Re: What exactly is the point of the "I don't know"-vote?
Sebastian said:
Second I don't think a lot is won by requiring
the voters to know at least three of the five nominees. Because then every voter will know the
WOTC product and two other random products. This will still be a huge advantage to the WOTC product.
Absolutely. There's an assumption on everyone's part, I think, that no matter what you do, short of barring them from the race, WotC has an advantage. And the will of the majority of those people who've spoken up on these boards is that that's just fine. The point is to diffuse that advantage.
Let's say last year 100 people voted (I know this number is too low, but it makes percentages easy). Of that 100, 60 were EN World members and 40 were people routed to EN World by company websites. Of those 40 people, likely 3/4ths were from wizards.com. So, if you buy my numbers (which are arbitrary, but ther percentages are likely something like realistic), you had 60% "informed EN Worlders," the people you actually want the award to be from, and 40% partisans. And really, 30% of the voters were partisans from WotC in particular. Now, those folks just vote a WotC straight ticket. Which means for anyone else to stand a chance, they need half the "informed EN worlders" and some percentage of their own partisans AND they need none of the ENWorlders to vote for WotC. Those are bad, bad odds and very hard to overcome, no matter how good your product.
Essentially, as it stands, due to the power of WotC name recognition and marketing muscle, everyone else walks in with a huge deficit they have to overcome. I think we can just assume by simple "majority rules, one man one vote (MROMOV)" that WotC probably walks into the contest with about 30% of the vote in any category where it's running. Maybe a bit higher, maybe a bit lower. That's a powerful handicap for the guy who least needs handicapping.
But no one likes the solution of just barring WotC from the running -- and with good reason. Everyone wants the "informed EN Worlders," the people who the award is from, to actually be able to pick a wizards product if they want to.
A weighted voting scheme means that partisans cannot hand their product or company an enormous advantage because they're also being forced to give points to other products. Let's say, for sake of argument, that if the voting were done again on best setting (which I pick as an example not to offend anyone or because I have a horse in that race, but because I know it was a source of controversy):
The wizards.com visitors (30 people) rank OA at #1 and Rokugan at #2. That's 90 points for OA and 60 points for Rokugan. Then a third of the ENworlders (20 people) also ranked OA at #1 and Rokugan at #2, another 60 points for OA and 40 points for Rokugan. By simple MROMOV, OA has it in the bag -- 50 of 100 people have voted for it as #1. But let's suppose (as I think was the case last year) the remaining 2/3rds of ENWorld voters (40 people) don't even think that OA belongs on the list, so they vote Rokugan #1 and something else #2. That's 120 points Rokugan and 0 for OA. Add them up and you have Rokugan with 240 points and OA with 150. How the remaining 10 voters break may decide who's in second and third, but at this point, Rokugan wins -- and it wins because it's on 90% of the ballots, and OA is only on 50% of them.
This is obviously a simplistic model -- the thing about having three weighted votes is that it defies simple modeling, so please forgive what may seem like a straw man argument above, but I have a job and can't really spend all day modeling out a 3 vote race in all its permutations (more importantly, I actually lack the mathematical skill to do it).
But the point remains: a weighted voting scheme allows a product that makes it onto the most lists to beat a product that is loved by a large group of the voters and not thought highly of by another large group of the voters. It does a very good job of expressing the opinions of the mass rather than the simple majority. It also makes it very possible for a product that 55% of the people ranked as #1 to still lose. This is something a lot of people might take issue with. I don't. It can only happen if another large percentage of the voters are really unhappy with the product and don't rank it at all.
Yes, Wizards will still have an advantage. They will still be #1 on a lot of peoples' lists. The will of EN World as expressed on these boards appears to be that people are fine with Wizards having that advantage. The goal is not to destroy it. The goal is to come up with a voting system that does not privilege that advantage to the point that it becomes nearly impossible for anyone else to win.
Hope that makes sense.
Aaron
P.S. The averaging scheme would either have no effect statistically (dividing all the totals by the same number -- the number of people who voted in the category -- just yields a smaller number at the same... gah... my algebra is failing me. Ratio? That's not the word. You know what I'm saying. Where X > Y, if you divide X and Y by 100, X is still greater than Y. I think. Unless I've forgotten everything I learned in high school -- a very real possibility) or would, yet again, privilege the minority (if you divide the total by the number of people who voted for the product, the fewer people who vote for the product, the better off it is.) The weighting system is good because it allows unchampioned products to get into the limelight by being number 2 or 3 on the vast majority of lists rather than #1 on more lists than any other product was #1.
(EDIT: fixed poor wording and typos and hid three easter eggs deep in the text)
