Daily95pips

Quoting bullrock

Disliked

I have made some statistical analysis. Here are the HARD FACTS. Probabilities or odds if you wish. You remeber? I always stress.
Now here is the problem

Out of 2073 days analysed of the GU.
The probability that it will reach 15 pips from previous high is 831 to 199
The probability that it will reach 30 pips from previous high is 641 to 389
The probability that it will reach 50 pips from previous high is 439 to 591

See the problem??? It is 57% probable that it will !!NOT!! reach the 50 pips from top... Now we also have an extra 5 pips.

DOOMED to FAILURE if you do not push the SL to BE when the second TP has been hit. It is even better if there is a fourth unit to increase the odds... The mas rise from previous high during this analysed period was 359 pips in a single day. Are you going to leave that on table. Nope. 4th unit is the answer.

Regards
Bullrock or "Billrock"

Ignored

Quoting jones247

Disliked

Based upon my visual observation of the daily candles over the last 3 months, only 1 out of every 5 - 8 trades results in a loss under the following conditions: (1) SL: 100; (2) TP: 30 - 50 (depending on currency pair); (3) No trailing stop; (4) allow for swing trading (1 or more days) before TP is reached; (5) enter long at prior day's high + 5 pips & enter short at prior day's low - 5 pips.

I've observed that about 60% - 67% of the daily candles have trended in accordance with the prior day close. In other words, if the prior day's candle closed higher than its opening, then the next day's candle will have a 60% - 67% probability for also closing higher than its opening, which was the closing price of the prior day.

Simply taking advantage of this pattern can easily yield four wining trades out of every five trades, or better. Based on standard lot sizes, that's a profit of about $800 - $1,000 for every 5 trades - per day or per week.

I have been testing the following currency pairs: GBPJPY, USDJPY, EURJPY, USDCHF, & GBPUSD.

Please let me know if your backtesting or forward testing supports this assertion...


Thanks,
Walt

Ignored

Quoting mikelath

Disliked

Wow ! We have some quality contribution from bullrock and walt (jones247).
THANKS GUYS

Bullrock, please excuse my ignorance. I assume GU refers to GBPUSD ?
If your stats pertain to cable only, then we should not rule out the probabilities for other pairs. This strategy is obviously more suited to more volatile pairs that can provide big price moves. Take EURJPY for example.
When first concieving the idea of D95P I was more or less of the same view as you in that I was seeing so many more pips in the trades after I'd taken my 20pip or 30pip profit target. Thus I thought to have 3 targets to capture more pips. Your idea of a 4th TP with a trailing stop is very good.
The only reasons that I haven't yet adopted it myself is that firstly my broker Oanda doesn't provide trailing stops and secondly a 4th order will increase the total amount at risk, and from a money management perspective it may be more of a challenge...it also may not be...I just haven't explored it yet.
Wal, your observations are invaluable. I'll admit that I've been purposely holding back from adjusting the system just to firstly see which of the pairs that I'm testing perfom well with this strategy. So far EURJPY and EURUSD are going well, though it's still early days.
Adjusting the stops to 100pips makes sense if it increases the probability of more winning trades. However, I am very cautious of risk/reward ratios, so that's why I've tried to keep it at approx.1:1. It seems that 30pips SL is too tight for pairs like GBPJPY, though my initial thought was that when pairs like this move they can virtually sky rocket or plummet in a short timeframe, breaking thru entry and profit targets without practically any retracements.This is where ANY stop loss would be superfluous, though of course still necessary.
As we've seen though, GBPYJPY HAS been retracing and stopping out. So yes, eventually we'll need to make adjustments for pairs like this.

Bullrock and Wal, myself and all those attending this thread will greatly appreciate if you include updated results of your approaches to this thread.
I've tried to be regular (more for my own clarity) in updating my results, and if you guys choose to be more committed to tweaking this strategy, then our combined efforts will certainly prove fruitful.

rgs
mike

Ignored

Quoting bullrock

Disliked

We could use the ATR as a basis for calculations. I have to go now. Seeya tomorrow.

Ignored

Quoting jones247

Disliked

No problem Mikelath, I try to start posting my results as of next week.

thanks,
Walt

Ignored

Quoting bullrock

Disliked

No offence. PLEASE PAY ATTENTION TO THIS POST. There is a lot of information.

1. You're talking about the probability of the previous days close vs the todays close. I am talking about the size of the difference between previous days high and todays high and the probability of its size. That is a different context and calls for a different strategy

2. I've done precise calculations and it is not a visual observation. I don't trust my eyes.

3. You've analysed 3 months I have analysed 8 years. IMHO 3 months is not enough. You know the reason? One business cycle in the banking industry is 3 months (the larger traders of the currencies) So you need to analyse at least 1 year or even better to analyse 4-5 years which is 1 economic cycle. I wanted to analyse 2 economic cycles. That is 5-10 years.

