True. But the premise of the idea is : You must have some knowledge in order to be a successful trader. If that is true you must lose money if you pick randomly and use big enough sample of trades. . You can take out beginners luck out of the equation if you run a simulation of multiple clueless traders and thousands of trades. Then the opposite of this random outcome must be a successful trader, I think but not sure about it. You may run a simulation where 1000 clueless traders make 1000 trades each a year. From those 1000 traders you may have X amount of really successful traders but if you average all of their returns they should still be losing money . I would be really curious if someone who knows how to back test run a back test simulation with 1000 accounts and 1000 random trades or just 1milion random trades.
I think backtesting is the only way to find out and we can have only these outcomes: 1) random picks made people lose money on average. so doing the opposite would make you money . Conclusion would be: You don't need any knowledge to trade just take the opposite side of a random trade. 2) random picks turned out to be profitable Conclusion would be: You don't need any knowledge to trade ,just do random trades. Based on those 2 outcomes we could then conclude : We don't need any education to trade. We would also find out running a simulation that from 1000 clueless traders we would have let say 10 very successful traders in the long run . We also have very successful traders who use their knowledge and experience and make consistently money , but based on statistics from brokers only 1% of traders(people with some system and knowledge) make consistently money ( if I remember correctly ) . How can successful traders be sure that it is their knowledge and not luck . I personally don't trade too much , but I can say that It is extremely difficult to beat sp500 and if I was taking opposite trades I would be beating sp500.
A third possibility is random trades are profitable with some market conditions and unprofitable with others. Conclusion would be: you just need knowledge of future market conditions to take advantage of random trades.
Nothing is Guaranteed involving trading. When will people stop searching for fixed, set it and forget it, formulas to try to make money in the market? Risk and Reward in the markets, and in life, are generally directly correlated -- it's a Universal law, that cannot be broken, molded and bent. Why are some hunters better than others? Why are some fishermen better than others? Why are some flirters better than others? Why are some traders better, or more successful, than others? Is it because they have a magical formula, or they're blessed with Luck from the Gods, or they understand the overall landscape and approach of things,
I agree that some random trades would would identical to the trades made by knowledgeable traders. We would see some profitable month made by random traders but the more trades random trades would make the less of those profitable month we would see. I wish somebody run a simulation and show the results. I am not smart enough to figure out this kind of backtesting. We might just find out that after running 10mil of random trades we would have a return identical to total stock market minus the spread and commission .
To do this, there is a wide range of options. You can either turn your monitor upside down, or you can stand on your head or hang by your ankles while trading with eyes glued to your right-side-up monitor. Any of these methods should work equally fine. Hopefully someone will soon try one of these equivalent methods and report back on the results. I've just completed a thorough analysis using Ricardo's equivariant tensors on polar manifolds on my liquid nitrogen cooled decicore® AMC server and can confidently report above recommended methods are in fact equivalent. What has occurred to me in the meantime is that the same result should be attainable by simply painting over any data and graphs on your monitor and proceeding to enter orders as you see fit. I shall test out that next. And yet another equivalent occurs to me. Why not simply use a felt tip pen to draw a pretty price-time graph on the monitor and trade accordingly? This method recently worked for repositioning hurricanes, so why not give it a try?
Amazing all the answers ... when the problem lacks critical information: How often is a new trade picked/entered? Can it pick a new trade, while already in a trade? But if it did that, it'd have to close the prior trade? "You would only close the trade if you made at least 1%" What if it buys at (what will be) the all time high, and never reached the 1% profit? You don't provide enough information to generate an adequate response.
You can pick any criteria you like. You can create your own stop losses and sell signals .I don't think it matters as you are simply testing a success of a totally random trading system. Your choice of criteria. You can just randomly pick a stock from sp500 to open the trade, then after 1 week you gonna close the trade regardless of a profit or loss if you like. That sounds like a totally stupid trading system. So doing the exact opposite should generate profit if we back test it on 1mil trades or so . Maybe even after 10,000 trades we should start seeing only losing months
A familiar trader knows how to make money by the opposite trade. At a certain moment before a possible rollback, it goes in the opposite direction from the trend. Naturally, he begins to lose money, at the next rollback he doubles, so as to go to zero, then he loses money again, and so on. Then, at a certain moment, when the trend unfolds or a rollback begins, it gets a snowball - it can take a very large profit. But this is a dangerous game. I have not been able to repeat this in the paper trade.
Many years ago, I experimented and designed profitable systems that used random entries; and stop-and-reverse entries/exits. The edge was in the loss, profit, and/or time stops. I also ran a trading contest and entered a dummy (control) trader that traded based on the first digit of a state lottery pick-three daily draw. It outperformed all traders (several) for three months, and was freakishly profitable (on paper, of course) with low draw downs. That said, It is just as difficult to make a system that is wrong 80% of the time, as it is to make a system that is right 80% of the time. Think about it, they are one and the same. Most losing systems are around 50% (i.e. random), imo. slippage, commissions, and what I call the percentage paradox is why reversing such a barely losing system, will not create a winning system. I call the fact that losing 10% (or any percentage) will require winning more than that initial percentage (or increasing risked capital) to get back even, the percentage paradox.