[AC]

Quote: Originally Posted by WaitTilNextYear

I don't know, ZiPS and Steamer tend to exhibit nearly the same predictive value. You can cross check one with the other and it's striking how similar they can be. The reason I switched from ZiPS to Steamer about a year ago is because Dan Szymborski's developed something of a grating personality.

On the aggregate yes but it isn't particularly hard to get a good aggregate result (I personally spent ~10 minutes bull****ting projections for an ISU and didn't get too far off ZiPS). The methodology is flawed. They don't regress peripherals properly in any way shape or form. It ends in guys with ridiculous true talent projections like 2 HR/9 (Scott Baker, Gavin Floyd) or BABIP (Odubel off the top of my head).