Pulsar System | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Please suggest a simple & effective Method/System of "Trading System" Forum to newbie 182 replies

The System Behind the System, by Merlin Jeffries 16 replies

200 Ema system - help me build a system ! 19 replies

Knife system, Vegas 4h system related questions. 4 replies

Would you trade this system? New system #2. 10 replies

Attachments: Pulsar System
Exit Attachments

Pulsar System

  • #1
  • First Post: Jan 19, 2008 10:00am Jan 19, 2008 10:00am
I first saw this system over twenty years ago. It was intended for futures trading.
I have recently dusted it off and found it seems to work for forex.

I did a manual backtest on GBP/USD from 12/01/07 through 01/18/08.

There were 25 wins and 13 losses for a 66% win rate.

Risking 2% of account size on each trade would have resulted in an increase of 38% of the account.

A $1000 starting account would have ended with an account of $1380. This includes a 4 pip spread per trade.


RULES
The following terms will be used in describing the rules:

Up Close - A close that is higher than the close of the last candle.

Down Close - A close that is lower than the close of the last candle.

Four Bar Low- A low that is lower that the lows of the past three candles.

Four Bar High - A high that is higher than the highs of the last three candles.

I used 4H bars.

--------------------------------------------------------------
LONG


A four bar low followed by an up close (the up close may be the same bar as the four bar low).


Then a down close that does not take out the low of the four bar low.

Now you begin watching the last two bars.

Enter long when price exceeds the highest of the two previous closes by 1 pip.

Exit on the first profitable close.

Stop loss is the low of the bar previous to entry - 1 pip.

-------------------------------------------------------------

SHORT

A four bar high followed by a down close (the down close may be the same bar as the four bar low).

Then an up close that does not take out the high of the four bar high.

Now you begin watching the last two bars.

Enter short when price falls below the lowest of the two previous closes by 1 pip.

Exit on the first profitable close.

Stop loss is the high of the bar previous to entry + 1 pip.


--------------------------------------------------------------
Stay in a trade until a profitable close or the SL is hit.

If a trade in the opposite direction occurs, take it also. Treat it as an independent trade.


I'm sure this system can be greatly improved on with better stops and a better take profit method.
Attached Image (click to enlarge)
Click to Enlarge Name: gbp.gif Size: 40 KB
  • #2
  • Jan 21, 2008 2:59am Jan 21, 2008 2:59am
hello billbss,

Do you have an EA for this? if not,What you think make an EA?Mabe others pair can be interested too... How long you are using this TS? iŽll not charge nothing, just share

Kind regards,
Marcelo Marques
  • #3
  • Feb 25, 2008 1:23am Feb 25, 2008 1:23am
Quoting billbss

SHORT

A four bar high followed by a down close (the down close may be the same bar as the four bar low).
Hi B,

Looks neat, just noticed though, that your second example on the chart (the short) is only a 3 bar low, not a 4 bar low as your rules say. Just an oversight, or was there a reason for that?

Thanks for sharing,

VBro
  • #4
  • Feb 25, 2008 1:28am Feb 25, 2008 1:28am
Quoting vbro
Hi B,

Looks neat, just noticed though, that your second example on the chart (the short) is only a 3 bar low, not a 4 bar low as your rules say. Just an oversight, or was there a reason for that?

Thanks for sharing,

VBro
Just figured out that the signal is in fact the four bar low, it's just that the fourth bar is also the up close.

Got it, my bad...

VBro
  • #5
  • Feb 26, 2008 3:39pm Feb 26, 2008 3:39pm
Quoting billbss
SHORT

Stop loss is the high of the bar previous to entry + 1 pip.
Hi again Bill,


1.) Would the second trade have been stopped out (see quoted rules above) because the stop would have been put at 9712ish (high of upclose candle)?


2.) Would the following closes be a valid four bar low?

Short:
0:00, 9810
4:00, 9820
8:00, 9830
12:00, 9800


3.) If a 4th bar is both a 4 bar low and a 4 bar high, do you take both trades?


Thanks Bill
  • #6
  • Feb 26, 2008 4:15pm Feb 26, 2008 4:15pm
Quoting vbro
Hi again Bill,


1.) Would the second trade have been stopped out (see quoted rules above) because the stop would have been put at 9712ish (high of upclose candle)?

It was a bear candle. I assumed the high occurred before the entrance occurred, which means it would not have been stopped out.


2.) Would the following closes be a valid four bar low?

Short:
0:00, 9810
4:00, 9820
8:00, 9830
12:00, 9800

Yes.


3.) If a 4th bar is both a 4 bar low and a 4 bar high, do you take both trades?

Yes. Take them as they occur. They won't both occur at the same time.


