READ: Alliance Changes Part II

Okay, so there have been many suggestions/comments/concerns since our first run. I will try to address as many of them as possible:

1) Accelerating Voting time.
-Every turn, we try to vote sooner and sooner but somehow it just messes up. We want a good head start, but we also want ample time for people around the world to be able to vote. I am thinking we should start voting at 2 hours into the game (to finish up any analyses once we find out Natalia's move) and allow voting time for four hours. Is this okay?

2) We need a coordinator.
-nola2172 and Elroch have taken over the last two moves in my absence but we are running things a bit differently every time. I think we need official coordinators, preferably one who is available during the first 6 hours of the Pogonina game. I nominate nola2172 and Elroch for this position if they accept. We can go over a set guideline so that there will be no confusion as to what to do during voting.

3) Messaging the IMs for a second option before voting
-Sometimes, what we see and what the IMs see are completely different. As such, I think we should assign someone who closely follows the discussion to copy the best lines in analysis and message the IMs, asking for their opinion. This way, they will be able to evaluate the same lines we are evaluating. We should ask them asap so that they reply to us before we vote, else it will be too late.

4) Final Variations- Is this topic even helpful? Should we assign someone to keep track of the best lines presented and post it in this topic? Is there anyone willing to do this?

I think this is the bulk of the suggestions/concerns this turn. If there is anything else anyone would like to add, please post it below. Thanks to everyone for their continuing support.

On weekdays, as long as the start times do not move up any more (it has been slowly sliding earlier and earlier, and is now around 7:30am my time, which is GMT -5:00 or EST in the United States), I can coordinate voting between 2 and 6 hours into our move on weekdays, but weekends are hard for me as I am usually busy during most of the morning.  I will be able to be online around 1 hour or so into the Pogonina game opening up until 9-10 hours into voting.

My ability to contribute much to the analysis during this time is somewhat limited, but I can count the votes and send out messages indicating what is going on.

^True. How about if we have 50% plus/minus 5%(?) margin on the votes, then we play the move the IMs recommend to us? I don't think we've established a method to break ties, anyways.

I think 6 hours after Pogonina moves is simply too late. Any advantage of longer analysis is wiped out by the fact that about 40-45% of votes will have been cast, this makes our contribution insignificant unless the best move is close to the lead already (recall the move before last).

For this move and any other moves where Pogonina's likely replies are 2 or less, I recommend simply doing (conditional) voting before she moves, and publishing the results immediately after the move.  The aim should be to make sure everyone who comes to the discussion early in the move sees the choice, stripped down analysis for clarity. Conditional voting could easily start 8 hours before Pogonina moves, so it would suit most time zones. It is essential that the voting schedule gets told to everyone in the next few hours, and stuck to.

I agree with not presenting close results in the same way as clear decisions. But we should forget about the "rating weighted" vote, at least with the current entirely arbitrary formula. Two 1400s do not balance one 2800 or whatever, and it is misleading to suggest that strength of voters has been taken into account. The development of a weighting to arrive at the most reliable choice would be an interesting research task, but I prefer the idea that the strongest contributers might convince others of their views, based on all the contributions.

The final variations thread should be the most useful output of discussions, but it has not worked well so far. Certainly, we need people to gather the lines together and present them here, right before voting starts.

I think the simple answer about the thought process, Estragon is that there are many people who can see no harm in just voting for whatever move they feel like individually, without either any realisation or any concern that if everyone did that a loss would be almost inevitable. It's not a point of view that is definitively wrong. Like people who think that studying chess takes away from the pleasure of playing without any learning from other people, the arguments against are better results and that study and dialogue make chess more interesting.

Though a lot of the voters do seem to vote randomly, I will note that on the Qd8/Bxd5 move, it was 3-1 in favor of Bxd5 before we recommended Qd8, and about 2-1 in favor of Qd8 after we recommended it.  So, if we have a strong recommendation in the very early stages of voting, I think the group will follow it.

If we have conditional voting in advance, I will not be able to coordinate that (it is far too early for most of us in the United States), but I will submit my opinion the night before.

Yes, the idea is that if we have a long enough period of conditional voting (8 hours minimum) it will allow everyone to be awake at some time in it!

Therefore I suggest conditional voting between 0400 and 1200 GMT/UST. This gives us another 14 hours to discuss the alternatives.

In a worst case scenario that we change our minds later, we could inform everyone of a changed decision, and I don't see how we could be worse off than when the first 6 hours of voting is based mainly on the average drive-by player's personal inclination.

Hm... so voting 8 hours before our turn? Won't any further analysis be useless then? My concern is that you can't persuade people who already voted if they're not there to change their vote. This why I wanted to do voting after we finish analysis (which varies from move to move). If all members voted move A and sometime in-between we find move B is better, move A may still end up being the Alliance move due to the large number of pre-analysis votes.

If I vote one way, then change my mind, I can edit the post containing my vote to reflect my new opinion.  Then the totals shouldn't be made until our move starts in the pogonina game

We may need an electoral college.

I generally agree with all of you:

(1) Conditional voting looks very practical, though much more labor-intensive in some situations. In moves where white has a wide range of moves, we have to stretch our timings a bit.

(2) We should declare our verdict in the main game within 1-3 hours (previously I thought 3-6 hours, but it seems that is also too late) of our time to vote if we want our suggestions to be really effective.

(3) Our voting time should span sufficient time to cover all time zones

(4) Editing of one's vote should be allowed till finalisation.

nola2172 and Elroch, can you guys tally the votes 3 hours after Natalia moves and announce it here and in the main discussion? Also, could you give some "X hours left to vote" etc. warnings before time is up? You can check how much time has passed more accurately if you go to this page

I should be able to assist with that.  At present, the final vote will be 10:30am my time (if we vote 3 hours after our move starts), so I am awake for that part of the voting.