Sklansky's Handicap

[David Sklansky]

This is a really, uh, dumb questions. Being smarter makes you a FAVORITE. Genuises who take up poker are perhaps a fifty to one favorite to beat a randomly chosen above average thinking person who has also taken up poker. Since at least 100,000 of those people exist, 2000 figure to be better than the (random) genius. You can't choose them after the fact.

[BeerMoney]

So the answers are Yes, Yes, Yes, I don't know?

[David Sklansky]

So you were you asking me a poker question rather than trying to refute my point about intelligence?

Intelligence is to poker as 100 yard dash speed is to being a defensive back. Its a big factor but not everything.

[southerndog]

[ QUOTE ]
So you were you asking me a poker question rather than trying to refute my point about intelligence?

Intelligence is to poker as 100 yard dash speed is to being a defensive back. Its a big factor but not everything.

[/ QUOTE ]

I agree with most of the things you say about intelligence. I was hoping to get a non-defensive answer from you.

Something like "I'm a tell-bag" or, "I have no gamble" or "These guys don't respect my raises.." But seriously, I would like to know.

BTW, I think anybody that is incredibly fast enough play football.

[VanVeen]

Dude, how can you make the argument that one's ability to solve randomly chosen problems has no correlation with his ability to solve a randomly chosen problem?

Of course it's true that if we define intelligence as one's ability to solve randomly selected problems that those possessing a greater helping of intelligence are more likely to be correct than persons possessing less intelligence when they encounter a novel problem and have the same data to work with. Admittedly, there is no reason why 'intelligence' must be the most important criterion when handicapping a debate, but to say it 'may not be relevant' is nonsensical.

[ QUOTE ]
And doesn't the fact they're all in disagreement indicate that jumping to conclusions based on personal experience might be hasty?

[/ QUOTE ]

Actually, any psychometrician worth anything would agree that the data conclusively demonstrates a causal relationship between one's 'raw neurologic potential' as measured by standardized tests of fluid g and one's financial success and self-reported level of personal satisfaction. How you define 'success' determines what kind of conclusion you will draw, but using popular objective metrics of success there is no debate whatsoever.

little data correlating 'confidence in performance' (after the fact) and actual performance, and i think the correlation is lower when the respondent is 'very confident' and 'very unconfident' (at the extremes). mentioned in some popular sci books and i'm sure the studies are online somewhere if anyone cares to look.

[FortunaMaximus]

So you're saying a more confident intelligent person has better results?

Huh. People go to university to learn how to define that too?

I certainly missed out on some awesome stuff then.

Wonder how Kim Peek did on his SAT's. Oh, wait.

[madnak]

[ QUOTE ]
Dude, how can you make the argument that one's ability to solve randomly chosen problems has no correlation with his ability to solve a randomly chosen problem?

[/ QUOTE ]

I can't, that's an absurd argument that I'd never make.

[ QUOTE ]
Of course it's true that if we define intelligence as one's ability to solve randomly selected problems that those possessing a greater helping of intelligence are more likely to be correct than persons possessing less intelligence when they encounter a novel problem and have the same data to work with.

[/ QUOTE ]

Depends on the problem. Assuming we know nothing about the problem, and must treat it as a random problem, then yes. Assuming, on the other hand, that we have some information about the nature of the problem, not necessarily.

But intelligence isn't "one's ability to solve randomly selected problems." In fact, "randomly selected problems" don't exist - every time a problem is selected, it must be selected according to some criterion or another. The set of all problems is very infinite. And actually, I believe the "no free lunch" theorem can be applied here: over the set of all possible problems, no problem-solving strategy has any advantage over another.

[ QUOTE ]
Admittedly, there is no reason why 'intelligence' must be the most important criterion when handicapping a debate, but to say it 'may not be relevant' is nonsensical.

[/ QUOTE ]

I'm saying it may not be relevant in a particular debate, not that it may not be relevant at all. I admit my language was messy there.

[ QUOTE ]
[ QUOTE ]
And doesn't the fact they're all in disagreement indicate that jumping to conclusions based on personal experience might be hasty?

[/ QUOTE ]

Actually, any psychometrician worth anything would agree that the data conclusively demonstrates a causal relationship between one's 'raw neurologic potential's measured by standardized tests of fluid g and one's financial success and self-reported level of personal satisfaction. How you define 'success' determines what kind of conclusion you will draw, but using popular objective metrics of success there is no debate whatsoever.

[/ QUOTE ]

I very much doubt that. Either you're using a true scotsman standard in talking about psychometricians and whether they're "worth anything," or you're confusing correlation and causation. Social science can't even isolate the variables, much less establish causal relationships.

But I'm not talking about correlation between success and metrics, I'm talking about the definition and nature of intelligence. And no, the psychometricians are definitely not in agreement about that. Maybe moreso than about personality, but still not much at all. And there are actually some very interesting trends that defy your oversimplification. The high incidence of mental illness in geniuses is one of the big ones, I think it's a huge consideration and one of the most important relative to this discussion.

[ QUOTE ]
little data correlating 'confidence in performance' (after the fact) and actual performance, and i think the correlation is lower when the respondent is 'very confident' and 'very unconfident' (at the extremes). mentioned in some popular sci books and i'm sure the studies are online somewhere if anyone cares to look.

[/ QUOTE ]

Your paragraph got chopped off, but it sounds like you're agreeing with me.