Thanks Bill
.
  • #7
  • Feb 26, 2008 4:30pm Feb 26, 2008 4:30pm
Hi,

Thanks for the quick reply! A few more...

1.) What server time zone did you use for your test?

2.) How did you handle really small SL's - did you ever change your SL because it was just too small?

3.) Any more testing yet? EA?

4.) You mentioned that you assumed the second trades entry was after the SL price had already come and gone. Does that mean that your results are based on a backtest of the 4 Hr chart only (no smaller timeframe verifications)?

I'm going through your test period right now to get a feel for it and checking smaller time frames to verify (Alpari: GMT +1). I'll let you know if there are any differences that I find.

5.) Are you trading this at present?

Regards,

VBro
  • #8
  • Feb 26, 2008 4:54pm Feb 26, 2008 4:54pm
Quoting vbro
Hi,

Thanks for the quick reply! A few more...

1.) What server time zone did you use for your test?

I used 0GMT - IBFX.
It doesn't matter which one you use.

2.) How did you handle really small SL's - did you ever change your SL because it was just too small?

No

3.) Any more testing yet? EA?

No

4.) You mentioned that you assumed the second trades entry was after the SL price had already come and gone. Does that mean that your results are based on a backtest of the 4 Hr chart only (no smaller timeframe verifications)?

The backtest I did for this thread used only the 4H bars.


I'm going through your test period right now to get a feel for it and checking smaller time frames to verify (Alpari: GMT +1). I'll let you know if there are any differences that I find.

5.) Are you trading this at present?

I trade it off and on. I've traded it for years, mostly on futures.

Regards,

VBro
.
  • #9
  • Feb 26, 2008 5:54pm Feb 26, 2008 5:54pm
Once a signal has been completed, how long does the system wait for price to come back and hit the entry target?
  • #10
  • Feb 26, 2008 6:29pm Feb 26, 2008 6:29pm
I remember this system!!

I also remember that this entry method was used on another system that used oscillators that represented different time frames.( I can't remember that system though).

I think we could build on this entry technique to make it even more accurate!

Thanks for bringing this back!!!

Everything old is new again!
  • #11
  • Feb 26, 2008 6:52pm Feb 26, 2008 6:52pm
Quoting vbro
Once a signal has been completed, how long does the system wait for price to come back and hit the entry target?
I'm not sure I understand what you're asking.
  • #12
  • Feb 26, 2008 7:08pm Feb 26, 2008 7:08pm
Quoting billbss
I'm not sure I understand what you're asking.
Yeah, I didn't make that as clear as I might have.

A signal is validated, go short @ 9800.
Then price ranges 9810 to 9850 for 8 hours (no entry).
Then price runs up to 9910 over the next 8 hours (no entry).
Then price ambles down to 9840 to 9820 over the next 8 hours (no entry).
Then back up to 9880 to 9910 for the next 8 hours (no entry).
Then back down to 9830 over the next 8 hours (no entry).
Finally, 40 hours later, price crashes down to 9700.

In the time between the original signal and when price triggered the signal's entry, there were other fresh signals and entries. But now, finally, price has triggered the original signal. Would the trader still have a limit order in place for that somewhat stale signal, or would they have discarded that signal as too old?

Thanks.
  • #13
  • Feb 26, 2008 8:14pm Feb 26, 2008 8:14pm
Quoting vbro
Yeah, I didn't make that as clear as I might have.

A signal is validated, go short @ 9800.
Then price ranges 9810 to 9850 for 8 hours (no entry).
Then price runs up to 9910 over the next 8 hours (no entry).
Then price ambles down to 9840 to 9820 over the next 8 hours (no entry).
Then back up to 9880 to 9910 for the next 8 hours (no entry).
Then back down to 9830 over the next 8 hours (no entry).
Finally, 40 hours later, price crashes down to 9700.

In the time between the original signal and when price triggered the signal's entry, there were other fresh signals and entries. But now, finally, price has triggered the original signal. Would the trader still have a limit order in place for that somewhat stale signal, or would they have discarded that signal as too old?

Thanks.
I still don't get it, but maybe I can answer this way:

Take every signal. If you have a short set up and a long set up intervenes, take it. There may be times you are both long and short. The length of time it takes isn't an issue.
  • #14
  • Feb 29, 2008 8:45pm Feb 29, 2008 8:45pm
Hi Bill,

In the picture below, would entry price be the blue or red line?
Attached Image
  • #15
  • Mar 1, 2008 5:14pm Mar 1, 2008 5:14pm
Quoting vbro
Hi Bill,

In the picture below, would entry price be the blue or red line?
After the down close you enter a pip above the highest close of the previous 2 bars...I believe if there is no entry generated you look to enter at the highest close of the most recent 2 bars.
  • #16
  • Mar 1, 2008 6:23pm Mar 1, 2008 6:23pm
Hi,

Would this system not fail in the long run due to the lack of a defined reward againt risk?

