12 Aug '09 14:57>1 edit
Originally posted by JonathanB of LondonI think you're actually saying the direct opposite of what atticus said? according to your scenario, a 1900 OTB can reach 2200+ RHP by playing loads of low rated players, right? (edit: oh, I just noticed you later said you think fundamentally it shouldn't affect that much...)
I think I would agree with this.
A while back I was asking about average strength opposition when I noticed for some of the top players the figures are very low.
I occurred to me that one way to get a very high rating here is to bash up 100s of players a long way below your strenght. the odd accident taken into consideration, the point or most simple explanation for examples of that phenomenon is very likely to be the corret one.
it's hard for me to imagine how differently a 2200+ thinks chess, but if I transfer the example down to my own level, it seems to me a 1700 OTB can easily reach 2000 RHP. there seems to be loads of people with even greater difference between OTB & RHP, and I mean people who are very unlikely to cheat or lie about OTB rating.
then again, I don't play OTB myself, so it's all hypothesis & hearsay. but I'm quite sure I wouldn't be anywhere near 2000 OTB if I took it up, 1700-1800 seems much more realistic guesstimate to me.
one think though about CC which I've seen with many high rated people already, is that they can delay resignations for quite a long time, and then resign all lost ones at once. my hypothesis is that their real strength is their rating at the bottom of that pit. I'm not sure if that holds, but their pre-resignation rating sure isn't the right one. -but point was that in OTB you can't do that, you play one game at a time, and you can't delay the resignations. and maybe this multiple simultaneous games vs. singular games thing also has much more effect on the rating differences than we even realize? parallel vs. serial processing, you certainly couldn't mix those kind of scenarios in any mathematical context without getting crazy results?