many Forexrebatekingdividual traders in the design of their own cashback forex system, will choose the cashbackinforex following strategy to build a trend following system there are usually two types of methods: one Forex rebate king the price breakout, the other is the linear crossover use price breakout to do trend following friends, usually looking for some typical price, such as the stage of the high forexrebateindonesia low, or the number of recent trading days of the highest and lowest point, etc. (for example, the turtle rule of 20-day high and low) once the price of such a typical price breakout to follow the trade. For example, the turtle rule of 20-day high and low points) once the price of such typical price breakthrough will open a position to follow the transaction and use the linear crossover to do trend following, mainly using the average price crossover of different time periods, such as double average crossover after closing to determine the opening of positions to follow the price of a single average crossover, MA (1) and a single average MA (n) crossover, which is also a deformation of the double average crossover and Or three SMA crossover filter or use MACD to do trend signals essentially MACD is also a deformation of the double SMA crossover and if you use the zero axis to filter the crossover signal, it is actually a deformation of the three SMA crossover filter Many friends design their own trend trading system after backtesting, the results are often unexpected Clearly there is a trend in the market, but why the backtest data does not show significant gains? The usual solution is to optimize the system is the first question is the parameter is not the parameter everyone uses, so it is not working? Try and try to find a variety of parameter combinations of backtesting results are similar in a period of excellent performance in another period of performance is very bad in a variety of good performance in a variety of parameters in another variety and few returns so again to optimize the system is not to add some restrictive conditions will be better? For example, how much floating loss after mandatory stop loss, or how much floating win after stopping out of the field or part of the profit but the effect is still very difficult to be satisfied with some profitable transactions were stopped out of some of the transactions that can obtain excess returns are also limited by the profit limit again optimization will basically involve the issue of position, such as continuous profit, or net value growth after the number of positions to reduce the opening position after continuous losses to increase the size of In this way, a trend-following system slowly embarked on a data-fitting routine If you are now facing such a predicament, I would like to remind you of the misconception that you are not walking into a trend-following This misconception is the trend signal as a trading signal We can consider that in a certain time period, if the price of a typical high breakout, then this is an upward trend signal Similarly We can also think that if a combination of averages crosses at a certain time period, then this is an uptrend signal. When we believe in the power of the trend and track the trend, we are essentially trading the process of the trend, not the point at which the trend forms and ends. In other words, the golden cross of two averages is the point at which the uptrend forms, and the dead cross of two averages is the point at which the uptrend ends. Trading, you can conclude that: in these two points in time to do more than one single and only do more than one single Therefore, the trend signal is a prerequisite or necessary condition for trading signals But, is it possible to do more than one single at any time within these two points in time? The answer is obviously not this relates to a profit and loss ratio of the odds problem we can all understand, in the late trend to do more than the trend is obviously much more risky than the early do more so the question arises, the position of the formation of the golden cross, obviously the trend is formed at the very beginning, then whether you can directly use the trend signal to do trading signals? Before answering this question, we first deconstruct the premise that trend tracking can be profitable If you want to be profitable in a trend tracking transaction, you need two prerequisites: a large trend running space, and a long trend running time we all have experience, if a trend overall running time, but not much space, which is likely to be just a larger cycle of adjustment, the trend trade ultimately pocketed The other situation is that the trend formed after running the space amplitude is very large, but the trend duration is extremely limited, the fall back more quickly, the market speedy battle this case tracking trend trading is likely to end up losing money out of the game or even possible back and forth face The most ideal situation is that the slow bull and negative market trend low-key formation, after a long time, out of a wide range of volatility, and eventually also low-key However, such an opportunity is very rare (the following chart you feel under) Unfortunately, the trend signal can provide us with information about the direction of the trend, but can not tell us the trend in the direction of the continuation of time and space or even, if we are more careful to describe the trend signal can provide us with information only the possibility of the direction of the trend in the case of sufficient samples, about 1/3 of the trend Some of these signals can be identified and filtered out by your trading experience. About 1/3 of the signals, although correctly pointing in the direction of the trend, are either limited in space or short in duration, and in general barely break even. However, it is extremely rare to be able to give a 3200 point swing like the above chart and eat 80% of the entire swing. It is because of this 1/3 of the valid signals that some trend traders believe that a 3:1 profit/loss ratio is all that is needed. Therefore, they prefer to use trend signals as trading signals. If they track every trend signal, the win rate is basically locked at 30%, and it is difficult to control whether they can get an average 3:1 profit/loss ratio in these 30% trades. Damage, and not through the history of backtesting control of even if you maintain a 3:1 profit-loss ratio, and fortunately did not encounter the damage of the Black Swan, but the problem is that this profit-loss ratio is only enough to preserve capital, where the profits come from? There will be friends cited the turtles way of trading turtle law is great, but if you read carefully, you will find that the author clearly pointed out that if the funds are limited, it will be difficult to replicate this success if you only focus on a market, it is likely to face 70% of the time unprofitable, and even wait for the trend to come account is hanging and if you focus on multiple unrelated markets at the same time, the huge capital requirements and become an obstacle which is why some This is why some trading stock trend system backtest data is great, the actual trading on the blind reason backtest is the assumption that there is enough money to track every profit point and in the real market, a few shots out, there is only a full position dry eyes Therefore, in the case of limited funds, the design of the trend tracking system, try not to simply brutal trend signal as a trading signal in general, the mechanized trend In general, mechanized trend tracking systems use trend signals directly as trading signals for ease of execution, and will be mandatory for the execution of each trading signal This can avoid missing the effective market, but in the long run, only to obtain β gains If trend traders want to obtain α gains, it is necessary to use some small strategies to improve the win rate or profit-loss ratio However, the problem is that these small strategies are very difficult to This is why I prefer manual trading