I really can't understand your vulgurity Mike. You show a lot of hatred towards people. It is puzzling to me. Are you posting from inside a mental institution? Are you under some kind of treatment? Do you take medication? Well, let me laugh...What you pointed to is not a quantifiable edge. It is extremely hard to backtest, if not impossible, and it may go for months without a position. More importantly, you cannot size positions to control risk, all these things are admitted by the OP so Mike, tell me something, have you read the post yourself? Can you distinguish between technical and fundamental systems? Is that your edge Mike? Is that your edge really? Where are your backtest results? What is the expectancy of the model Mike? You are crazy Mike, you are the troll and the crazy one. You should seek help if you are not under treatment already. Bye Mike... Only one suggestion....if you keep insulting people like that you will soon have problems of all kinds...Volgurity has the property of coming back and hitting you hard in every respect...
There was nothing absurd about Bill's idea. You are not well...have a doctor take a look at you...Bye...
Do you have anything to say other than personal accusations? I'm still eagerly awaiting the explanation of why my small, yet non-zero probability idea is absurd, other than that you keep on saying it's absurd, but bill's small, yet non-zero probability idea is not absurd. Let me guess, this is one of those things where if I don't agree with you, it means I'm "not well", right? My God, what a transparently rhetorical stance, implying you don't even know why my idea is absurd, you only know that calling an idea absurd is a good way to attack it and that you "feel" that you must attack me for some reason. Because, truly, there is zero logic in your position. The fact that you are defending bill's idea as non-absurd while attacking mine as absurd shows how little you understand about what statisticians do. They use a variant of that metric I mentioned to determine whether things are random or not EVERY SINGLE DAY, thousands and thousands of times in thousands and thousands of workplaces, universities, government departments, etc. A trading method which gets a 6 on that metric, for example, could, in a completely accurate way, be described as a "Six Sigma" process because, as I've tried to explain multiple times without it seeming to penetrate your thick skull, that metric is from standard statistical theory. Are you now going to say that "Six Sigma" is wrong-headed because it doesn't take into account bill's absurd scenario? Van Tharp only tried to market it as something else. The fact that he couldn't get a trademark because of this proves my point. Yet, you and bill have somehow figured out they are all wrong to do so because some dude could really be flipping a coin and generating all the outcomes those statisticians are measuring. You also don't seem to understand that statistics is about measuring the probability that something is random, not a statement of certainty about randomness. If the OP's returns showed up as statistically non-random at some confidence interval, that doesn't mean they aren't random, just that the probability is low. So, bill's "idea" was simply a restatement of what is already implicit in probability distribution theory. Obviously, since you are unaware of this, bill's idea seems like some brilliant insight. Because I am aware of this, I see that his idea is a restatement of the obvious and an extreme outlier not worth worrying about in any practical context of note. And I'm "not well"? And, since you feel free to give me advice, here's my advice to you: Next time you feel like responding, don't. You are only embarrassing yourself, assuming someone as thick-headed as you are is capable of embarrassment.
Dave, I'm calling a spade a spade here; you're acting like a stubborn child. You seem to be fixated on ignoring anything to goes counter to your belief system, and in the process, offending a few successful traders (talon, myself, and a couple others I've recieved PM's from) via your blatant ignorance of certain concepts. Take this simple system as an example. You're calling it non quantifiable because you don't know how, or, are too lazy to quantify it. I'll admit it takes time, but, its definetely quantifiable. To that effect, I'm not going to do the work for you, and, furthermore, are you really that lazy? I know you're intellectually lazy, but with a paper pencil and a few google searches, this edge is not that difficult to quantify. Just admit you're wrong dave. No shame in that. I'm not going to punish you for it. Change these ridiculous beliefs you have and then maybe, just maybe, you'll be able to make some money in this business. Otherwise, why do you even bother to visit this site? ET, although not as sophisticated as sites like nuclearphynance and wilmott, is a very good resource. People with better knowledge exchange info and try to guide those without that knowledge. It's like you think I messing with you on purpose for no reason... that's not it at all. I'm messing with you because you're stubborn and wrong and you won't admit it. Mike
Imbecile logic_man, I have no time to educate you. If you are lookiing for education you must enroll in a college and study statistics. Your understanding of statistics and its objectives is distorted. I am not here to talk about statistics in general. Our subject is the metric you proposed. I agree with Bill that this is not a good metric but for different reasons. Here is an example: There is a trading system that generates entry sisgnals and one trader gets the following trades (in point units): -1, -1, 1, 1, 5 mean is 1 and stddev is 2.19. The value of the metric is 1.01, which is less than 1.65 and indicates a random system according to you and Tharp of course. Now, a different trader with the same system and entry points, meaning the same exact timing of the market, uses some type of risk control to cut losses and ends up with the following trades -0.01, -0.01, 1, 1, 1 mean is 0.49 and stadev is 0.50. Mertic is equal to 2.21. In this case it indicates a non-random system. So according to you imbecile and to Van Tharp, the same market timing, meaning the exact same entry points produce a random and a non-random system depending on stops. That should ring a loud bell in your retarded brain. If it doesn't, don't worry. Your are not alone. Stupidity is so prevelent in this world. You will survive. Now, Imbecile Mike805, I understand you have limited market experience. This "edge" you proposed was a very popular strategy in the mid 1980s. I knew then poeple who lost fortunes with this approach to the markets. Try again...
