With the 2014 MLB draft upon us, you will likely here the following phrase a lot in the next few days: "you can't scout a stat line." It's a cliche, and like most cliches it contains a great deal of truth, but it isn't the whole story.
Ideally, you want a player's scouting reports and stats to match up. If the guy has great scouting reports and great numbers, well hey, you're (probably) set: the guy is a good or great prospect and his story is (probably) going to turn out OK, provided that something weird like a serious injury or emotional breakdown doesn't occur. That's true in regular life, too, isn't it?
But what happens when the scouting reports and the stats don't match up? There are two ways this can happen: a guy with good stats but mediocre or weak scouting reports (say Matt Carpenter in the spring of 2009), and a guy with weak stats but strong and optimistic scouting reports (say Matt Holliday when he was a mediocre Double-A outfielder in 2003).
In such cases, do we just throw out the stats? Of course not: nobody does that, even people who say they don't scout a stat line still look at the stats. The point here, in cases where the stats and the scouting don't agree, is to figure out why they don't agree, and to project which side of that disagreement will win the battle in the end.
The examples of Carpenter and Holliday both had positive outcomes. Carpenter was extremely successful in college, but lasted until the 13th round due to mediocre physical tools. The numbers said that Carpenter could play and they turned out to be right even when the early scouting reports were not optimistic.
Holliday took three years to get out of A-ball and had problems keeping his SLG above .400 in the minors, but he was a good athlete and eventually figured out what he was doing, to the point that his major league career OPS is 130 points higher than his minor league career OPS. The scouting reports about his physical potential won that case.
There are many other examples.
If a scouting report says that a guy is going to be a really good hitter, but he's currently hitting .206/.292/.312 in rookie ball, that report needs to explain why he should be expected to improve and give a plausible explanation for the case. It needs to explain the data available, not just ignore it. And the good explanation may very well turn out to be the correct projection, as it did in the .206/.292/.312 case.
If a scouting report says that a guy who hit .478/.558/.862 for a Division I school in a major conference, then followed up by hitting .355/.431/.525 in a difficult pro-pitching A-ball park, won't be able to thrive at higher levels, at least not to the point where he would be worthy of an early pick, it needs to explain why. And in that case the good explanation why stardom should not be expected was correct, too.
You can, of course, find numerous examples of guys with great tools who never figure out what they are doing with them. Ruben Rivera. Drew Henson. Joe Borchard. And a bunch of A-ball flameouts you never heard of.
My ultimate point here is that we shouldn't throw out data. All information, including statistical information, needs to be examined when looking at a prospect, even at the amateur levels. And that's doubly important when individual bits of that information seem to contradict. Sometimes the numbers will win over the scouting reports, and sometimes the scouting reports will prevail over the numbers.
It is true that you can't scout a stat line. But you sure as hell need to explain it if it contradicts the scouting reports, for good or ill. If the guy is better than his numbers, or worse than his numbers, we need to know why.