FTSE 100 trading system - only 10 days data but 100% hit rate so far - need more data

[counter_violent]

Good...looks like wer'e all in agreement then. (except BJ who as usual has added nothing to the debate)

 

[Liquid validity]

Good...looks like wer'e all in agreement then. (except BJ who as usual has added nothing to the debate)

I am.
I certainly wouldn't knock the idea or put him off at all.
Drawing conclusions from 10 days of data is a fair point though.
What I'm running now is hitting something like a low 60% strike right now.
I know thats an aberration based on forwards testing and backtesting.
As long as he bears that in mind, which he does appear to be, good luck to him.

In fact if he is interested, I will show him how to load free data into
Ninja Trader and test it if he is interested (won't be for a couple of weeks due to time though - no promises either...)
I'll admit I am curious myself.

 

[Liquid validity]

In fact if he is interested, I will show him how to load free data into
Ninja Trader and test it if he is interested (won't be for a couple of weeks due to time though - no promises either...)
I'll admit I am curious myself.

Having said that, looking at the OP, this looks as if it does really need pure tick data.
http://www.tickdata.com/products/futures/available-futures-data/

Free m1 time series is highly likely to be inadequate for this task.
I don't currently have any decent FTSE tick data.
I may still give it a shot with the m1. although I wouldn't expect it to be that
accurate, and if typical stops and targets are small, it will do more harm than good...

 

[counter_violent]

The system is a simple opening range breakout system that I've been back testing.

However my spread betting company only give 10 days of past data on the 30 minute time frame. So wondering if anyone with more data can help me test it.

Currently for the past 10 days it has captured 153 pips, so 15 points odd a day on average.

The bar used for this system is the 8:00 - 08:30 bar.

You go short on the break of the 30 minute bar, stop placed 2 points above the high of the 30 minute bar. The profit target is the high minus the low of the 30 minute opening ftse bar.

The reverse for going long.

Over the past 10 days non of the trades have been stopped out. I'm sure this is just a good run, so would be grateful if someone could test for the past year or so to give a better idea of the hit rate.

Cheers.

Might be worth checking out Mark Fisher (logical trader) as he has done extensive work in this area. Stuff all over teh webs.
http://vimeo.com/15967639

 

[j2o]

Thanks counter_violent, I will download a demo.

Another wining trade on this system today, so 11 in a row now.

Opening range was 19 points.

High -6388
Low - 6369

Buy @ 6388. Target of 6407

Hi Rupert,

I have a similar trading style to yours, I place my trade after 1 hour rather than 30 minutes. I would be interested in comparing notes, how has your performance been since you posted this?

j2o

 

[cablemonster]

Here is a bone I will throw you to chew on if you will. Your 8 to 8.30 data creates a price range. If this price range is lower than the average of the last 100 ranges does this mean your target is more likely to be hit?

What happens to price after it leaves the range?

Good luck

 

[ordinaryguy76]

Hi Rupert,

I have a similar trading style to yours, I place my trade after 1 hour rather than 30 minutes. I would be interested in comparing notes, how has your performance been since you posted this?

j2o

Been there, done that. Sometimes it works, sometimes it doesn't. There is no real edge here.

Good luck anyway. You might be lucky to get a good run.

 

[Hoggums]

All the FTSE data you can process down to the tick level at Forex Historical Data - it's CFD data but should mirror the underlying pretty closely.

 

[j2o]

All the FTSE data you can process down to the tick level at Forex Historical Data - it's CFD data but should mirror the underlying pretty closely.

Thanks for the link, will have a look at that later.

 

[j2o]

Been there, done that. Sometimes it works, sometimes it doesn't. There is no real edge here.

Good luck anyway. You might be lucky to get a good run.

Thanks for the feedback, I really appreciate it. You wished me luck and ultimately, that's probably what it will turn out to be, however because of the success I've had so far, I feel I should continue on an experimental basis.

My system is slightly different to the one described by Rupert. I am spread betting and my stake is very small so I'm unlikely to go bankrupt. I always use a stop loss and my initial target is only 10 points. I am experimenting with moving the stop loss to breakeven when my target has been reached. Sometimes I automate and sometimes I close manually. I have only tried this for a short period so I cannot provide accurate data at this stage but the results so far indicate: 20% losers, 20% closed at breakeven and 60% winners.

I will continue with this as long as it remains profitable. Any other suggestions for improving the system would be most welcome. Thanks.

 

[Splitlink]

:!: I am strongly advising you: Save yourself from future emotional and financial heartache and TEST your system first with a minimum of 10 years of historical data.

I'm sure I'm not welcome here by saying the above, but you will bitterly regret not 'listening' to me.:!:

Ten years? You have to be kidding. I advise OP to keep going with small stakes. If you have had a good start if woukld be a waste of time to do excessive backtesting unless the system starts to go pear shaped.

I realise that I am answering a March post and we are, almost in May! I'll keep reading and my fingers crossed for you---just watch your stake size and you will be ok.

 

[Vasco da Banana]

Thanks for the feedback, I really appreciate it. You wished me luck and ultimately, that's probably what it will turn out to be, however because of the success I've had so far, I feel I should continue on an experimental basis.

My system is slightly different to the one described by Rupert. I am spread betting and my stake is very small so I'm unlikely to go bankrupt. I always use a stop loss and my initial target is only 10 points. I am experimenting with moving the stop loss to breakeven when my target has been reached. Sometimes I automate and sometimes I close manually. I have only tried this for a short period so I cannot provide accurate data at this stage but the results so far indicate: 20% losers, 20% closed at breakeven and 60% winners.

