hmmm, throwing out non sequiturs in an futile effort to discredit, typical tactic of a scoundrel. comprehension of the written word, not one of your strong points? i'll state it again, i admit that LW may have had some unsavory situations in the past regarding this business--- however, his cogent responses to the allegation makes sense and your the one who will not take the time to find out for yourself, nor believe anyone who doesn't scorn the man. best wishes, surf
i kiss no mans ass. if LW didn't fully satisfy my initial cynicism toward him, i would be singing a different tune. here's his contact information should you choose not to be driven by your bias: http://www.ireallytrade.com/ best regards, surf
Quote from oddiduro: In short, this is a pretty good admission that trends only become trends after the fact. The reason why so many people do not succeed in trading is because they all think they know all there has to been known. So, per definition, everything that is said and does not correspond with their own opinion is rejected. They place their own ego above anything else. A better approach would be to listen to others, try to understand and verify what they say and try it out. The mathematical results will objectively and clearly show if the statement was correct or not. If the statement would appear not to be correct, there are two possible conclusions: First: the statement was NOT correct. Second:maybe the person who did the test was not able to achieve what someone else could. I included 40 trades on the ES, for which I show the distribution of the P/L. My conclusion is that markets are not random because the distribution is not equally divided between profits and losses. There are much more profits and they are also bigger. From -4 to +4 we see the an almost identical distribution, but on the positive side it continues till 26. This clearly shows that there are trends that can be exploited. Can trends be seen in advance? Not with a 100% accuracy. But there are ways to maximize the probability BEFORE entering a trade (to me this is equal to predicting trends but with a lower accuracy). That is clearly proven by the chart as we see between -4 and +4 the failed attempts to catch big trends, but the trades between +5 and +26 clearly prove that there are trends. 14 trades are in the failed attempts and 26 trades are +5 points trades, which to me, are considered as trendfollowing trades.
How many data points (sample size) do you need to reach a conclusion like that ? I think you used way to few for your conclusion to be valid. Then, I think most people do not consider a four or five point gain the result of catching or identifying a trend. I think when people refer to trends they are talking weeks, months, or even years. Noise trading is far from trend following.
The sample is part of an +1000 trades sample. The chart is almost identical except for the fact that in the extrememe profits (+15 points) the frequency is a bit lower. The negative figures are identical to the bigger sample. All depends on the definition of the word TREND. To me trend is the major move within a specific timeframe. Trend has nothing to do with days, weeks or months. Trends do also exist in minutes or hourly timeframes. Trend gives the direction of a move, time is irrelevant. (Or is a 20 points move intraday noise, but the same move in 3 days a trend?) Noise is the data that disturbs a perfect trend in a certain timeframe. In my example this is certainly not noisetrading. Noise must be defined in correlation with the timeframe your are trading in. An intraday trend can be qualified as noise if you trade on the longer term, but in daytrading it will remain a trend.
Spike, Your chart is interesting, but what does it really and truly show? The only thing that they really and truly show is that you are able to use a set of rules to try to catch directional price movement. As traders, we have dozens of patterns that try to say that if you follow them, then a trend will ensue. But that cannot possibly be the case with any predictability. Some traders will try to increase their odds by using a cluster of these patterns, whether mathematical or pictorial. Still, certainty in this business does not appear to exist. I am not a quant, I am a chartist. I can tell you that every pattern that has been created has been falisfied. Trading is not a science, because the rules that we create produce variable results. H2O is always water in science ( heavy water included), but a rickshaw doji at a fib retracement on very high volume relative to that market on the day after the first full moon in October does not always mean a price reveral of 20 points in the ES on the next day. We have very sophisticated ways of doing it these days, but the end result is the same........ we are only guessing Best Regards Oddi
I agree completely. I think it is a bit presumptuous for others to dictate what time frame justifies a trend for everyone else. Regarding your sample size, I would have to agree with the previous poster who suggested that the sample size is too small to draw any meaningful conclusions, especially when it pertains to the financial markets. Even so, I don't believe that the markets are completely random, although they certainly seem to have a fairly large random element.
Well, there is TA and then there is voodoo. I would avoid the overlap. Very true. Although some approaches allow for better guesses than others.