In response to a poster on Silva's blog

I wanted to know what you guys thought of my argument on yet another stats v scouts propaganda instigated by Silva.

If any holes or inconsistencies are found can you guys point em out for me?  I try to be as succinct as possible in real life situations when this comes up and i can always use the help of AA.  My argument follows

I truly wish you could have an argument with the likes of Quine, Putnam or Hempel because your reasoning is laughable. Effort, work ethic, game savvy are all accidental generalizations that almost appeal to an authority and are not falsifiable which turns this into a Creationism vs Darwinian theory argument.

We will NEVER find the “perfect” statistic much like Quantum Mechanics may one day be falsified like Aristotilian Physics and Newtonian Mechanics.

For a few years win shares was a big one, then VORP, now WAR. Each one albeit imperfect was a MASSIVE upgrade over W/L, Saves, Errors etc which at one time were very useful in gauging player evaluation. The SABRmetric community is attempting to progress and understand baseball better while willingly admitting and discarding out-dated or flawed statistics. Why are some people SO fearful of progress?

I don’t believe WAR is the unified theory of baseball evaluation due to the relatively strong arguments against UZR and +/-. I agree a run saved against a run created can be equal (and there are strong arguments saying a run saved can be more important) but we can’t take these advanced defensive stats in isolation, they play into the entire picture.

Also, the SABRmetric community EMBRACES the scouting aspect of baseball as shown by explosion of Pitchfx. Go to amazinavenue.com or fangraphs and at least once or twice a week a thread is created discussing certain players pitch location and stuff for the game and in regards to the playoffs on fangraphs each Burnett start was broken down by pitchfx.

Finally, the Dan Murphy argument is the ultimate straw man. You create some blanket statement with zero research. UZR is a statistic that the creators emphasize is best utilized on a three year sample size. Murphy played about 2/3rds of a season at 1B and prima facie looked pretty good over there. It is a mild indicator of future success at 1B but were he to turn into Giambi, nobody would be surprised.