We had a discussion in the main forum. But to not make it off-topic, I moved it here. This is the link to that discussion:
http://en.lichess.org/forum/general-chess-discussion/why-am-i-not-in-the-rankings-in-the-bulletSo, my last argument.
Cynosure, you have contradiction in your words.
For example, you say:
"I'm saying in 1+0, an opponent who is 100, 200, 300 even 400 rating points below you can be more assured of victory than if you were playing a 90+0."
But if it's so, it only means, that for Singer__Marta it's only difficulter to gain rating. Because if, as you say, in bullet, those who is weaker can be more assured of victory, then 2000-rated players, who play with Singer, are more assured of victory, so it means, that Singer has less chances to win. And it only means that he has to work harder to gain his 2800. And it would mean he is more than 2800.
No, you are just so obviously adverse to logic that even posters who are well versed in statistical methodology and mathematical logic think you're incorrect and have stated why several times. Each time, you refuse to actually engage with their points, but restate yours. The reason I would close the discussion is because you are, in effect, trolling, and not producing any new material for discussion, just obstinately repeating your same points over and over again as if that is some mantra which will deny your intellectual opponents of their well-reasoned points based upon their study.
My own point still stands. A weaker opponent is more likely to win a 1+0, but there comes a point where the weaker opponent will make so many more inaccuracies, mistakes and blunders than you then you will win every time. For myself, I could win, without really trying, against a sub 1500 player in 1+0. Around 1650 (250 rating points below my average) it becomes more of a challenge, but I'd still usually win 90% of the time. 1700, maybe 80% of the time, 1750 maybe 75%, 1800 around 65%, fellow 1900s is more of a 50/50 split, probably a marginal 53% of the time.
However, if I were to play them in a 90+0, that sub 1500 player would make less mistakes and produce a more consistent, strong game than I would expect in a 1+0. Similarly, a 1750 may win against me 30% of the time with a higher prevalence of draws, and an 1850 may win up to 55% of the time, if you include a draw as a weaker players win (which I would).
Someone my level, in a 90+0 should theoretically produce about 40% wins each, with 20% draws.
What I am saying is that in a 1+0 you will not face the same level of competition as you would get otherwise. A 2000 player can expect victory against a 1700 more than the formula used for 90+0 games would assume. As I've stated, in a 1+0 a 2000 rated player will beat a 1700 rated player without really trying. On the other hand, the 90+0 game will be much closer. This goes back to my original point, and to yours.
You assume that, regardless of the amount of time played, the same predictions for victory can be made. I disagree. Time is a factor in bullet as much as skill, understanding of theory, and endgame knowledge. To ignore time as a significant factor, impeding a knowledge of chess is grossly ignorant.
@Cynosure I'm going to use your first para, as is, in so many places! I can already think of so many forums and people that would be annoyed ... yum yum.
The answer has been presented and no one seems to have anything constructive to add. I'm sealing this with a ritual.
