Here is where I stand on the whole "stathead" vs "old-school" debate.  I am not a stathead because I really enjoy watching the game and thinking about the strategy and the chess match of a live a game.  I am also not very good or diligent about researching and understanding the stats that I am looking at.  However, I love reading about stats and stat comparisons.  I admire that ability.

His blog, I mean article is dumb.  The everyday fan is not driven away by things like VORP.  They just ignore it.  People that don't enjoy stats are not paying attention to the "statheads".  If they were then the true fans would be the ones that have the deciding votes for All Stars and not the journalists.  Unfortunately the journalists have the ear of the everyday fans and that is why David "Professional DH" Ortiz is the starting first baseman for the All Star game.  The Yankees and Red Sox and there journalist lackeys are the ones driving away the fans.

To the extent that some people use them as "proof" that one player is "definitely" better than another, I can understand his criticism.  It annoys me when somebody quotes 2-3 years of VORP like its a Holy Grail stat.  There's certainly parts of the game that are near impossible to measure (defense and clubhouse morale in particular).  Dismissing these factors as "made up" is as unfair as Chass's assertion that measuring VORP is a bad idea.  The guys at Fire Joe Morgan are probably just as guilty of perpetuating ignorance as the writers they blast.

Finally, I've always felt there's something elitist and disingenuous about quoting VORP, especially to a regular fan that's not familiar with advanced math.  Win modeling is the best way to measure a stat (i.e. - a stat is "good" if the team that has the higher number usually wins the game).  "Runs" is the perfect team modeler for wins because the team that scores more runs that its opponent, wins 100% of the games.  Therefore, you want to find players that will increase the number of runs that your team scores per game (and/or decrease the number of runs allowed).  Hits, actually tracks wins pretty well at 71%.  Batting average is actually, a really good statistic all things considered because it tracks wins at a 79% clip.  OBP tracks wins only 1.2% better than batting average.  The difference between Batting Average and OPS is another 5% of improvement.  Once you get around the 85% mark, the laws of diminishing returns kicks in.

Stats like VORP, WARP, and even RC only serve to tack on an extra 1-2% more "certainty" to the stat.  If you delve into the play by play data you can squeeze another 3-4% of win correlation out of the numbers, but again, you're really not talking about a huge leap in "truth".

My feeling is that VORP is an excellent academic exercise, but I definitely question the reasons for its growing usage.