Vintage Draft Scoring

[johnnya24]

I FORGOT TO MAKE THE POLL PUBLIC ... if this is too close to call, we may need to do it over again. We will not know who voted for what, so people will not be able to change their votes (which is how we have traditionally solved inconclusive polls).

Which scoring method would you like ... additive or multiplicative?

Explanation for the newcomers:

Additive: we add up the category ranking scores just like normal roto scoring.

Multiplicative: we multiple them together instead.

-------------

Two people have voted in the other poll, and not voted here yet.

Long John

[johnnya24]

I voted for additive ... felt multiplicative was getting very stale and a bit boring ... with everyone basically drafting the same (with minor variations).

------------------

On separate point ... I think we need to find a better balance between the two scoring systems. I ran the number previously trying to adapt multiplicative scoring to allow for more variance in strategy: deleting the 1's and the 1's & 2's from the scoring ... but it made little difference I seem to remember. I also experimented with combining additive & multiplicative together, but the harm 1's and 2's can have on your multiplicative score cannot be offset by the higher number of big scores in the additive rankings.

Edit to add: maybe simply eliminating the 1's and the 1's and 2's would work, because the numbers I ran were based on a completed multiplicative draft ... so there was no strategy variance in play to properly test the numbers.

If not, I'm sure there is a way to establish a formula that would prevent extreme punting, while not requiring ABC down the middle drafting. Possibly by bringing in decimal points and docking the average score by 1% and eliminating the 1's or the 1's and 2's. For instance, in a 16 team draft, if we eliminate the 1's, the top ranked team would get 17 points and the bottom ranked 2 points. If we can find a way to keep the top ranked team at around 16 points, and still eliminate the 1's, that could allow for more strategy to come into the multiplicative game.

So for argument sake, something like this (based on eliminating the 1's, and keeping the top score at around 16):

2 (bumped from 1 to 2)
2.5 (bumped from 2 to 2.5)
2.9 (the remainder docked by 0.10714%)
3.9
4.9
5.9
6.9
7.9
8.9
9.9
10.9
11.9
12.9
13.9
14.9
15.9

I'll try these numbers on an old additive draft at some point and see what comes of it. Of course this could all be a load of hokum

[The Feral Slasher]

I voted additive. If you are looking for a variant on multiplication maybe you could add 5 (or some other number) to the score before multiplying. That way the lower scores wouldn't kill you. Or throw out one or two low scores ? I'm fine with whatever

[johnnya24]

I voted additive. If you are looking for a variant on multiplication maybe you could add 5 (or some other number) to the score before multiplying. That way the lower scores wouldn't kill you. Or throw out one or two low scores ? I'm fine with whatever

We can't do anything like that for this draft ... we need to run some numbers and make sure that these changes won't skew things too much. Hopefully by the time VD2 comes around, we have some data to fall back on.

When I deleted the two lowest points, it only made a marginal change to the scores, but like I said, that experiment did not account for strategy variance. But if I remember right, one or two of the teams had a lot of 1's and 2's, and it didn't make that much difference to their position.

[SeaDogStat]

I voted multiplicative

[heyelander]

additive. I like the strategies even though I don't know how to implement any of them.

[johnnya24]

Someone has voted in the other poll, and not voted here yet. I'm guessing the following, plus one other, have yet to vote in this one.

MJL
Long John
TSGarp