4. I never trust "visual observations". Needs to be backed up by calculations.

So here they come:
I have analysed long and short. So if prev day up this day up probability. And if prev day down this day down probability. Your observation does not hold up in case of USDJPY in the last 3 months hold up for the 8 years and USDCHF does not hold up at all!!!

Notation:
L+ -> if prev day close higher today close higher
L- -> if prev day close higher today close lower (pattern failure)
S+ -> if prev day close lower today close lower
S- -> if prev day close lower today close higher (pattern failure)

GBPUSD
Total: (L+) 51.32% (L-) 48.68% (S+) 46.63% (S-) 53.37%
3 months (L+) 57.14% (L-) 42.86% (S+) 55.56% (S-) 44.44%

GBPJPY
Total (L+) 53.53% 46.47% (S+) 47.83% (S-) 52.17%
3 months (L+) 50.77% (L-) 49.23% (S+) 42.31% (S-) 57.69%

EURJPY
Total (L+) 54.68% (L-) 45.32% (S+) 46.52% (S-) 53.48%
3 months (L+) 56.72% (L-) 43.28% (S+) 44.00% (S-) 56.00%

USDJPY
Total (L+) 50.42% (L-) 49.58% (S+) 46.60% (S-) 53.40%
3 months (L+) 47.06% (L-) 52.94% (S+) 35.09% (S-) 64.91%

USDCHF
Total (L+) 46.93% (L-) 53.07% (S+) 45.54% (S-) 54.46%
3 months (L+) 47.06% (L-) 52.94% (S+) 40.98% (S-) 59.02%


Regards
Bullrock

Ignored

Quoting forexdummy

Disliked

Hi Mr. Mikelath

Thanks for sharing D95 Pips system. I tried the system once using my virtual account and it was working fine. However, I came up with a serious problem. I humbly hope you (and also anyone in the forum) could help me solve my problem.

I am trading GBP/USD pair at the moment. The problem that I find is, when the movement is going downwards and touching BUY point (prevoius day’s High+5) I immediately open BUYING position at the market rate. Later, I find out that it is too risky to do that. What if the movement really stroke down to the SELL point and continued going down ?

The same problem also occurs when the movement is going upwards and touching SELL point. I don’t know whether the movement ends around SELL area, or it’s still striking upwards.



How do I know that a movement really ends around BUY / SELL area and reverses its trend when it reaches BUY/SELL area ? when is about time to open a position ? If I could get this answer correctly, I can make a combination that will increase the probability of success up to 85 - 90%. I hope we all can enjoy the fruit of the system

Ignored

Quoting bullrock

Disliked

No offence. PLEASE PAY ATTENTION TO THIS POST. There is a lot of information.

1. You're talking about the probability of the previous days close vs the todays close. I am talking about the size of the difference between previous days high and todays high and the probability of its size. That is a different context and calls for a different strategy

2. I've done precise calculations and it is not a visual observation. I don't trust my eyes.

3. You've analysed 3 months I have analysed 8 years. IMHO 3 months is not enough. You know the reason? One business cycle in the banking industry is 3 months (the larger traders of the currencies) So you need to analyse at least 1 year or even better to analyse 4-5 years which is 1 economic cycle. I wanted to analyse 2 economic cycles. That is 5-10 years.

4. I never trust "visual observations". Needs to be backed up by calculations.

So here they come:
I have analysed long and short. So if prev day up this day up probability. And if prev day down this day down probability. Your observation does not hold up in case of USDJPY in the last 3 months hold up for the 8 years and USDCHF does not hold up at all!!!

Notation:
L+ -> if prev day close higher today close higher
L- -> if prev day close higher today close lower (pattern failure)
S+ -> if prev day close lower today close lower
S- -> if prev day close lower today close higher (pattern failure)

GBPUSD
Total: (L+) 51.32% (L-) 48.68% (S+) 46.63% (S-) 53.37%
3 months (L+) 57.14% (L-) 42.86% (S+) 55.56% (S-) 44.44%

GBPJPY
Total (L+) 53.53% 46.47% (S+) 47.83% (S-) 52.17%
3 months (L+) 50.77% (L-) 49.23% (S+) 42.31% (S-) 57.69%

EURJPY
Total (L+) 54.68% (L-) 45.32% (S+) 46.52% (S-) 53.48%
3 months (L+) 56.72% (L-) 43.28% (S+) 44.00% (S-) 56.00%

USDJPY
Total (L+) 50.42% (L-) 49.58% (S+) 46.60% (S-) 53.40%
3 months (L+) 47.06% (L-) 52.94% (S+) 35.09% (S-) 64.91%

USDCHF
Total (L+) 46.93% (L-) 53.07% (S+) 45.54% (S-) 54.46%
3 months (L+) 47.06% (L-) 52.94% (S+) 40.98% (S-) 59.02%


Regards
Bullrock

Ignored

Quoting mikelath

Disliked

Hi forexdummy.

