Skill or Chance?

[J-Pav]

[b:ca0e4943b3]Marcus:[/b:ca0e4943b3]

Good post. I'm gonna have to spend some time doing my own digging. I like the direction it's going though.

I'm not a game theorist or a true stats guy. You seem to be on the right track; however, I'm wondering about two things:

If Joe Average plays .500 over 54 seasons, from his perspective he has irrefutable evidence his game is pure chance, even though several players may be both above and below .500. Neither of us would argue though that the game is pure chance because we're on either side of .500.

Since you have a perfect bell curve around .500, how do you know that the farther right and farther left are not the statistical outliers of many, many coin flips? I know from some "fun with numbers books" you can get outrageous streaks of heads/tails in large numbers of coin flips.

The same arguments are made in mutual fund manager results. Do you get paid in the way of return for hiring a certain money manager? Peter Lynch would say yes and have a track record to back it up; however, statistically, the evidence is overwhelming that overall you do better dollar cost averaging with a low load index fund.

Warren Buffett argues against Random Walk with [i:ca0e4943b3]The Superinvestors of Graham and Doddsville[/i:ca0e4943b3].

Food for thought...

http://www.investorhome.com/coinflip.htm

Here's some irony. If the "Old Guard" keeps winning, then it demonstrates that chance is less involved. But if the "Old Guard" melts into the pack...

[J-Pav]

[b:d2541b6271]C2:[/b:d2541b6271]

This one's for you (your guy Paulos).

http://www.investorhome.com/scam.htm

Here's another one...

http://www.investorhome.com/mutual.htm#do

Raising the question, if you're a newb trying to win free credits, who would you pay what to manage your team?

[MARCPELLETIER]

[quote:e82994d342]Since you have a perfect bell curve around .500, how do you know that the farther right and farther left are not the statistical outliers of many, many coin flips?[/quote:e82994d342]

This is exactly what the p value means. It is the probability that the specific result you are looking at results from statistical outliers of many many coin flips.

For your 2005 season, for example, the p-value is 0.00001, which means that if your coin is not broken or anything, the probability that you flip this coin and obtain 53% of heads (or tails) after 54*162 tries is about 0.001%. It is so extremely low that it is much more probable that you have done something with your team which had better impact than random decisions.

[cummings2]

[quote:ab9bf228a3]
[quote:ab9bf228a3]AAAAAAAAAAAAAAHHHHHHHHHHHHHHHHRRRRRRRRRGGGGGGGGHHHHHHH.[/quote:ab9bf228a3]


You complicate your life way beyond reasonable questionnings. If you have a wife, or a child, it is time to take them in your arms and feel their love. [/quote:ab9bf228a3]

:lol: :lol: Ooooh boy, this one really cracked me up. Very nice one M.W. :lol:

J, thanks for the links.

Like the direction this is going...just need some time to chew on things a bit :D

[J-Pav]

[b:95c892153f]Marcus:[/b:95c892153f]

This is a pure stats question, I think, and maybe the Fisher thing doesn't answer this question.

If we use .05 as our baseline, applied to ALL participants, what number of managers do you think we'll find from the total number of managers in the pool (1800, say) under .05? Then, when we have this number, how do we [i:95c892153f]prove[/i:95c892153f] it isn't a number you'd see anyway in a regular bell curve distribution?

I know you get the answer if you apply the formula to an individual manager. But what happens when you apply it to the pool (again, zero sum game equals .500)?

Maybe we're on to a new and improved manager ratings system and HAL will lob us some free credits. At the very least, it will give all these fanatical newbs a true target to measure their game by.

[J-Pav]

I like the fact that the more you play, the more you have proven (mathematically, even as you [i:93e8f13abb]approach[/i:93e8f13abb] .500). I'm as sensitive about this Old Guard thing as you are about your beta teams! I'm offended when I hear things like "former SOM legend [b:93e8f13abb]RiggoDrill[/b:93e8f13abb]..."

At least I like the company at the old guard's table!

:D

[cummings2]

[i:8b17a984a8]"former SOM legend RiggoDrill..." [/i:8b17a984a8]

:shock: :shock: Who the katookies has said this :?:

I like poking Riggo, he'll beat me anyways so I might as well have fun while doing so.

[J-Pav]

[b:ccec645955]C2:[/b:ccec645955]

You're not listening. You've got a 50/50 shot! :wink:

[J-Pav]

Bump, to move the [b:9d8573261b]Jewish Prophet[/b:9d8573261b] further down the list. :D

[J-Pav]

Wow. The moderators cleaned out all those goofy posts rather quickly. Nice job, crew!

:P