THE BOOK--Playing The Percentages In Baseball

Additional discussion here:
http://ballhype.com/story/you_too_can_be_a_scout/

I agree. The fan’s report needs to be closely analyzed to see how much merit it likely has. Kudos to Walsh.

There is another similar type of fielding rating system--the Fielding Bible Awards, published in the last 2 B James Handbooks. The main procedural difference is that, AFAIK, the voters are allowed to consider and subjectively weight any info they have, including their own observations plus whatever stats they look at. They don’t have to play this phony game of trying to put the stats out of their minds and vote on observation alone. I think this idealistic treatment of scouting and stats as distinct and independent sources of info is not realistic, or at least not optimal in terms of getting the best results. Every observer has seen X number games as observation, and every observer has seen X amount of stats such as FLD% and UZR.

I say, let each voter mix the observation and stats components as he/she sees fit (according to their knowledge and competence in each area).

As an example, I see most Cubs games on TV. I am well aware of A Ramirez’ reputation and past stats as a bad or at least below avg. fielder. But watching the games in 2007, he looked much better. Actually, he looked impressive. So, when I filled out the fans form, I think I rated him very highly, trying my best to ignore the prior information I had. But, perhaps my best rating would have been somewhere in between. IOW, I would have mentally regressed my ARams’s 2007 observations towards the mean, based on that prior information.

So, are you really getting the actual wisdom of crowds, or are you only getting ‘some’ of their wisdom?

Thanks, Tom.

You’re right about the underrated players. In fact, a straight line is a poor model in this case. So, those lists of underrated and overrated players should be taken with a grain of salt.

As a followup to my crappy post #2 (I would have deleted it, as I was half drunk), I would simply ask-- in addition to the skill category stuff in the fan scouting report, why not have an additional category of “overall” rating, where each fan gets to rate each fielder as he sees fit, using whatever combination of observational and statistical information he has, and in whatever weighting he feels best about? Then, you could still keep the scouting and overall ratings separate, for any desired purpose, and still get the maximal input from the ‘crowd’.

To answer the last sentence in post 2, you are only getting the
observational “wisdom of the crowd”,
and not the
ability to mix observational and objective data “wisdom of crowd”

I’m not sure how beneficial the maximal input is.  The more I ask, the less response I’ll get.  It’s just one more, but then, I might lose some voters too.

If I’m going to expand it by 1, I’d rather it be something else.

---"If I’m going to expand it by 1, I’d rather it be something else”
___________

Well, I certainly will trust your judgement, Tango--but still, for me, the overall rating would have been my favorite (or favourite, to a Canadian ) part of the process. I mean, instead of trying to figure out how good A Ramirez’ first step was (who knows?), I could’ve been balancing my stat info with my overall visual impression of Ramirez in 2007. A much better use of whatever I have to contribute, IMO.