Sunday Breakout Strategy

[pirateboy]

it's all down to whether the 6400 holds....still hanging on!!!

Originally Posted by phil838

I'm not going to enter the short trade if it hits. The amount of pips it went up and the day of the week are right in the middle of my enter/don't enter criteria, but the fact we're in a strong uptrend tips my decision over to "don't enter."

[pirateboy]

I'm stopped out at 6386. my luck was pip perfect as it rebounded at that exact spot. either that, or because my $5bn stop loss order turned the market back...

Originally Posted by pirateboy

it's all down to whether the 6400 holds....still hanging on!!!

[phil838]

Originally Posted by pirateboy

I'm stopped out at 6386. my luck was pip perfect as it rebounded at that exact spot. either that, or because my $5bn stop loss order turned the market back...

Mine missed the SL by one pip and is still going. It's at +40 pips now.

I'm glad I didn't move my SL up. I thought about doing that when it bounced directly off the trendline but decided to stick to the rules!

[mumpips]

Originally Posted by phil838

I'm not going to enter the short trade if it hits. The amount of pips it went up and the day of the week are right in the middle of my enter/don't enter criteria, but the fact we're in a strong uptrend tips my decision over to "don't enter."

Hi Phil,

If my target is only 100pips per week and i exited everytime when the trade hit 100, will there be any unforeseen issues that I will face in long run?
Pls advise.

Thanks

[phil838]

Originally Posted by mumpips

Hi Phil,

If my target is only 100pips per week and i exited everytime when the trade hit 100, will there be any unforeseen issues that I will face in long run?
Pls advise.

Thanks

Yeah, if you consider losing money an unforeseen issue...

Seriously though, I'm 95% sure this strategy will end up losing money in the long term if you go with a set TP.

The main problem is different market conditions. You might have made more money in the last few months (or even this whole year) using 100 pips, but in years like 2008 you would have lost big time. You need some type of criteria that moves with the markets, like ATR, ADR, candle sizes or something along those lines.

If you still want to change the numbers make sure you backtest it over at least a few years. If it comes out profitable then go for it, but I don't think it will with a static TP.

[pirateboy]

hi phil is it possible to share with us your results if they are on excel? would like to calculate the probabilities to reconfirm. give me something to do till next sunday...

Originally Posted by phil838

Yeah, if you consider losing money an unforeseen issue...

Seriously though, I'm 95% sure this strategy will end up losing money in the long term if you go with a set TP.

The main problem is different market conditions. You might have made more money in the last few months (or even this whole year) using 100 pips, but in years like 2008 you would have lost big time. You need some type of criteria that moves with the markets, like ATR, ADR, candle sizes or something along those lines.

If you still want to change the numbers make sure you backtest it over at least a few years. If it comes out profitable then go for it, but I don't think it will with a static TP.

[orpips]

Nice clean bounce off the trendline to keep those of us who weren't stopped out still in the game!

[orpips]

...and now a bounce off the overhead downtrend line. Looks like we're squeezing into a wedge. Will the breakout take us to profit or stop us out? Who knows?

[phil838]

Originally Posted by pirateboy

hi phil is it possible to share with us your results if they are on excel? would like to calculate the probabilities to reconfirm. give me something to do till next sunday...

My spreadsheet is in the zip file on the first post of this thread. I just updated it and it's current as of last week.

Also, I don't remember if I mentioned this or not but back when Orpips and I were discussing backtesting results I found some backtesting errors I made in 2008. Apparently while backtesting mid 2008 I switched to a 1H chart to look at something and forgot to switch back for a few months!! So what I thought was the Sunday candle was really only half of it and my stoplosses were smaller than they should have been!! I corrected the problem on the new spreadsheet, and it changed the systems backtesting results over the 5 year period to be about 400 pips lower (7500 instead of 7900).

It's a small mistake that doesn't really change the systems profitability, but I thought I should point it out in case anyone is using the old excel file to help with their own backtesting.

[pirateboy]

Hi phil, thanks, running through ur spreads, Sunday breakout gives a sharpe ratio of 1.2 (since 04), and actually, comparable to cowabunga (since 08, which i backtested). But we know which one gives us more time to take our holidays!

by comparison, S&P 100 yr historical sharpe is 0.4, and even Madoff based on my readings "window dressed" a sharpe of 2.5 (with 11% return, so extraordinarily low volatility beyond realms of possibility).

caveat is that don't rely on my figures and run them yourselvs, as sharpe fluctuated over the 4 yrs depending on which period you took.


--

Ok, i've calculated the sharpe ratios for all the years:
05 - 1.5 (excellent, could maybe get into nominated for a hedge fund award...)
06 - 1.4
07 - 0.94
08- 1.5
09- 0.78 (a few more months data to come, implying we ought to see better results in coming months to boost sharpe to 'normal' levels?!)

Originally Posted by phil838

My spreadsheet is in the zip file on the first post of this thread. I just updated it and it's current as of last week.

Also, I don't remember if I mentioned this or not but back when Orpips and I were discussing backtesting results I found some backtesting errors I made in 2008. Apparently while backtesting mid 2008 I switched to a 1H chart to look at something and forgot to switch back for a few months!! So what I thought was the Sunday candle was really only half of it and my stoplosses were smaller than they should have been!! I corrected the problem on the new spreadsheet, and it changed the systems backtesting results over the 5 year period to be about 400 pips lower (7500 instead of 7900).

It's a small mistake that doesn't really change the systems profitability, but I thought I should point it out in case anyone is using the old excel file to help with their own backtesting.