THE BOOK--Playing The Percentages In Baseball

Monday, January 11, 2010

Forecasters Challenge 2010 - Update 1

Change #1 - different style
As I mentioned last month, this time, I will be pitting each Pro in a league of 21 Joes.  So, the Pro drafts first in a league like so:
Pro1 v Joe1, Joe2… Joe21,
Pro2 v Joe1, Joe2… Joe21,
...
Pro22 v Joe1, Joe2… Joe21

Then I repeat with Joe22… Joe42, this time each Pro drafts second.  And so on.  In all, everyone involved, Pros and Joes, will be in 22 different leagues.  The only difference is that the Pros will draft in each slot once, while the Joes will have their voting slots fixed.

Change #2 - 4 different forecasters
Here are the 22 Pros.  Those without an ID means they are new this year.  Those that are “inactive” (MGL, PECOTA) means that I will receive limited involvement by the forecaster (a trustee will need to compile the draft list).  MGL doesn’t forecast saves, runs, RBIs, so I have no choice there.  I’m still trying to deal with BPro to get active involvement.

102 Ask Rotoman Active
ANONYMOUS2 Active
Baseball Info Solutions Active
204 Baseball Primer (ZiPS) Active
105 Brad Null Active
106 CAIRO Active
108 Chone Active
109 Christopher Gueringer Active
110 Cory Schwarz Active
111 Fantasy Scope Active
112 FantasyPros911 Active
113 FeinSports.com Active
Future of Fantasy Active
115 John Eric Hanson Active
116 KFFL Active
217 Marcel Active
203 MGL Inactive
119 PECOTA Inactive
120 Razzball Active
122 RotoWorld Active
218 Steamer Active
Wells Oliver Active

The following declined to participate:
101 ANONYMOUS1
121 RotoExperts
207 CBS (*)
214 Hardball Times

(*) Not declined, but way too much trouble for me to compile myself.

If a Pro wants to join the Challenge, send me an email, and you’ll be on the wait-list. 

Change #3 - Better Joes
I had to come up with an algorithm last year to try to come up with “reasonable” Joes.  Basically, I took the consensus picks, and for 95% of them, gave the picks a random dollar value of +/- 5$, and for the other 5% gave them such a drastically low valuation that it made them undraftable by any Joe.  I repeated that process for every new set of drafts.  This forced the Pros, as a group, to be as good as I decided they should be (though within their group, they stayed the same, relatively speaking). 

This time, I will rely on the Fangraphs readers.  Since David is compiling a bunch of forecasts for a bunch of fans, I will no longer need to guess the kind of valuation Pujols might get.  Now I’ll know for sure.  I’m going to generate 462 Joes out of all the Fangraphs forecasts.  It won’t matter if everyone has enough forecasts.  I’ll make it work.

Change #4 - Weekly tracking
Once again, Fangraphs to the rescue.  Just as David tracked the Clutch project for me two years ago, he has agreed to track this challenge on a weekly basis.  I’m not sure how many bells and whistles I’ll get, or that I’ll program myself.  Maybe this is the impetus I need to get more involved in a stronger partnership with Fangraphs, rather than the very loose one I enjoy.  David is fantastic to deal with, as anyone who works with him knows.

***

Anyway, feel free to add thoughts or suggestions, as always.