WasWatching.com

My opinion on stats projections has improved a lot since I've been frequenting WW and other baseball blogs, but this strikes me as going way overboard.

This is guesswork. It's slightly informed guesswork, but if you have a model where you're projecting season long injuries and team changes and all that this crystal ball aspires to do, I think you need to let it go. Can't we just say, we don't know what kind of hitter he'll be, so let's just wait and find out?

Sometimes it seems like the statheads would rather baseball didn't even exist outside of a theoretical simulation.

I certainly agree DFLNJ. If all these projections and computer simulations are going to tell us what's going to happen, why both taking the field?

John does this just for fun and to start discussions.

I think he sells Cano a little short. He only has hit hit over .300 twice, doesn't give him the home run power he's sure to develop (he has huge opposite-field power), and he only sniffs two hundred hits a few times. John seems to assume Cano is going to be a seventh/eighth/ninth hitter his whole career. If he keeps developing, he could be a three/four/five.

I think people were getting a bit flummoxed by these projections so John put this up a little while ago when he posted one for Papelbon:
http://www.minorleagueball.com/story/2006/5/4/145633/6618
"I get questions occasionally about how seriously to take these Crystal Balls. Remember, these are intended as a starting point for discussion. I put some time into them and try to make them at least semi-realistic, based on what we know about the player in question at the time. Am I seriously projecting that Papelbon will have a 126-85 record with a 3.44 career ERA? Not really. But I do think that his career has a fair chance to look something like what we see above."

That being said, Sickels' trade centers around prospects so it is his job to analyse and project what these kids might do. Putting numbers to his thoughts seem to be more useful than simply stating "this kid has a high ceiling".

I don't think he presents it as anything other than slightly informed guesswork. There's nothing wrong with that. I have strong issues with people applying projections to create statistics, but that's not really the case here.