Why Second Gaining Zero Will Dramatically Change Higher-Level FFA

Context: Indi's thread regarding the starting position. Martinaxo made the suggestion for 1st: +2; 2nd: 0; 3rd: -1; 4th -1

tl;dr In Martinaxo's rating system, second and first will just team to the end, and dramatically change FFA

Note: this is a partial rehash of arguements I have made before

I think the best way to make this arguement is to use an example. 

Blue: Strongest player

Green: Weakest player

Yellow: In between

Let's say green is really weak and green and yellow have just enough to match blue if they play very well. What will green do? Will green take the risk of losing, and gamble that yellow is smart enough and he somehow wins the endgame? No, he will throw the game to blue.

What about a more balanced game? I will posit, and this, if my memory serves me, is what happened the last time (2019?) we had a similar rating system, green and blue will team and then divide the spoils. If the weaker player loses, no problem, as he still does not lose anything. Players will play it safe, and rarely betray and risk third.

Is this change bad? I don't know if this would be better than the current rating system, but I argue that we should not change to this until will understand its effects. We don't comprehend its effects, and as such don't know enough to be able to tell if this is the direction we want to go. As such, I oppose this idea currently, and argue that we should stick with the current rating system or go back to 3 0 0 -3

Well here it was necessary to publish the direct link to the Thread of the other conversation with Indipendenza, where I explain why this rating system is better with more arguments.

Conversation thread: BY Setup & System Rating 
https://www.chess.com/clubs/forum/view/the-set-up-really-must-change?page=1#last_comment

Whether it’s the best or not, when it’s down to human psychology there are always people playing for 2nd. So some don’t like Solo, maybe for different reasons than what I said though.

If there's one thing I've learned from the merge, it's that players will throw the game whether they get punished for it or not. Changing the rating system won't change that. And I hate Solo, but at this point I think Solo (or reverting) is better than a new system.

I've spoken here about that already many times.

a) I believe that 4th place shouldn't be punished too much: quite often you're 4th without having done anything actually wrong, but just because your opp played incorrectly,

b) if the 2nd gains rating it encourages playing for 2nd which ruins games ; same if the 2nd has 0 : it encourages at some point playing for 2nd, once one estimates he can't win anymore at all, so prefers to help someone in order to avoid being 2nd or 3rd,

c) the difference between 2nd and 3rd and 3rd and 4th in my mind shouldn't be large, so my favourite choice would be something like +3 -0,5 -1 -1,5, (but I would also accept the pure solo 3 -1 -1 -1),

d) on the other hand, it's not good for new players' retention that the 2nd gains nothing. And it is clear also that high rated games, low rated games, middle rated games are different stories.

So yes, I am satisfied with the current morphing solution, but I believe it should morph earlier,

e) maybe the solution that we had 3-4 years ago (when it was becoming WTA (=Solo) automatically if the average rating was above 1800 (if I remember correctly) was correct. What if we reverted to that? Today it could be something like: +3 0 0 -3 for all, but if two or more 2000+ players are present, it's automatically +3 -1 -1 -1.

Makes fully sense; but we also have to think about the "marketing" considerations. I mean the players retention. Many beginners will be discouraged by a pure Solo from start. Because if they lose rating in 80-95% of cases (because it won't be 75% from start: they are beginners!), they will be discouraged to stay.

That's why I believe it shouldn't be the same for all levels, cf. above.

I'm not really refering to throwing (tangent:throwing will always be part of the game but its serverity is influenced by controlable factors)

My point is, this rating system, because 2nd is "acceptable," will lead people to abandon a solo style 3 player stage and team with their opposite to the end. Which isn't really new, but since 2nd: 0 is being proposed I think it's important to be readressed

why not +3  0  -1.5  -1.5 or +6  0  -3  -3   ?

the first player win, the second is draw and the 3rd and 4th lose equally 

I don't think you understand how math works.

+80 would be possible in regular games without Glicko RD inflation with +6.

Having 2nd getting 0 loss would still encourage playing for 2nd for 0 repercussions.

Perhaps there's no 1 size fits all :/

a) 6 0 -3 -3  doesn't look good: I think everybody would agree that a level of a player can't miraculously increase a lot within 20-40 min. The rating variations after just ONE game should remain modest, ratings shouldn't be too volatile (otherwise it would mean that the rating is a random number around the real level which in fact is not very accurate....). Sometimes you win simply by luck. It can't be recompensed THAT much. But the idea 3rd = 4th is correct.

b) 3 0 -1.5 -1.5 is the same of course, but with smaller variations. Could work. BUT.

There are 2 "but".

1) 0 for 2nd still encourages too much the play for 2nd,

2) it becomes far too tempting for RY (or BG) to try to eliminate the sides TOGETHER. That would make much more boring the 2nd stage (3 players), whereas currently it is rather interesting, that's when the games are won actually.

this will be very strange...

i am not sure that is the best

-7 -2 +3 +6  lets go

It exists in fact, try Giveaway (former Antichess).