Optimal Profit-taking

[trendie]

Once a trade has been initiated, there is a stop-loss, which is a quantifiable value.
Fine and dandy.

Then there is the potential for profit.
You may have a profit target, which might be a reversal pattern, or a Sup/Res level.
However, there is also the possibility that the target may not be reached. Under these circumstances, you need to have a mechanism to account for targets not being reached. This maybe in the form of moving stop-loss to B/E.
It may also be in the form of a trailing stop.

I think Newtron Bomb places multi-lots, and then takes profits quickly to break-even, and then allow the remaining amount to become a home-run.

Phil Nel, who has a highly-regarded 5-min system, seem to go for 10-20 pips.

What is the optimal profit-taking scenario?

Multi-lots? Home-runs? Trailing stops? Pre-defined "target or bust"?

I suspect the mechanism is determined by time-frame, so lets say its a day-trade, with anticipated target of a quarter of daily range, trades taken between 7-am and 9-pm.

 

[fxmonk]

There is no optimal profit traking scenario. If you set small targets you will win often and if you set large targets then you will lose often but win big to make up for it. (These two cases are analogous to negative and positive gamma in the options/macro world)In both cases your expectancy is 0 if you don't have an edge. If your edge is based on capturing the large moves in currencies, you will want to use a profit limit that is several times your stop size. If your edge is based on mean reversion strategies, then you will want to take profits very quickly. Just keep in mind that optimization has little value when the market is ever changing and nearly random.

It becomes hard to distinguish between trading off of actual P/L and mark-to-market P/L, meaning that one would get stopped out only to see the move continue in the right direction, and then say, "Darn, I should've held on longer" when in reality it wouldn't matter over a large enough sample of trades. The market is just too random for these "rules of thumb" to work.

I hope this made sense

 

[trendie]

Thanks for reply.

I have been thinking about those who go for a risk/reward of about 1:1, but high hit-rate. But I wonder if the occassional home-run is what makes their system really work.
For example, I want to do some work on what would happen if I remove the top 5% of trades and see how that affects overall returns.

For example, in a run of 100 trades; if I remove the best 5 trades, how much they affect overall profitability. I recall reading Curtis Faiths "Way of the Turtle", and how many of their trades are small losers or average trades, but they rely on catching the occassional big trade to give them their profits. That means taking every signal without fail.

I am beginning to think that small bite trades with targets can help to reduce whippy equity-curves, but at the expense of having to relying on outliers for true growth.

Again, as per my previous thread, the assumption is of a positive edge already established.

 

[Chorlton]

Thanks for reply.

I have been thinking about those who go for a risk/reward of about 1:1, but high hit-rate. But I wonder if the occassional home-run is what makes their system really work.
For example, I want to do some work on what would happen if I remove the top 5% of trades and see how that affects overall returns.

For example, in a run of 100 trades; if I remove the best 5 trades, how much they affect overall profitability. I recall reading Curtis Faiths "Way of the Turtle", and how many of their trades are small losers or average trades, but they rely on catching the occassional big trade to give them their profits. That means taking every signal without fail.

I am beginning to think that small bite trades with targets can help to reduce whippy equity-curves, but at the expense of having to relying on outliers for true growth.

Again, as per my previous thread, the assumption is of a positive edge already established.

Trendie,

Out of interest, would your own "positive edge" still work if you changed your strategy from (i) low win-rate / high R:R to (ii) high win-rate / low R:R, or vice versa?

I would suggest that the two approaches are completely different. For example, with the former one needs to take every trade (as you stated) whereas one can afford to miss a few with the latter.

By using Targets you will definately smooth out your Equity Curve but how then are you going to achieve the very profitable outliners, if all trades are exited at a low R:R (say 1:1)??

 

[mrsoul]

Once a trade has been initiated, there is a stop-loss, which is a quantifiable value.
Fine and dandy.

Then there is the potential for profit.
You may have a profit target, which might be a reversal pattern, or a Sup/Res level.
However, there is also the possibility that the target may not be reached. Under these circumstances, you need to have a mechanism to account for targets not being reached. This maybe in the form of moving stop-loss to B/E.
It may also be in the form of a trailing stop.

I think Newtron Bomb places multi-lots, and then takes profits quickly to break-even, and then allow the remaining amount to become a home-run.

Phil Nel, who has a highly-regarded 5-min system, seem to go for 10-20 pips.

What is the optimal profit-taking scenario?

Multi-lots? Home-runs? Trailing stops? Pre-defined "target or bust"?

I suspect the mechanism is determined by time-frame, so lets say its a day-trade, with anticipated target of a quarter of daily range, trades taken between 7-am and 9-pm.

I've experimented with many ways to take profits.
The way I am most comfortable with is to take half my profits when price is half way to my target, then set a break even stop and try to get close to my target on the other half.

 

[h_pfinance]

One of the profit taking methods I have found to be effective is the "SMA" technique. Suppose I am long in a trade and lets say 50 bars have gone since I entered the trade. Then from that bar on, I calculate the 50 SMA of the Lows and I am in the trade till 2 bars close below the 50 SMA of Lows IN SUCCESSION.

If the trend is strong .. I will be in the trade till the trend lasts. I have found this technique to be effective. One may choose 50,60 or even 100 SMA. Combined with this technique, a high profit target can also be used to exit the position once 50-60 pip profit target is reached.

Also, I move my stop Loss up once I hit 10 pips to breakeven and then after every 10 pips, I move the stop loss up by 1 pip. So if the trade dies down fast, at least I breakeven or make a little sometimes.

Same but opposite is true for short entries. I calculate the SMA of Highs and if 2 or more bars close Above the SMA in succession, then I exit. I have found this to be effective for 25, 50, 100 Tick bars.


hp

 

[trendie]

Phil Nels two strategies that seem to be popular.
(this post for anyone wanting to know who Phil Nel is)

5Min Intraday System - Forex Factory

4 Hour Strategy (MACD) - Forex Factory

 

[Rossini]

(this post for anyone wanting to know who Phil Nel is)

lol. Cheers