Regards,
lucidlamp
  • #17
  • Mar 1, 2008 9:18pm Mar 1, 2008 9:18pm
Quoting lucidlamp
Hi,

Would this system not fail in the long run due to the lack of a defined reward againt risk?

Regards,
lucidlamp
Where did you learn that?

It will fail if it does not make a profit.
  • #18
  • Mar 1, 2008 11:01pm Mar 1, 2008 11:01pm
Quoting RickW00716
...I believe if there is no entry generated you look to enter at the highest close of the most recent 2 bars.
Thanks, Rick. Bill, I'd love a thumbs up from you about this if this is right...

(I see my picture is wrong. The blue line should have been one candle earlier, then entered on the following candle, when the two candle high is breached... sorry for any confusion this might cause).
  • #19
  • Mar 2, 2008 7:43am Mar 2, 2008 7:43am
Quoting billbss
Where did you learn that?
Hi billbss,

If 30 pips are risked in any given position and the position, which risked 30 pips, is closed for a 15 pip profit a loss is incurred, no? Does not every trade taken have to have a reward target (TP) in mind that will outweigh the risk (SL) taken? Eventually the 66% correctiveness, stated in the first post, will make the system loss in the long run if pips risked outweigh pips won. So if 30 pips are risked on every trade and every profit is just 15 pips the system has won 15 pips 25 times for a total of +375 pips, conversely 13 losses at 30 pips risked would be -390 pips. Since the reward pips (TP) is not defined there is no way to measure the systems profitability? Just does not make sense how this system can be considered profitable as it is now designed as it can not be tested, well you can foward test it..., which is blind trading (flip of a coin)? Even backtesting can give deceptive results without stringent parameters?

I am just making an observation, about systems in general. I am not singaling out your this system. I looked back and the set-up seems very viable, as it produces accurate short term forecasts, especially if a more solid method of managing money, amount risked against reward, SL & TP is used. As it is now, just taking profit at the close of the first profitable bars close, seems like the system is not living up to its full potential. Of course this is if the system is traded objectively to the tee...

Maybe I'm just being too much of a pessimist?

Regards,
lucidlamp
  • #20
  • Mar 2, 2008 11:09am Mar 2, 2008 11:09am
Quoting lucidlamp
Hi billbss,

If 30 pips are risked in any given position and the position, which risked 30 pips, is closed for a 15 pip profit a loss is incurred, no? Does not every trade taken have to have a reward target (TP) in mind that will outweigh the risk (SL) taken? Eventually the 66% correctiveness, stated in the first post, will make the system loss in the long run if pips risked outweigh pips won. So if 30 pips are risked on every trade and every profit is just 15 pips the system has won 15 pips 25 times for a total of +375 pips, conversely 13 losses at 30 pips risked would be -390 pips. Since the reward pips (TP) is not defined there is no way to measure the systems profitability? Just does not make sense how this system can be considered profitable as it is now designed as it can not be tested, well you can foward test it..., which is blind trading (flip of a coin)? Even backtesting can give deceptive results without stringent parameters?

I am just making an observation, about systems in general. I am not singaling out your this system. I looked back and the set-up seems very viable, as it produces accurate short term forecasts, especially if a more solid method of managing money, amount risked against reward, SL & TP is used. As it is now, just taking profit at the close of the first profitable bars close, seems like the system is not living up to its full potential. Of course this is if the system is traded objectively to the tee...

Maybe I'm just being too much of a pessimist?

Regards,
lucidlamp
You have some basic misconceptions of calculating profit expectancy. The misunderstandings that you have are fairly common among traders.

You do not have to have a greater reward factor than risk factor to be profitable.

The formula for profit expectancy is:

(W%*(average reward/average risk)) - L%

If the win% is high enough you can have a Reward/Risk ratio well below 1 and still be profitable.

Conversely, you can have a low win% and be profitable if the R/R is high enough.

You cannot single out either factor and say that it must be anything.

The reward and risk on any single trade does not have to be clearly defined in order to be profitable. You only need to know that the average R/R coupled with the average win% gives a positive profit expectation. You never can know that on any system. You can only rely on past statistics.

This system can be backtested. The parameters are no more stringent that on any other system.

I have not done a lengthy backtest on it because I've traded it long enough to know that it has worked in the past and has done so for a long time.

I wouldn't take anyone else's word for it. If someone wants to trade it, they need to do their own testing.
0 traders viewing now