Daily Hi Lo

[SanMiguel]

So if you’re going to trade this hilo idea then maybe best to only set up pending trades away from support or resistance. Unless price is currently somewhere inbetween in which case you would set up orders in both directions.

I don’t know if you can test that in a strategy tester.

This is the big thing with breakout strats, you have to have some pattern to it. Quite simply going short at the zone where the last 2 times buyers have come is just a recipe for disaster. I agree that the more an area is hit, the more likely to breakout of it but I’d be looking to get in way before the breakout so I could have my stops at breakeven.

I haven’t done enough analysis but any breakout strateies that do work are very likely to be quantitatively based on trading volume and ATR. There is a reason why the 24 year old “pro” over at F4N suceeded in trading breakouts when GJ had a daily range of over 400 pips and why his method failed substantially in the last 18 months (GJ range much much smaller now).

[dobro]

Ever heard of “Sell in May and then go away”?? Summer volume and action historically has always been lower and given the economic uncertainties of 2010 many strategies may very well underperform until fall. My .02 cts worth. d.

[Synergistik]

Ever heard of “Sell in May and then go away”?? Summer volume and action historically has always been lower and given the economic uncertainties of 2010 many strategies may very well underperform until fall. My .02 cts worth. d.

I have now :o

[finally]

Holy Grail
Sorry I couldn’t resist saying that. The only thing I can’t resist is temptation.

This is not a highly polished well developed trade strategy. Just something I’ve been looking at that I thought I would share and get your thoughts on.

Take a look at a daily chart. I’ll use GBP/USD in my example.
Do you see it?
Kinda sticks out like a sore thumb doesn’t it?
Did you ever notice how the candles are all the same length and lined up in a horizontal flat line across your screen like a bunch of dominoes?
No? You didn’t notice that?
Well that’s because they don’t!

So what does price do? It moves! up and down. Even when it’s in a narrow range it bounces up and down inside that range. (good for boll bounces but that’s a different thread).

So most of the time a candle will be offset from the previous candle. It will extend either above the previous high or below the previous low or sometimes both.

On a daily candle even a small extension can be worth enough pips to make a trade worthwhile. But not on time frames less than daily. So that’s why this is going to be a ‘daily’ high / low system.

after looking at the last 10 years of daily candles…
96.32% of them will extend either above the high or below the low, or both, of the previous daily candle.

The trade strategy is obvious. Place a pending buy stop at yesterday’s high and a pending sell stop at yesterdays low. You will get at least one order triggered 96 days out of a 100. Of those 96 days, 10% of them will get both orders triggered.

So Of those that go above or below the previous daily High or Low, price will go at least 20 pips 78.66% of the time. This will give you an average of 1573 pips per 100 trades, (78.66 times 20 = 1573). But this is not the optimum.

The optimum Take Profit is actually somewhere in the range of 45 to 75 pips. This gives you in the neighborhood of 2377 to 2489 pips per 100 trades.
Specifically a take profit of 55 or 60 will each yeild 2489 pips per 100 trades.
This only takes into account wins, not losses.
When taking losses into account the optimum TP may be entirely different. (caveat emptor).

That may not seem like a lot, only 24.89 pips per day but with compounding it can add up in a hurry.
And also there will be days when price moves a lot farther than 20 pips so maybe some kind of trailing stop would be worth using?

Notice I haven’t said much about losses or Stop Loss distance. That’s because I haven’t figured that part out yet. Obviously there has to be a balance between how small you can make your stop loss and still leave enough room for price fluctuations so that price will hit your TP. That’s the tricky bit.

An alternative that I’ve been looking at. (no statistics yet) is a break of the daily open or close rather than the high or low. A cursory look gives me the impression that this might have a higher pip yield.

The problem with back testing a Daily high/low breakout method using Daily chart data is that you can not know if your stop got hit before your profit target.
You would need 5 min. data to test this.

A while back I tried this method using 20 SL and 20 TP but it was not profitable and the win rate was barely 50%.

If you find a way to make a simple method like this work, I’d love to hear it.

[Synergistik]

The problem with back testing a Daily high/low breakout method using Daily chart data is that you can not know if your stop got hit before your profit target.
You would need 5 min. data to test this.

A while back I tried this method using 20 SL and 20 TP but it was not profitable and the win rate was barely 50%.

If you find a way to make a simple method like this work, I’d love to hear it.

The ST can do this for you.

[TalonD]

The problem with back testing a Daily high/low breakout method using Daily chart data is that you can not know if your stop got hit before your profit target.

yep, got to agree with you there. you can’t just say this candle went this far past the previous candle high or low and I could have made so many pips. It could go the wrong way before it goes the right way.
So what’s a solution to that?

[stevefromnaki]

yep, got to agree with you there. you can’t just say this candle went this far past the previous candle high or low and I could have made so many pips. It could go the wrong way before it goes the right way.
So what’s a solution to that?

I got excited about this method with the high win rates, so tried it on a demo but found it got stopped out, even though it moved in the right direction and would of hit profit levels later. I wondered how to test it, so downloaded minute data and put it into a database and wrote a huge SQL query to cycle through the lows (for a long trade, after i identified the long was triggered) to see if the trade would hit SL before TP.

I’m still working on the code, but will let everyone know how i go. I’m sure there are other backtesting programs that would do the same thing, but i like SQL

stevefromnaki.

