i quoted that one out of all of what was written in response to hazel’s point that (paraphrasing) “all their moves look good.” and jay’s point (at least argumentatively) is that maybe all their moves shouldn’t look good and be blatantly obvious to the armchair GM. the good teams appear to have some sort of “edge” – whether that be better at identifying talent, coaching, understanding something from numbers point of view that hasn’t hit the mainstream yet, so on. it could be the case that the indians are doing a good job process-wise and are just getting knocked by some bad luck or whatever you want to call it. as jay points out, though, perhaps their process isn’t as sound as it appears to be to the rest of us.

we’re not going to answer this – just food for thought.

I’m the author of that linked piece.

First, the piece specifically is presented as a set of theories, a six-course smorgasbord of food for thought. The utter failure of the Indians represents a challenge to the sabermetric community, which generally has admired Indians management and approved of nearly all of its moves. So, what next? That is the basic idea of the piece, not to present any conclusions, but to get the discussion moving.

Second …

Science tells us that the human brain is highly subjective and flawed, that our eyes fail us and that we remember certain things in very specific ways, which lead us to faulty conclusions.

… yes, I think we are all well aware that this is the sabermetric dogma. The problem comes when you conflate “we” the fans with “they” the baseball professionals, experienced people who make subjective judgments about baseball players as a full-time vocation. No doubt, the pros are still human, and their judgments are still subjective, and they are still capable of coming to faulty conclusions. But successful clubs have been built almost entirely on the subjective judgments of those professionals — not necessarily, in fact, in tandem with decent-quality objective analysis, but sometimes to the near exclusion of it. It is here where the sabermetric dogma falls apart, if that difference isn’t recognized. And I don’t mean just paying lip service to the importance of scouts.

Third …

I would say that certain traditional GMs understood how to balance scouting reports with statistical analysis. Maybe Gillick wasn’t using wOBA and UZR, but I’d bet that his scouts used rigorous reporting under very specific guidelines.

… maybe so and maybe not. As it happens, one of the changes Shapiro instituted when he took over from John Hart was to standardize scouting reports and coaching nomenclature across all levels and functions of the organization, so that a 55 rating means the same thing from any two scouts, and one man’s slider isn’t another man’s slurve, etc. My understanding is that standardizing to that degree was a little bit radical, as scouts reported with a wide array of variation within the 20/80, five-tools system. Gillick and his organizations weren’t necessarily rigorous at all, or at least not in the way you’re suggesting.

Fourth …

"Numbers rarely tell us the full story"? "Anything else is pure luck"? These don’t sound like statements from someone who has a deep understanding of math, science, and sabermetrics.

Let’s make sure you’re understanding the context of those statements. The first one is absolutely genuine — the numbers represent only an abstraction of what happened, not a complete description. There is also real, tangible, relevant information that is not found in the numbers at all. Finally, despite our best efforts, the best and most coherent understanding of what the statistics are suggesting still may not turn out to be correct. That is, the statistical interpretation may be as correct as it can be, but what it suggests may turn out to be wrong. And by “wrong,” I just don’t mean that variance blew the wind the wrong way. I mean wrong, as in, wrong. Because numbers do not, in fact, tell the whole story. Hell, with official scorers involved, they’re not even purely objective.

The second statement, “anything else is pure luck,” is meant to be a characterization of sabermetric devotees, not a statement of mine. My statement would be that what is commonly characterized as luck and variance in performance data should really be considered luck and/or random variation and/or other factors which as of yet have not been characterized statistically.

And finally …

I’m extremely reluctant to throw the baby out with the bathwater when saber analysis has gained such a foothold in successful organizations.

… I never suggested throwing out the baby with the bathwater. What the piece suggests is that sabermetrics alone absolutely does not cut it at the major league level. The piece also hypothesizes that maybe the Indians’ analytic viewpoint is not all that sophisticated, and also that Indians scouting and player evaluation may be deeply flawed. A lot of hypotheses are thrown out there.

Anyway, thanks for writing a three-paragraph response despite apparently not having read the article.