@ dpruess
It think it's easy to tell Morphy was better than a 1900 rating, you said a GM once estimated him at 1900, but I doubt a GM could tell much of a difference between 1700-2200, they'd all be about the same. I'm close to 1900 and I play many players between 1800-2000 and I can easily tell Morphy would win against this level of play.
@baron http://www.chessville.com/misc/PsychologyofChessSkill.htm
That I think is the article you are talking about (but I'm not sure).
Yes, that has definitely been my weak point, trying to understand positional evaluation. Every chess engine, calculates positional advantages alongside tactical advantages and piece advantages.
The calculation is complex, but moving the bishop to one spot may lead to 0.87 while another 0.95, and this type of sensitive calculation is done intuitively by grandmasters.
I notice when I watch grandmasters play and speak while they play (on blitz games they play), and they think similarly to all other players (there's no genius wisdom in the way they talk, you would expect them to rapidly tell you the various advantages with precision--but instead they are like "oh if i take this, he will do this..." very simple stuff). Yet, they sometimes are not vocal about the intuitive calculations like making the choices they make (Because it's so ordinary for them).
Example "hmm, if I move the knight here, it's a good spot for it... but wait I'm just gonna move it right here that's a better spot"--- whaaaa? How did he arrive at this conclusion? (apparently rybka agrees!) To the normal player, both moves have 99% equal value!!!
@dpruess I think I might agree with that assessment. Of course some time before Steinitz, was there even such thing as elo ratings? When did that come up, so it might be that he was simply the first while ratings were "young".
I too would be curious how they got so strong. My suspicion is that most had masters teaching them or playing with them all the time. If they were the strongest they might have had a close friend who is almost as good that play with them all the time and create theories together.
This is also why you don't see Super GMs messing around on chess websites, because usually they have other super GM friends or regular IM, GM, NM friends that play them constantly and perhaps even fans who are extremely high rated that would play them the second they called.