Increasing proposal difficulty

[paul]

Dammit! OK let’s try again:

What should the new Minimum Conviction be? (currently 2.5%)

• 3%
• 3.5%
• 4%
• 4.5%
• 5%

[sem]

The results of the poll are:

• 45% of votes on increasing to 3%
• 22% of votes on increasing to 3.5%
• 33% of votes on increasing to 4%

3% is the option with most votes, but if we consider that people who voted 4% would prefer a 3.5% than a 3%, we can assume that increasing the minimum conviction to 3.5% is what voters actually prefer, even though it is the one which received less votes (22% of votes), it is the stronger choice (22% + 33% = 55% of votes).

[Monstrosity]

Yes, agree. Or put another way, if we weighted the votes the average comes to 3.44% so 3.5% keeps it simple to remember and a cleaner number.

[D0SH]

I like the the approach of taking the average but these results are skewed. 45% voted 3% and 22% voted 3.5% which equals 67% of the votes. A proper solution would be weighted average of all not just the 3.5+4% of the voters

[Monstrosity]

that’s a really good point I overlooked. We never gave the options for lower than 3% so of the choices given skewed the results in favor of a higher min conviction.

[sem]

Doing the average is what I initially thought, but there was this problem of not having requested lower values, so it would be unfair to do it. I think we should read the data overall, and that’s it.

• In @Monstrosity poll we have a 73% of people who requested increasing the minimum conviction, and a 27% who preferred to keep it as it is or to reduce it.
• We can assume that all those who voted not to increase the minimum conviction voted on the 3% option of the last poll. So tentatively we can think:
  • Less than 3% - 27% of votes
  • 3% - 18% of votes
  • 3.5% - 22% of votes
  • 4% - 33% of votes
  • More than 4% - 0% of votes
• The vote is concentrated on two extremes: keep it as it is (2.5%) or increase it to 4%. If we join the two first groups (less than 3% to 3%) and the two last groups (3.5% or more), we obtain a relation of forces 45-55%, which is defines where the parameter should be. I think it would be safe to put it just the middle: 3.25%. Does anyone else have a better solution to interpret the results? It is impossible to find the best one, but I’d like to hear other ways to see it.

[D0SH]

Thank you for the explanation Sem . I think going towards the 3.25% would cover some of the lower percentage voters. We could do a final YES/NO poll to switch to 3.25% that could be the final deciding factor based on majority. If NO we could re-poll using a bigger range

[sem]

I agree. Let’s do the poll.

• Should we increase the minimum conviction to 3.25%?
• Should we repeat the poll with more results?

[sem]

The vote to increase minimum conviction to 3.25% is on in the garden.

The evmscript used in the terminal is:

connect 0x8ccbeab14b5ac4a431fffc39f4bec4089020a155 disputable-voting.open @context:Increase minimum conviction
exec disputable-conviction-voting.open setConvictionCalculationSettings 9999799 1000000 3250 200000000000000000

The proposal will be open during the next 5 days.