Should we have a prediction contest?

[Friar]

I think it is a good idea and will be very entertaining.

Every active member have one entry and predicts:

1. Bush's and Kerry's popular vote percentage.
2. EV in each candidate's column.

Closest result in each category receives 1 point, second receives 2 points etc etc...

Picking the correct number of EV will receive a bonus of -5 points and picking the correct percentage of popular votes will receive a bonus of -10 points per candidate (popular votes percentage needs to be in the form of 47.2 and not 47.23).

Each member's points are then calculated and final standings announced.

The winner will have the bragging rights till election 2006 when we can play it with the midterms He will be also declared Atlas Forum Top Pollster for 2004

Contest should start Oct 26th and ends Nov 1st.

Anyone interested in taking part? It can be real fun. Post suggestions to make it more perfect and run as smooth as possible.

[MODU]

hehehe . . . sounds like a fun idea.  Work up the guidelines a bit more, and we'll fine tune them, and we'll put it in place before the 26th arives.

Maybe Vorlon or Dave can put the prediction contest thread at the top of the page, and edit the thread so it's only the figures, no dialogue.

[Friar]

hehehe . . . sounds like a fun idea.  Work up the guidelines a bit more, and we'll fine tune them, and we'll put it in place before the 26th arives.

Maybe Vorlon or Dave can put the prediction contest thread at the top of the page, and edit the thread so it's only the figures, no dialogue.

Rules are easy... just the scoring system is little confusing. That's because I have no idea how many people will enter the contest...

Points can be reversed - 100 points for the closest one, 99 for the second closest... And Bonus points can be 5 instead of -5...

So, maximum points is 200 plus 5 (correct number of EV) plus 2x10 (both candidate's percentage of popular vote) = 225.

I agree scoring system can be worked on but the rules seem quite easy.

[elcorazon]

sounds good to me.

Are we going to predict each state's winner or just total ev's? I think we should do state by state and that'll cut down a bit on ties.  It'd be tougher, but I'd even be up for picking specific margins in each state, with a scoring system designed to award the closest to accurate (of course, that'd make the final tally a bit slower to determine).

[Starbucks Union Thug HokeyPuck]

Yea, make it predict each state, the margin in each state, and the PV.  Closest wins. 

[A18]

Sounds good to me

In that case, I'll finalize my prediction in late October.