A better guide to surplus value

Anyone who knows me has to deal with the fact that I think surplus value (SV) is a pretty clever concept, combining player performance (WAR) with a player’s salary to determine how good a value that player is to a team.

Yeah, I can see it getting a little unseemly, like the nonsense this summer about the Chicago White Sox gaming Eloy Jiménez’s service time to squeeze another season of control out of him? Maybe. I’l admit, SV is a little bit ledgery.

But you know what, teams are prized these days (probably always have been) for finding productive talent on the cheap. And that’s what SV measures: How good or bad a team’s gambles were, whether by draft, trade, or free agency.

One catch with the system is that the dollar value per WAR, from FanGraphs, is mired perhaps not in the “real world” of play, but skewed to free agency and prime players.

As was pointed out in the comments at some point this season, the concept that 1.0 WAR carries a value of $8 million may not be ludicrous in and of itself. But mixed into the batter is the idea, apparently, that a fair amount of any team’s WAR is going to come for “free” — i.e. as cheap, minimum-salary labor. Literally, a 25-man roster of 1.0 WAR players would cost a team $200 million — while having a 73-89 record.

I don’t like the looks of that.

So, I’ve made some changes to the SV system going forward.

First, I’m going back to my original method of averaging all three WAR measurements: Baseball-Reference (bWAR), FanGraphs (fWAR) and Baseball Prospectus (WARP). Yeah, I know only WARP incorporates catcher framing, while bWAR is generally the more generous WAR measurement of the three. As you’ll see in the end-of-year SV articles to come, WAR can wildly vary from measure to measure, and not just for catchers. So while SB Nation allies itself most closely with FanGraphs, and we tend to rely most on FanGraphs in our own analysis, a combination of all three WARs sits best with me.

Second, and more significantly, I’m no longer tying WAR value to FanGraphs’ estimate, which this year for the White Sox lands exactly at $8 million per WAR. It’s not a “realistic” measure, and while you could pick any number out of a hat and the truths of how the White Sox compare to the rest of baseball would still be valid, I’d like something that feels reasonable when rolled out on the spreadsheet.

How did I accomplish a more realistic value? Dumbly; I’m not mathematician.

I presumed that each “replacement” player on our 48-win baseline replacement team was making the minimum salary of $545,000 (that isn’t realistic I suppose, there would be some lousy veterans on a replacement team making more that the minimum, but this is the best I can do, bub). That’s a replacement team salary of $13,625,000.

Average team wins are 81, replacement wins are 48, so wins above replacement are 33.

Average team salary of $138,562,313, replacement salary is $13,625,000, so “salary above replacement” is $124,937,313.

That pegs the true value of a win above replacement at $3,785,979.18 — not $8 million.

I’ll have three articles coming soon, breaking down SV for White Sox hitters, pitchers, and team performance. Within each, I’ll also dig into discrepancies among the three WAR measures, and any other little trivial tidbits I can unearth as we close the books on SV 2018.

More to come.