It is necessary for trend-following traders to accept that trends sometimes don't work - in fact, trends fail more often than sustain. Prices do have a tendency to revert to the mean.
One of the saddest situations for a trend-following trader is a ‘whipsaw series’. The trader sees a trend and takes a position.
The price “whipsaws” the other way and triggers the stop-loss. This may happen many times in succession, and as the losses build, it leads to questioning of the system.
It is necessary for trend-following traders to accept that trends sometimes don't work — in fact, trends fail more often than those sustain. Prices do have a tendency to revert to the mean.
One may see five or six trend failures in succession before a trend sustains and takes off. Trend-following pays only because when trends do sustain they earn vast profits.
Since there is no way to tell which trend will work, trend-following systems demand action on every valid signal. The definition of ‘valid signal’ will, however, differ from system to system.
Traders can set as many initial filters as they want to define validity. But, once a valid signal is defined and it pops up, it must be acted upon.
Given persistent whipsaws, the trader should also check stop-loss methods. Stop losses must be set close enough to the price to suffer losses in comfort. But, a stop must also be wide enough to ensure that small random price changes don't trigger it.
When a stop is set too close, it leads to unnecessary whipsaws. One sign of a stop loss too close to entry price is that the trend sustains in the correct direction after triggering the stop loss and closing the trade out.
If you're confident the stop is wide enough and the system has enough filters to process valid signals, keep taking the trades as they come. Don't allow your emotions to affect the situation.
Very few traders use both trend-following and mean-reversal systems. If a robust trend-following system is persistently throwing up trade-failures, it is not a trending market. In that case, switching to a mean-reversal system may work.
These trades go in opposite directions. A trend-following system will trade a breakout on the assumption that the breakout establishes a new trend. A mean-reversion system will assume breakout failure and resumption of range trading.
It should be possible to marry the two systems. For example, let's say a stock breaks out from a trading range of 90-99 and a trend-follower takes a long position at 100 with a stop loss at 98.
The stop is triggered. He may then go short, with a stop at 100, hoping for range trading back to 90 levels. Or, he could wait to confirm trend failure and short, if the price falls to 97 with a stop loss at 100.
Employing such a hybrid strategy means trading frequently with high brokerages. It is also possible to lose money on whipsaws in different directions. Filters must be set up to control losses, if such a hybrid system is used, but it should be possible in theory.