Firstly, NO system can demand the market to conform to its parameters. In other words, we can never predict with 100% accuracy which way price will go.
The best we can do is devise a system based on probabilites provided by historical and forward tested data.
The D95P strategy is based on the probability of price continuing to our profit targets once it has moved beyond 5pips above or below the previous days high or low.
Some have mentioned that it has been historically shown that the market tends to 'range' (stay within the previous days high and low...or if it does breakout, retrace again within the previous days H and L) more than 'trend'(break above or below the previous days high and low and continue to trend). So according to this view, the probabilities are in favour of trading within the range of the previous day. You'll notice that discussion of BOTH approaches are contained in this thread.

Regarding your following statements:

'The problem that I find is, when the movement is going downwards and touching BUY point (prevoius day’s High+5) I immediately open BUYING position at the market rate.'
If price is going down we would expect the SELL orders to trigger, NOT the buy orders.
Please clarify ?

'Later, I find out that it is too risky to do that'
The risk is the total amount of our stop losses.

'What if the movement really stroke down to the SELL point and continued going down?'
If price triggers my SELL orders and continues to go down to my profit targets I will celebrate.

'The same problem also occurs when the movement is going upwards and touching SELL point.'
You seem to be describing the 'ranging' strategy that bksujal has shared...?

Is your concern that what if BOTH buy and sell orders are triggered ?
I'm sorry...I'm not really sure what your question is?

If you haven't done so already, I encourage you to read thru all the posts...I know that as the thread progresses that increases the amount of reading...however, it can be worthwhile.

'I can make a combination that will increase the probability of success up to 85 - 90%.'
Hey, if you can do this then I'm with you all the way !

mike

Ignored

Quoting forexdummy

Disliked

Thank you Mr. Mike for your response
Sorry to make you confused about my question. I try to make it simpler. What I noticed when conducting a virtual trading with this system was :
1) The price occasionally, though not very often, moved downwards and touched the BUY POINT. Then it didn't break through the buy point. It was just reversing and going up. I have attached a screenshot from MT chart, hopefully it could help give you a picture of my question.



I found out during trading virtually, that if we can break this problem and solve it, we can increase our probability of getting net greens (profits).

Ignored

Quoting jones247

Disliked

Hi Bullrock,

I appreciate your technical analysis. Your data is very precise & detailed. However, you are not including a few variables that would change the percentage allocation for wins and losses. To use your terms (L+, L-, S+, S-), it's shortsighted to assume that all L- & S- equate to a loss. That is why a 100 pip SL is so important. With a SL-100 you can have a L- or a S-, but it can reverse to a L+ or a S+ within a few days as long as the SL was not reached. Another variable to consider is the fact that is some instances, although it may be a L- or a S-, the day's high or day's low was good enough to reach the TP before retracing to the L- or S-. Therefore, although it was a L- or S-, it was still a profitable day because it trended strong enough to reach the TP before reverting back to a L- or S-.

Once you factor in these other variables, I still believe that most of your trades in the trend will win 67% - 80% of the time with a TP-50 & SL-100. I will try to provide emperical backtesting data to confirm my assertion.

Thanks,
Walt

Ignored

Quoting jones247

Disliked

Hi Bullrock,

I appreciate your technical analysis. Your data is very precise & detailed. However, you are not including a few variables that would change the percentage allocation for wins and losses. To use your terms (L+, L-, S+, S-), it's shortsighted to assume that all L- & S- equate to a loss. That is why a 100 pip SL is so important. With a SL-100 you can have a L- or a S-, but it can reverse to a L+ or a S+ within a few days as long as the SL was not reached. Another variable to consider is the fact that is some instances, although it may be a L- or a S-, the day's high or day's low was good enough to reach the TP before retracing to the L- or S-. Therefore, although it was a L- or S-, it was still a profitable day because it trended strong enough to reach the TP before reverting back to a L- or S-.

Once you factor in these other variables, I still believe that most of your trades in the trend will win 67% - 80% of the time with a TP-50 & SL-100. I will try to provide emperical backtesting data to confirm my assertion.

Thanks,
Walt

Ignored

Quoting vanupips

Disliked

Hi Bullrock,
According to your data and analyse, wouldn't the simplest way of trading be just to buy on up-days immediatly at the close of the candle(day), with T/P 20 pips and S/l 20 pips ,and sell on down days at the close with 20 pips T/P and S/L.
We would leave a lot of pips on the table some days, but that is not the goal, the most important is to score the more pips on a regular basis.
You most probably did allready some research on this idea?
regards,
vanupips

Ignored

Quoting bullrock

Disliked

Ok I will recalculate and consider your views.

Ignored