Although I generally agree with your statements I do not appreciate the offensive language you use. I think you are a genuine and bright guy, however you should express yourself with more cool. Please do not lower yourself to the level of the average ET barking. Try to deliver your knowledge in a friendly manner. There are a very few people here (and anywhere else as the matter of fact) that are fit (or trained) to understand probabilities (Monty Hall experiment proves it). However, the ones who do understand probabilities should always be patient and try to propagate the knowledge so that there is a chance for others to learn. The probabilities, games, random walks, chances etc. are by far more difficult subjects than anyone here led to believe. Keep it cool! Cheers, MAESTRO
OK, then, the probability that you win every trade of over 100 trades in a random experiment is the same probability that Gisele will leave Tom Brady for a studly internet tough guy like you. Happy now?
I've seen your posts with the stats for your systems. There is no help I could get from you other than help stating the null hypothesis. You've exhausted your ability to help me already and I've only been here for a couple of days. I was happier when I thought I was going on your ignore list, believe me. No need to test the null hypothesis on that one.
Then I'm glad I don't use his methods. Whew, that was a close one. Look, I didn't think the OP, who said he didn't have a lot of stats training, really needed to know that what I was talking about was a t-test, blah, blah, blah, so I put it in layman's terms. You got a problem with the "system quality number" take it up with Van Tharp. I was just using it as an example of a number grounded in something other than the OP's original metric, which was being up 7-something percent in a couple of months and wanting to know if that was chance alone. And, since the OP was happy with what I posted, that's all that really matters to me.
Thanks for saying this in a polite way. I give bill a hard time because he comes across as an ASS. That said, he actually does contribute the occasional post with pertinent content. Learning how to interact peacefully with various personalities is a simple skill. Most children learn that in the first few years of school. Mike
Hi Maestro, You are a good man. First of all I want to let you know that I rarely insult anyone first. I only get angry when people are not appreciative when you contribute to their questions even if you are wrong. As far as logic_man, I spent 1 hour from my life to undertsand his problem in this thread http://www.elitetrader.com/vb/showthread.php?s=&postid=2977563#post2977563 and to figure out the correct null hypothesis because he had it wrong. I spent my time for him. At exactly the next post, where I tried to pinpoint to him a flaw in some metric he was proposing, he attacked me like this "I realize after only seeing a couple of your posts that you have a compulsive need to be smarter than everyone, but you might slow it down a bit and r-e-a-d the posts before trying to prove it. I was only trying to give the OP something a bit more to go on in his quest. Don't have a conniption over it." http://www.elitetrader.com/vb/showthread.php?s=&postid=2978756#post2978756 Now in a couple of posts he saw, the first was a post for which he thanked me and the second a warning about some esoteric metric he found in the internet. Now, how do you deal with an idiot like that? I am not a saint. I have a temper I admit it. I trade you know. I did not have a temper before I started trading. Anyway, I cannot stand people who first thank you because you helped them and then try to get back to you because they actually took personally the fact that they were wrong and waited for you around the corner to attack you. I give up my friend. I will take a vacation from ET and I doubt I will ever come back. I met some good people here and I learned a lot I admit, people who know things, even Mike805 is a pain in the head but deep inside he is a good man and knows stuff but we got off to a bad start together. I may come back one day. So long.
Don't leave on my account. Clearly others value your input and there is no reason for you and I to interact in the future. It would have been quite easy to ask if I meant "average gain" in the sense you took it, rather than assume I did and then post some sarcastic comment. Especially since all I was trying to do was give the OP a relatively easy way to understand the value of his method, i.e. to be helpful. As far as I'm aware, no one made you arbiter of how to use the English language and decreed that "average gain" could only be a synonym for "average win". I also find it funny that you call what I posted "some esoteric metric" and consider yourself some kind of stats whiz when clearly what I posted is part of any Stats 101 class at a community college. Why do you think Van Tharp wasn't able to trademark it? It would be like trademarking E=MC2 or some other already known formula just because you happened to use it as part of your marketing. And, your condescension is also all over older threads, which I've been reading to get up to speed on what's been discussed here over the years. Just because I've only experienced it for myself this one time doesn't mean I'm not aware that it's a pattern for you. Maybe others put up with it because they are hoping to learn some trading advice from you because you do seem to have some fairly sound methods, but I don't need any, so I have no reason not to speak my mind.
Logic_man, are you Van Tharp? Only then it would make sense you defending so hard such a stupid formula that measures the dispersion of trades around the mean but tells nothing about drawdown, which is in my view one of the most important system parameters. It is the wrong number to use in evaluating the performance of a system.
Here is what the OP originally stated: "I would like to determine the probability of my success being due to chance alone (and thus not my skill)." Whether you like Van Tharp, hate him, never heard of him, think drawdown is the most important statistic in the history of the universe, whatever, the FACT remains that the metric I provided measures EXACTLY what was originally requested. Didn't you ever do statistics and discuss being "3 standard deviations from the norm", i.e. your results are most likely (99th percentile confidence interval) not chance? That's what a "system quality number" of 3 means. Again, that's why Van Tharp couldn't get a trademark for it. If I were Van Tharp, wouldn't I have tried to claim the number is some kind of original formula I whipped up? The reason I'm "defending" the metric is because it is part of the standard ways in which statisticians measure the non-randomness of events. If Van Tharp abuses it in some way, that say something about Van Tharp, not about the metric. As far as drawdown, I mentioned that as a limitation a few posts ago, actually, so I'm not unaware of that fact. It also doesn't measure trade frequency and opportunity, but, hey, you know what, the OP didn't ask about drawdown or trade frequency, so I didn't mention them. If, instead of mentioning Van Tharp, I had just said, "Do this test, it will tell you how many standard deviations you are from chance", this would have been a non-issue.
I'm not an expert in statistics. No such claim to make here. It is a tough subject. I've just taken a couple of courses during my college years. In statistics, std. dev/mean is defined as the coefficient of variation. Now, you invert that as in: mean/std. dev This tell you how many standard deviations the mean is. The higher the number , the less the variation. In the original ratio, the lower the number, the less the variation. Basically the same thing, inverted. Now, you take the inverse of the coefficient of variation and you multiply it by the sqrt(T), where T is the number of trades. This is what Van Tharp did. I have no idea what that means. If anyone does please let me know.