Jens 'n' Frens
Idle thoughts of a relatively libertarian Republican in Cambridge, MA, and whomever he invites. Mostly political.

"A strong conviction that something must be done is the parent of many bad measures."
  -- Daniel Webster

Monday, October 02, 2006 :::

I previously commented on the election contest I'm planning — in the loosest sense of the word "planning" — to hold again this year. The entry form is likely to be approximately this Excel file; I'm likely to offer a text version just because I'm not a fan of mandating the use of a microsoft format, but it's likely to consist of a list of races and possibly some specification as to the format in which I'd be expecting the entry to emailed to me.

You'll note the first column is simply a list of states that are holding Senate races this year. Years past I've found ways of brushing under the rug any possibility of a race being won by someone other than the Republican or the Democrat; since it's not at all unlikely to happen this year, I'm actually asking for both a probability of the Democrat winning and the Republican. Years past there's been at least an implicit encouragement not to mark probabilities as zeroes or ones; in any case, if you claim a probability as 0 and the event happens, the ordinary scoring procedure gives you a score of negative infinity, which is awfully hard to make up in other races. I intend to mark my own entry with zeroes and ones in many cases this year, and encourage others to do so; the Democrat/Republican/Field probabilities will in fact have ε added to them and 1+3ε divided from them, and then epsilon taken to zero from above. Most likely all scores will converge nicely as this takes place, and the whole exercise will be pure pedantry, but in the event that not just one but multiple entrants do hit the whammy, I can still rank such unfortunate souls, so long as there's some δ>0 such that, for all ε such that δ>ε>0, the same rankings obtain.

The standard rule that any ambiguities will be decided by me, with the possibility of appeal only to me, applies as usual. And, as alluded to above, I do intend to enter. And, in case you'd bother to ask, this is in fact a conflict of interest. If anyone has anything constructive to suggest in the next week in particular it would be more than welcome. Rules, such as they are, should be finalized by early next week, and I seem to be entertaining in my head the notion that entries will be due October 15; that's one of the rules that will be finalized by early next week, and the ambiguity as to what time on October 15 will be decided by me by that point, as will the ambiguity as to what "early next week" means.

::: posted by dWj at 12:11 AM

Comments: Post a Comment

Comment Policy

Dollars and Jens
Steven's web-site

Kitchen Cabinet
Colby Cosh
The Volokh Conspiracy
The Corner
The Bleat from James Lileks
Tim Blair
Daily Ablution
Mickey Kaus
Dave Barry
How Appealing
Virginia Postrel
Reason's "Hit and Run"
Captain's Quarters
Roger L. Simon
Power Line
IWF's InkWell
Blogs for Bush
Chetly Zarko
Signifying Nothing
Cosmo Macero
Hub Blog
Ex Parte from Harvard Law's Federalists
Harvard CR blog
Priorities & Frivolities
Daley News
Emil Levitin
Politica Obscura
Wave Maker
Town Watch
Worcester County Repubs

Election '08
Don't Vote
Dave Barry
John McCain

Other Sites of Note
Townhall columnists Cambridge Republican City Committee
Cambridge Chronicle
Robert Winters
Boston Herald
Boston Globe
Boston Metro
Channel 5
Commonwealth Mag
Fox News
Massachusetts Republican Assembly
Robert Benchley Society

U.S. Constitution
9/11 commission report [7 Meg PDF]
Iraq Survey Group report
Fahrenheight 9/11 deceits


Idle thoughts of a relatively libertarian Republican in Cambridge, MA, and whomever he invites. Mostly political.

Powered by Blogger