Quote:
Originally Posted by KCUnited
Paul Bessire's Predictalator (cdcox style of simulator) has KC's win total at 79 this year, mainly due the Vargas, Gutherie, Chen middle of the rotation.
|
The problem with projections is that they don't work very well with young teams. They'll do the best they can with them, if you take a AAA ERA or a AAA OPS, we can convert those to MLB-level numbers, and they try to project improvement as a young player ages, but the variance is enormous.
If you take every rookie and 2nd-year player in baseball, and compare their projected 2014 season numbers with the actual combined total 2014 numbers of all those players, it may end up being very close, but the projections for any particular individual young player are going to be way off for many if not most of those players. Young players tend to crush or fail their projections, and those projections adjust until they get old enough to know what to expect.
Once you get a track record and you are looking at 29-year old players the projections get better. There are still some surprises and some variance, but not nearly as much. Those projections are better suited for older teams like the Yankees.
The scouts think Moustakas has fixed his swing and is ready to roll, but the projections have no way of quantifying that. He played decently in 2012 and looked like dogcrap in 2013, so they take something in between those 2 years and project a slight age-based improvement on that. The projections have no clue what to do with a Ventura who the scouts all say has really put it together and will be a ROTY candidate, the projections don't have anything to back that up, and just take his minor league numbers and assume he'll be a typical SP rookie with that minor league pedigree.
If you hard-coded "someone will hit 30 HR, and Ventura will be great" into the system, the results will change quite a bit.