I will continue with this as long as it remains profitable. Any other suggestions for improving the system would be most welcome. Thanks.

Not knocking you or your idea, but as others have suggested this could 'work' for a day, or a week, or a month, or a year, then it might stop 'working'. To improve it you'd just have to add more criteria for entering or exiting trades... but that would also have to be based on something proven. If exits are already discretionary, your system isn't really definite, anyway.

 

[Splitlink]

Not knocking you or your idea, but as others have suggested this could 'work' for a day, or a week, or a month, or a year, then it might stop 'working'. To improve it you'd just have to add more criteria for entering or exiting trades... but that would also have to be based on something proven. If exits are already discretionary, your system isn't really definite, anyway.

The idea is to keep it simple, Once more criteria, etc, is brought in one needs a university degree to sort it all out.

I bet Livermore did not know what he was starting! Neither did the bucket shops that he used, either,

 

[malaguti]

The idea is to keep it simple, Once more criteria, etc, is brought in one needs a university degree to sort it all out.

I bet Livermore did not know what he was starting! Neither did the bucket shops that he used, either,

I go along with this, the idea is to constantly monitor the system and "refine" the criteria..typically it will be indicator period adjustments, as cycle lengths change.

 

[j2o]

Not knocking you or your idea, but as others have suggested this could 'work' for a day, or a week, or a month, or a year, then it might stop 'working'. To improve it you'd just have to add more criteria for entering or exiting trades... but that would also have to be based on something proven. If exits are already discretionary, your system isn't really definite, anyway.

Thank you Vasco, Splitlink, Malaguti, for your suggestions.

Vasco, you're suggesting that I need to add more criteria and it has to be "proven" i.e. I need an edge. I'd like to ask if this edge has to be a mathematically proven formula or if I provide say 6 or 12 months of profitable results, would that be regarded as an edge?

Do I really need to trade this system profitably for ten years before I can say yes, it works?

By the way I'm not claiming that this is my system, I read an article by Tim Wreford entitled "Developing a Trading Strategy" in the "Articles" section of this website. My system is based loosely on the strategy described in that article. Judging by the number of member comments in that article, many of the "older" members will have tried that strategy and possibly given up on it, indeed Tim Wreford himself doesn't seem to be claiming any success with it. Does anyone remember this strategy and had any success with it?

 

[Vasco da Banana]

Thank you Vasco, Splitlink, Malaguti, for your suggestions.

Vasco, you're suggesting that I need to add more criteria and it has to be "proven" i.e. I need an edge. I'd like to ask if this edge has to be a mathematically proven formula or if I provide say 6 or 12 months of profitable results, would that be regarded as an edge?

Do I really need to trade this system profitably for ten years before I can say yes, it works?

By the way I'm not claiming that this is my system, I read an article by Tim Wreford entitled "Developing a Trading Strategy" in the "Articles" section of this website. My system is based loosely on the strategy described in that article. Judging by the number of member comments in that article, many of the "older" members will have tried that strategy and possibly given up on it, indeed Tim Wreford himself doesn't seem to be claiming any success with it. Does anyone remember this strategy and had any success with it?

My belief is that if a simple day trading system somehow makes a profit, the market will tend to adjust itself so the same system gives it all back, sooner or later. If it takes ten years for this to happen, you'll be laughing all the way to the bank, of course, but even then it might just be luck.

Because the markets have steadily risen for years, possibly the best and most simple 'system' would have been to buy and hold something that tracks the index, for no hassle and effort.

 

[Splitlink]

Thank you Vasco, Splitlink, Malaguti, for your suggestions.

Vasco, you're suggesting that I need to add more criteria and it has to be "proven" i.e. I need an edge. I'd like to ask if this edge has to be a mathematically proven formula or if I provide say 6 or 12 months of profitable results, would that be regarded as an edge?

Do I really need to trade this system profitably for ten years before I can say yes, it works?

By the way I'm not claiming that this is my system, I read an article by Tim Wreford entitled "Developing a Trading Strategy" in the "Articles" section of this website. My system is based loosely on the strategy described in that article. Judging by the number of member comments in that article, many of the "older" members will have tried that strategy and possibly given up on it, indeed Tim Wreford himself doesn't seem to be claiming any success with it. Does anyone remember this strategy and had any success with it?

I think that everyone understands that this is not your system. It has been tried over many years, by many people.

However, it depends on whether you have that edge. If anyone has an edge, they are unlikely to tell us, because that is their secret between success and failure.

If I may criticise it, I do not like the size of the stops that a 30 minute bar requires. But that is a personal thing. My advice is that, as you are on a winning streak, you keep at it and see how it pans out.

 

[hacks]

Although I'm a bit sceptical, I'm willing to test this on Metastock. I have data from September 2008, mainly from FXCM and GKFX.

 

[j2o]

Although I'm a bit sceptical, I'm willing to test this on Metastock. I have data from September 2008, mainly from FXCM and GKFX.

I don't have the back testing facility so I would be very interested in seeing the results of your test.

I am in the process of forward testing my own version of this strategy and will report the findings later this month.

 

[hacks]

I don't have the back testing facility so I would be very interested in seeing the results of your test.

I am in the process of forward testing my own version of this strategy and will report the findings later this month.

Tim Wreford uses the following rules:

Do not trade on Thursday.
Do not trade if previous day’s
average trading range x 1.6 > average previous 5 days.

Do you apply these rules as well, or do you ignore them?