Oh this is just unbalanced. How is this 'narrow'? 90.9% chance to win is not 'narrow' in any reasonable sense of the word. A narrow chance for Schickelmooser to win perhaps.

When one side is expected to lose 9/10 times it might be time to give a handicap to the winning side/buff the weak side.

Yes, that makes sense.

But the main point is that fighting opponents in MM who are much better than you/worse than you should not be seen as a 'fair match', at the very least, for this low level of play.
This would be akin to games simply refusing to match higher ranked players to lower ranked players. Of course, ZK is too small for that, which is why we have a 'wide' MM bracket.

The point is that the definition of 'wide' MM should be widened so extremely slanted games where one side has 91.8% chance to win is a handicap game, rather than a normal game.
Its the idea that if fairness in MM is crucial to a good MM experience (no matching red dwarfs to Godde for example), then perhaps we should look about making extremely unbalanced games handicap.

At the very least, for this low level of play. The last thing newbs want to face is someone who will beat them for the entirety of their ZK session on a given day, with it counting as a 'fair' match.

There's no gloating or mockery from Double_Helix_DNA in this thread that I can see, if anything the perception is more for the other side (historically people are far more likely to be put off when they are the ones matched with opponents +400 than -400).

The question is given there are two matchmaking bands (wide and narrow), does a game that is predicted as 90-10% match player's expectations of being a narrow, close match? Or should this pairing only occur in wide since the MM system will normally give you a handicap for this kind of gap? Or where the boundary between the two bands be? These questions seemed pretty reasonable to me and do not imply disdain towards anybody, ratings exist for the sake of the matchmaker, not for our egos.

(The algorithm is of course not an oracle and ranking systems only have an accuracy around 55%-60% at best. The prediction may be very wrong for a single game (people behave very randomly and have good and bad days and maps), but will hold some truth statistically over all games. If your WHR is completely wrong vs your level of play, it will correct pretty quickly, the ladder rating may lag, but WHR jumps up fast.)

Wtf Manero?

Are you actually reading my post?

First you call for modaction against me on another thread, and then here, especially with the misinformation.

I am asking why is '90%' chance to win fair by the matchmaker. It would be preferable if it was not seen as 'fair', and for the weaker side to be buffed, the stronger side to be nerfed, or the game to not count for elo.

Would you vs @Godde be a fair match? No! Is this a fair match? No! Should it be seen as a fair match? I argue that no!
How
Hard
Is
That
To
Understand?!
Lay off mate, this is the same stuff that MrTumnus is dealing with.
Especially with the calls for modaction over something like this

The game was no higher than 88% when it started. At the time of posting it is now 95.7%. It looks like, at the time, you were underrated or your opponent was overrated. Rank updates fairly quickly so there isn't much more to do about this. I'll boost the threshold to 12.5% as my strategy is to make small tweaks until the complaints balance out.