[EM] The "Single Contest" method (pt 2)

[Kevin Venzke]

Hi Jameson,

--- En date de : Jeu 21.7.11, Kevin Venzke <stepjak at yahoo.fr> a écrit :
> I *think* this is what you do, or can do:
> 
> For each pair, find the best possible score this pair could
> have by
> moving the threshold. (So, for each pair you try every
> threshold. The
> best score ever achieved indicates the winning pair.)

So, I tried this with 3- and 4-slots, using only U. Not exactly your
method but it might give you ideas. I found that the truncation 
incentive was horrendous. If I just let the better of the two raw
ratings count as the ballot's vote for that pair, it cuts the 
truncation rate in half (but it's still bad).

I think the voters do not trust the automation...

Kevin Venzke