Originally Posted by Soylent Green:…Stats4gnats - Of course scouts use stats. But I'm talking about the more general "battle" in the game between the scouting eye test vs sabremetrics.
That’s my point. There really isn’t any “battle” at all. Both do the exact same thing but use different methods and perspectives. The problem is, amateurs more often than not try to look at things as though their perspective was the scout’s. Rather than seeing the numbers as nothing more than the average of all performances, they try to extrapolate Joey’s HSJV .575 BA into what would happen if he were in the ML.
And let’s not leave the impression that the “scouting eye” is omnipotent. They’re good, and they’re much better than any other group, but there’s no way their efficiency and therefore the quality of the chosen player couldn’t be improved. 
It's strange to be arguing this side of it, because I'm actually a stats nut myself and generally like sabremetric analysis. I just feel that the pendulum has probably swung too far in favor of stats in the past few years. I think a proper balance will develop between the two over time.
I don’t think that’s what’s happened in the venues where it counts, i.e. play for pay. Where I think its happened is in the amateur levels where technology had made great leaps, making numbers available to people who never had them. I see it all the time where someone started using something like IScore and is blown away by the raft of stats the thing can kick out. All of a sudden they think they’ve found the “answer”, but they haven’t quite learned that that “answer” has its limitations. So they start making all kinds of decisions based on something they don’t yet understand.
Then there’s always the problem of the validity of the numbers. At the amateur level there’s a very good chance the person scoring, or gathering the data to be used in generating the numbers, doesn’t do a very good job. So no matter how slick the software generating the stats is, its GIGO. But the good news is, things are getting better. The problem is, the expertise goes from the top down, and that’s a very long process because so few people from the top filter down to the bottom. 