[TalonD]

I got excited about this method with the high win rates, so tried it on a demo but found it got stopped out, even though it moved in the right direction and would of hit profit levels later. I wondered how to test it, so downloaded minute data and put it into a database and wrote a huge SQL query to cycle through the lows (for a long trade, after i identified the long was triggered) to see if the trade would hit SL before TP.

I’m still working on the code, but will let everyone know how i go. I’m sure there are other backtesting programs that would do the same thing, but i like SQL

stevefromnaki.

I use SQL some at work but not an expert with it for sure. Let us know what results you come up with. I haven’t traded this high low lately because I got interested in another trading idea.

[stevefromnaki]

I use SQL some at work but not an expert with it for sure. Let us know what results you come up with. I haven’t traded this high low lately because I got interested in another trading idea.

Will do, am currently working full-time and jump on at lunchtime to put on 4 trades on a demo (buy and sell limit orders on EUR/USD and cable).

Give me a week or 2 to get the code working properly and will post what I come up with

stevefromnaki.

[Ray_1]

This is the big thing with breakout strats, you have to have some pattern to it. Quite simply going short at the zone where the last 2 times buyers have come is just a recipe for disaster. I agree that the more an area is hit, the more likely to breakout of it but I’d be looking to get in way before the breakout so I could have my stops at breakeven.

I haven’t done enough analysis but any breakout strateies that do work are very likely to be quantitatively based on trading volume and ATR. There is a reason why the 24 year old “pro” over at F4N suceeded in trading breakouts when GJ had a daily range of over 400 pips and why his method failed substantially in the last 18 months (GJ range much much smaller now).

How about using candlestick patterns and trading the break of the pattern? I believe using candlestick patterns will reduce the risk. My 2 cents thought.

[polexa]

great…takes me a long time to read through 63 pages on this topic.

I just want to say the idea is great, statistically speaking it works. I sit in front of the screen full time, so practically it works.
However like everyone here say, when put on auto mode, setting the stop is the difficult part. Once you set a stop the winning ratio decrease so if you have a deep pocket, instead of setting tight stop and tight T/P. If you set 200 T/P and 20 S/L, as time goes by despite more losses and less win, overall the result should be better.

[TalonD]

How about using candlestick patterns and trading the break of the pattern? I believe using candlestick patterns will reduce the risk. My 2 cents thought.

hey Ray you mentioned that before but I have to admit I’m still not very good with candle patterns but I keep looking for them in the charts.

[TalonD]

If you set 200 T/P and 20 S/L, .

200 TP seems kind of high for an intraday thing.

[Master_Tang]

200 TP seems kind of high for an intraday thing.

I think he meant 200S/L, and 20 T/P…

He talks about a higher winning ratio, without tight stops, so that’s what I took it to mean.

We’ll see!

Sounds like a recipe for disaster to me. Two losses would set you back 20 wins.

[TalonD]

I think he meant 200S/L, and 20 T/P…

He talks about a higher winning ratio, without tight stops, so that’s what I took it to mean.

We’ll see!

Sounds like a recipe for disaster to me. Two losses would set you back 20 wins.

ah lisdexia! lol. yeah I don’t mind having a stop greater than a tp but that’s too much, you have to consider win ratio also as a factor.
I’m still trying your idea and liking it so far.

[jeroenva]

hey Ray you mentioned that before but I have to admit I’m still not very good with candle patterns but I keep looking for them in the charts.

Hello to you all, I think it’s a nice system, but to make profit consistantly you guys should take a look at this, I think you left a couple of things out…
Hopefully this helps…

forex-strategies-revealed.com/advanced/midnight-setup

[jeroenva]

and this…

forex-strategies-revealed.com/advanced/breakouts-of-breakouts-trading

[TalonD]

I like that site, there are a lot of good ideas there.

[stevefromnaki]

I have run the SQL and came up with the following:
Close,Count,SL/TP,Pips, Ratio
Profit,904,25,22600, 38%
Loss,1133,-15,-16995, 47%
Time,357,5.19,1851.6, 15%
Total,2394,N/A,7456.6, 100%

This is based on minute date from Jan-01-01 up to June 30th 2010, so has 10.5 years of data. I looped through each day and got the high and low, then looped through minute data for the next day, and determined if a high or low was breached, triggering a trade. Then once the trade was open i determined what was hit first, TP, SL, or neither (time close). Results are in table above.

Total of 2,394 trades, 904 closing in profit of 25 pips, 1,133 closing with a SL of 15, and 357 trades are closed at the end of that days prices (meaning neither TP or SL was hit), and on average are cosed out at 5.19 pips profit.

This is on EUR/USD only, with no pip cushion above or below (i.e. a pip buffer to buy, say 2 pips higher than the previous days high). Total pips profit of 7,456.6, might not sound like much for 10 years of trading, but thats the results with this strategy based on my input. (TP=25pips, SL=15pips, and pip buffer=0, and trade closed at end of day if not hit)

Comments/queries/suggestions?

stevefromnaki.

[TalonD]

7,000 in ten years isn’t much. the week before I went on holiday I made close to 200, that would work out to about 800 a month or around 10,000ish a year… That’s a different trade method though. I still think with some improvements this daily high low could be a good one. If you combine it with candle patterns or support/resistance etc. don’t know how you could examine things like that with an EA or sql though. Price does seem to react around daily high low / open as well as weekly lines…

I’m sounding more and more like a flat liner :eek: