The examples given are not examples that McMahon tournaments are somehow bad.
They are examples that McMahon tournaments are not ideal for tournaments that take longer than people need to go from beginner to dan player and that McMahon bars are often badly chosen.
In my particular example, I “only” differed from the player that won by two kyu, and we started off ranked first and second. I don’t think we changed ranks throughout. I was undefeated and beat them in our game… so IMO this case is an example where it’s unfair.
In your particular example are all Players with similar Rankings. There was no Need for a McMahon System.
The tournament leader just chose the wrong Option.
Yes, in your case the problem is that a McMahon is not really suited for players really close in rank, as a 1-2 point difference could mean that a player may win or lose the tourney on the base of the initial ranking.
So I agree, as @selphie said, in your case the tourney director just chose a really bad format. Or, better said, he chose the wrong McMahon bars. I can see that the bars were actually 11k-2k. Given that almost all the players were close in rank and the majority was around 5-9k, I would have chosen McMahon bars around 7k-11k, so that everyone above 7k would have started at 0, and also a 9k would have had a chance to win.
But I don’t know how the tourney was prepared… if the director was expecting dan players to join the tourney, then the McMahon bars would have been fine. It’s hard to tell.
In any case the McMahon system is widely used, especially in Europe, for many of the EGF official tournament, so it cannot be that “unfair”…
I know that the EGF is not the Nihon Kiin, but it’s still a large Go Federation, so I suppose they have well weighted pros and cons of the system.
Yeah, I guess the problem is that the organiser has to preempt what kind of people will join. Hard to know in advance.
It isn’t really a problem that one does not know in advance. I recently had a tourney where the McMahon bar turned out to be ridiculous at starting time, so I manually edited all starting values to 0.
Ah, I wasn’t aware that this could be modified retrospectively. Yes, seems like tournament directors need to be vigilant.