So what if the system metric is impacted by stops and not solely entries? Risk management can be a "value-add" to a trading system and, if part of a conscious strategy, isn't random at all. Exiting a trade is just as much "market timing" as entering, as anyone who has ever taken a profit too soon or stayed in a trade too long will tell you. Are you noticing a theme here, i.e. that none of your arguments withstands even simple scrutiny?
Like every other moron I have even met, you also deliver answers before reading carefully. Read carefully: Random system: Net profit 5 points Non-random system: Net profit 2.98 points You are telling us that "risk management can be a "value-add" to a trading system and, if part of a conscious strategy..", yet, money management turned a random system into non-random by impacting its profitability by 40.4%. Where is the added value? I think what you basically miss is logic. There are countless contradictions in your arguments. This is the reason you chose the nickname logic_man. It is common that average or below average ability people to select nicknames based on what they are missing or what they want to be and consistently avoid using their real name because they actually hate themselves because they are losers. That is what you are...a loser... Same timing but ---the money management that results in high profits you claim is an added value but that results in a random system ---the money management that results in 40% less profit you claim is an added value and that makes the system non-random No sane man could draw any valuable conclusions from your contradictory arguments, yet you continue jerking off with one hand while responding here with the other...
First of all, no one would apply a statistical test to a trading method after 5 trades. Secondly, let's say you've reached a point where you have a large enough sample of trades to have some confidence in the statistical validity, the metric is designed to answer the question, "What is the probability these results are the product of chance?" not, "Which of these two systems is more profitable?" Thirdly, the point about risk management being a value-add has been shown in a number of experiments on how exits and stops impact profitability, with some evidence that they have as much or more of an impact than entries. And, before you accuse me of being Van Tharp again, I'm just quoting this to make a point, so try to pay attention to the topic under discussion and avoid personal commentary. http://www.traderclub.com/forum/topic?f=10&t=70 "In the System Building workshops that I teach with Dr. Tharp, we play an exit game where everyone enters a series of trades at the same point and then implements their own exit strategy as prices are reported to the group. After about ten quick trials of this game the results typically range from one extreme to the other. A few traders make a lot of money, a few lose a lot of money and most fall somewhere in between. It would be rare for any two players to have the same results. The point of this simple exercise is to illustrate the effect of exits on our trading results. Everyone has identical entries yet the outcome of the simulated trading always ranges from big losses to big profits. "
I would suggest the following method: take the percentage profits you make on each trade and compare it to the equivalent profit you would have made buying and selling the index on the same days. Subtract one from the other, and then check the result for randomness. This calculator will check for randomness. For example the % profits on your trades might be in column a, the % change in the index in column b, and column c is the difference. Check column c for randomness. 5% 8% -3% 1% -4% 5% -3% 7% -10% 13% -4% 17% -8% -7% -1% -5% 1% -7% 2% 5% -3% 5% 3% 2% 16% -3% 19% -1% 1% -2% 2% 7% -5% -2% 0% -2% 0% 2% -2% 4% 4% 0% 8% 8% -1% -2% -1% -1% -3% 5% -7% 8% 8% 1%
Are you assuming you'd buy (or sell, if the trade is a short) on the open and close the trade on the close? I'm actually interested in running this test but want to make sure I understand the parameters.