Quote from Thunderdog:
Read my post again. You are "horribly oversimplifying" what I wrote. If enough people take flyers, a few will succeed.
Of course that is true. But we are not talking about "enough people." We are talking about one person. Hence, your your reference to the Central Limit theorem is a misuse of statistics, because you are not making a statement about a population. Also, you make a statement about existance, when you should be making a statement about frequency. To say that there will always be someone that has amazing results is meaningless. To say that in a population of x number of observations you expect to see y number of traders with a frequency distribution of outsized returns is what matters. What you should be asking is not whether a VN type result can exist (of course it can, obviously), but rather given some number of traders (observations) what is the probability of seeing outsized returns from some such a sample returns from a trader. In other words, can you tell skill from random luck? I claim VN results are a fat tail, and I further claim that successful traders don't fall under gaussian statistics. Note, I have not done the study, but I trade in a room full of maybe 100 traders, and I can attest to the diffculty of what he did.
It's just a statistical thing and not an indicator of competence. It is the much longer term consistency that separates the wheat from the chaff.
That is false and shows again a lack of mathematical maturity, although I admit to having made the mistake as well.
Statistics and probability are related, but their point of attack are nearly opposite. Their point of intersection is subtle. In probability the variables and the initial conditions are known, and the goal is to predict the most likely outcome. In statistics conversely, the outcome is known but the past causes are uncertain.
In the statement above, you make a statistical statement to say that VN results are within a normal cuve and hence not necessarily representative of skill (you don't say that explictly, but it is clearly implicit in your statement), and then use the statisticl argument then to say that the probability of seeing a VN type result is gaussian. LMAO! Sorry, but that is funny. This is like saying that three = 3. Of course if you assume that VN results are described by a normal curve you can then point to the point on that curve and read off a probabilty! What you should be reading off is the number of people you expect to see having those sorts of results, and then backing out the distribution it comes from.
VN made money because be overleveraged and floored the accelerator, completely disregarding the risk he was incurring. And, not surprisingly, he crashed and burned at the turn. What skill. Want me to give it a try? Give me your money.
P.S. You've read Taleb's Fooled By Randomness, right? And you know that he was talking about VN, don't you?
Yes I have read the Taleb books, and while I find them interesting, the whole premise is silly. How can you make a statistical argument about one thing? You can't as you saw above. Perhaps you have heard the joke, A man has his head in the freezer and his legs in the furnace, so on average he is very comfortable. You can however make a probabilistical statement about one thing, but only after you have worked out the distribution. Talebs "distribution" is "waving his hands in the air" (that may be a bit unfair as he has also been around a large number of traders in his life). Where is the data in the book? Where is the link to the data on the internet? This is not math, it is philosophy.
You have an affliction that most retail if not all unprofitable retail traders have. You cannot fathom that skill may be involved in making money consistantly from the markets, and still blow up, because you have never seen it with your own two eyes. So naturally you seek theories that fit your experience. That is normal. What is not normal is that people repeatedly tell you on this website that oversized returns from tading markets are seen all the time, and that given the sample, it is indicative of skill.
If you believed that assertion a posteriori, and you wanted to continue to be rigorous (which you appear to want to be) you should stop and think how this knowledge affects your belief system.
