If you’ll pardon the big graphic spread, the following is a simple bar graph showing the career WAR for every #25, #26, and #27 pick taken in the MLB amateur draft to date, sorted best (Alan Trammel, 70.3) to worst (Jeff Liefer, -.2.8); although you could make an argument that the giant list of NULL values tacked on after Liefer are “worse” than him considering that they never made it to the Majors to record a career WAR at all.
Note that only 24 of the 147 players listed here (I intentionally expanded the list to include #25s and #27s in order to have a better sample size) break the magic “3.5 WAR” that Dan Szymborski(*) says is “the pre-free-agency value for the typical 26th pick.”
What Dan wants to claim is that the aggregate value of the #26 pick is roughly 3.5 WAR. (We will bite our tongue and not take up the snake-eating-its-tail-logic wherein Syzm generates that 3.5 WAR number by aggregate not merely actual WAR a of Major League players, but also adding in the WAR that players who were not “properly” advanced by their teams would have contributed based on their Minor League Equivalencies, had they actually been in the Majors instead.) But that’s not the case. This is one of those times where “statistics” is properly listed as the third portion of “lies” and “damned lies.” Only one player on the list actually generated 3.5 WAR for his career; Sergio Santos. The rest either generated far more, or far less. This is one of those times where aggregate reductionism just breaks rational thinking altogether.
The #26 pick in the MLB draft is not “worth” 3.5 WAR. It’s worth a 16% chance of getting a guy that might generate or exceed 3.5 WAR. It’s worth a 15% chance of getting a guy that comes in between 0.1 and 3.2 WAR. And it’s worth a 69% chance of getting a guy that either contributes zero or less than zero WAR to your ML club. Rational GM’s would trade a 70/30 split on “nothing” for one year of Ervin Santana and a competitive rotation in the Bigs, every single time. That’s not a “panic move.” That’s thinking clearly about the purpose of running a Major League Baseball team — winning Major League baseball games.
(While we’re here, we might as well note that taking that absolutely made up out of thin air 3.5 WAR and then multiplying it back by the “cost per win” on the free agent market manages to shove the snakes tail back down it’s throat a second loop. But again, even if we accept this absurd loop-de-loop logic on its face and take Szym’s calculations as a given, the +/- $20 million “extra dollars” the Braves are “paying” for Santana is itself so much misdirection and smoke. The number is presented as if it’s rational to add that extra 20m back onto the 14 mil of the actual contract, thus making Ervin Santana “a de facto $33.7 million” one-year contract. The big gotcha there is supposed to be something like “ERMAHGAH, $33.7 mil for one year of Ervin Santana!!!” in comparison to the Braves limited 100 mil per year payroll. This, of course, fails to account for the fact that even if we assume a very fast track for the magic draft pick — say 2 years in the minor leagues — his 3.5 WAR won’t accrue until something like 2022. That’s 8 years. Over that time, the Braves will have invested $800 million or more in player payrolls, which significantly downgrades the BOOM from “losing” that made up from the ether $20 mil.)
(*)For the sake of disclosure, Dan and I were friends for many, many years, and I still consider him as such, though I’m pretty sure he’s ignored me for a year or so over at the Factory. Such is the way of the world.)