talking Andrelton Simmons and Willy Adames with guys like Paul DeJong, Jorge Polanco, Jose Peraza, and Carlos Correa missing the cut. Just to highlight a methodology issue, Tim Anderson ranks 26th by offensive WAR despite producing a top 10 fantasy shortstop season. If you want to mentally swap Anderson with Daniel Robertson (ninth), that’s fine by me.
At second base in shallow leagues, replacement level looks like Daniel Descalso, Adam Frazier, and Ozzie Albies. Although that list reads far more like an actual replacement level, we should probably note the flatness of production at the position. Plenty of noteworthy names rate below that trio. The missing piece is DJ LeMahieu. He ranks 35th on the list. Coors Field hurts when context is neutralized. He easily had a top 12 season among second baseman. So too did 20th ranked Jonathan Villar.
If we add a middle infield slot to our 12 team league, replacement level pushes out to players like Marcus Semien, Ketel Marte, Cesar Hernandez, and Yoan Moncada. Replacement level in a hypothetical 20-team with middle infielders would include bad seasons from still-useful players like Jason Kipnis, Ian Kinsler, Dee Gordon, and Dansby Swanson.
So what’s the lesson here? Almost regardless of league depth, there is no position scarcity in the middle infield. This is a point we at FanGraphs have been hammering for well over a year. Now we have evidence of reverse scarcity. Not only is it unnecessary to reach for a middle infielder, you have a modest incentive to prioritize other positions first. When all else is equal, get that big corner infielder.
As for the “historically high” referenced in the title, you’re welcome to try to call me on that. I quickly referenced a few time periods just to sense check, but I didn’t do an exhaustive search for other seasons like 2018. Even 2017 was far less extreme. Although I only expect shortstop and second base to add more good players, that doesn’t mean regression won’t return the replacement level to 0.0 WAR. I know I’m betting on another 1 WAR replacement level, but that’s just my bet.