Your post has a certain beauty to it, the way you put together the problem of how to trade the markets in such a clear, logical and scientific fashion. The "edge" of how to trade the markets has always been there, it's just a matter of the individual trader learning how to find it, and then trade it successfully. MIT Blackjack Team The essence of their strategy was to only place bets (make trades) when the odds of sccess were overwhelmingly in their favor. Those "rules of gambling" apply to trading as well, you just have to figure-out when those situations exist in the markets, and place your bet accordingly.
What worries me about this approach is you are basically data mining the market using your brain rather than a computer. In the process of watching the market and trying to work out what is causing price movements you will develop and test tens of thousands of hypotheses (some consciously, but most unconsciously). You will "pick" out of all these hypotheses a few that seem to work. The risk is that the vast majority of rules you pick will be false and will only appear to work because of chance. If this is the case you might as well take all your money and go to the casino - at least you will get a free drink
Thanks for the complement. I don't expect that trading is easy, or the skills required to be successful are easily acquired, but I have learnt from my day job (analysing complex data sets and predicting outcomes) that only by understanding the underlying physical causes do you have any chance of predicting anything. It is really easy to mine out "rules" that appear to work if you study the data long enough, but which have absolutely no long term predictive value. The human mind is really good at finding patterns in data even where none exist (eg faces in clouds, mythical figures in the arrangement of the stars, monsters in the shadows, etc).
I remember reading in Gary Smith's book about right-brained and left-brained people when it comes to trading. Left-brained people attempt to quantify the market and apply a set of mechanical rules, while right-brained people are more discretionary traders. I'm definitely a right-brained discretionary trader, so I am not qualified to comment about quantitative trading. There are multiple ways to pull money out of the markets. However, as a profitable trader on multiple timeframes I am proof that a quantitative approach is not necessary. Regardless of the path though, know that it will take lots of time. Topper and illiquid have already summarized the rest of my thoughts on this matter. Good luck!
piezoe you da man. I agree 100%. In fact all my pathetic postings on ET where I try to explain my take on trading, should be scratch and replaced by this post. Very well said indeed.
Since some people seem to have found my post entertaining I will continue I have been thinking more about the trading strategies that are used. Assuming that most traders (even the successful ones) are actually using random strategies that they have either mined out of back testing, or from staring at the screen for years, then the effect on the market would be to add what is effectively noise to the price signal. I guess another way to look at this is what would happen to the market if traders started using trading rules selected from the toss of coin. If true random strategies were being predominately used, then even if there once was a predictive signal present in the price data, it would be lost in the noise coming from the "random" traders. This random trading hypothesis has two interesting predictions: 1. Trader's profits and losses will be normally distributed assuming trading cost are ignored. 2. The price predictive signals embedded in a price data series will be less noisy the fewer trades have been made. Both of these predictions are worth looking at further. Starting with the first, the distribution of earnings is not normal. Some traders make or lose more than would be predicted from a random strategy (ie there are fat tails on the earning distribution). This implies that there must be a way of predicting price movements within a market (ie trading can work). While this might seem obvious to the everyone here it does question the efficient market hypothesis. The second prediction is more subtle and probably best illustrated with an example. Imagine Trader A buys a stock based on inside knowledge of an upcoming profit that will significantly exceed market expectations. They will buy the stock, increasing its price from the level it would have been at in the absence of this knowledge. This is a predictive signal to future prices (ie the price movement caused by Trader A is in the direction the stock will take once the news become public). In the absence of other traders in the market, Trader B could use this price movement to predict future prices. If on the other hand the market is dominated by traders applying random trading strategies, then the price signal from Trader A will become lost since the smallest signal that can be detected is limited by the noise caused by all the other traders. In other words, the random trades will swamp the predictive price signal coming from the advance knowledge trades. The second prediction has another interesting aspect, that is, the price prediction signal will be greater for those securities with higher volumes of insider trading. A security dominated by insider trading would reflected very quickly and strongly the future price movements based on the news becoming public. A security where nobody traded on private knowledge would have no future price signal embedded in the past price data (ie all news would be unexpected). Since it is possible to retrospectively identify those securities that do show evidence of insider or private knowledge trading by looking for significant price movements before announcements, it should be possible to identify those securities where current price movement is most predictive of future movement. While the average investor wants to avoid those stocks where the insiders are trading on their knowledge, traders would be best to trade those stocks where lots of insider trading is going on - I guess this would be a "jump on the crooks band wagon" strategy
"Am I on the right track?" No, most probably not, if you would have to ask someone else for confirmation! PS: Not actually finished reading most of your posts yet!
you are thinking way too much and wtf is this shit with some validity level. What you want is profitability. To achieve that you need positive expectancy. Most likely you will eventually want to achieve it trough moderate win-% and high RR. Another story is the tactics ie. different setups to achieve the positive expectancy.
Can you propose how a price/time chart would look for a stock where the big majority (80-90%) of the traders are insiders? Do you assume that all insiders act on the same knowledge? Do you assume that all insiders get the knowledge at the same time? Do you assume that all insiders will act on the knowledge at the same time? Can you accept that some insiders will act sooner than others? Is it possible that some of the insiders might be involved in creating the knowledge? Is it possible that some insiders might not buy into the knowledge 100% and have some doubts? Do you assume that all insiders have the same goals? Do you assume that all insiders have the same levels of greed and fear levels? Will the bigger insiders make their purchases at once moving the price to a new level, or will they try to get as much of the purchase on a lower level by making smaller purchases over some time? Can you accept that some of the insiders will start faking their intentions? Selling when they intend to buy? It seems to me that stocks traded mostly by insiders will show some level of noise that we as traders will need to interpret and bet on.