Ok Levijean point taken thanks for letting me know.
TB builder is on the shopping list, but, right now I suspect it would be like giving a 5 year old child an Indy car to learn to drive in. I'm happy finding the limits of my understanding of probability, simulation and reality using TB Turtle. Too many more parameters and dimensions would confuse rather than help.
Keeping up with the simulation and Spreadbetting?
Medius
I have run into the same problem. ie. placing spreadbets on UK open to see them stopped out and TB still has them active.
I can see 2 potential solutions, i think both are mentioned in this thread.
1) Only trade on U.S time, which i need to check but i think is possible.
2) Once in the market only trade on the next day open. ie stop loss only applied to end of day data.
Wondering if you found a solution that you were happy with.
Regards
Matt
I have run into the same problem. ie. placing spreadbets on UK open to see them stopped out and TB still has them active.
I can see 2 potential solutions, i think both are mentioned in this thread.
1) Only trade on U.S time, which i need to check but i think is possible.
2) Once in the market only trade on the next day open. ie stop loss only applied to end of day data.
Wondering if you found a solution that you were happy with.
Regards
Matt
Hi Matt,
I've been using a combination of solutions.
1. Firstly I try not to advertise stops to the market at all. (Unless I'm being compromised by my own account size related issues running out of margin etc). My account size is a lot less than the $1m recommended.
2. I bought TB builder.
3. I downloaded the Donchian MOO (MarketOnOpen) from
viewtopic.php?t=6977&highlight=moo
4. The resulting problem however is the mental agony of ignoring the gremlins which after an exit order is signalled a) tell you to keep the position open if the orignial trend appears to have resumed. Or b) if the stop is shot through: to keep the position open waiting for a better exit price.
Good luck ignoring the gremlins.
Which version of TB have you got?
Medius
I've been using a combination of solutions.
1. Firstly I try not to advertise stops to the market at all. (Unless I'm being compromised by my own account size related issues running out of margin etc). My account size is a lot less than the $1m recommended.
2. I bought TB builder.
3. I downloaded the Donchian MOO (MarketOnOpen) from
viewtopic.php?t=6977&highlight=moo
4. The resulting problem however is the mental agony of ignoring the gremlins which after an exit order is signalled a) tell you to keep the position open if the orignial trend appears to have resumed. Or b) if the stop is shot through: to keep the position open waiting for a better exit price.
Good luck ignoring the gremlins.
Which version of TB have you got?
Medius
-
Jens Albrecht
- Senior Member
- Posts: 40
- Joined: Thu Feb 08, 2007 3:19 pm
- Location: Dortmund, Germany
I am spreadbetting in the UK quite successful.
The key is to use wiiide stops. IMHO don't try to trade systems with 0.5 ATR stop or the like with spreadbetting. I use 6 ATR and this works fine. The drawback is that you have to go really long term. These systems don't tend to depend on the wider spreads of spreadbetting firms.
Also, take care when choosing markets. Calculate rollover costs in advance and choose markets then...there are some markets where rollover would cost really much ($1000 and more for me a year). Other markets are quite cheap.
Next thing: Place your 'betting stop' at the 'TB stop + betting spread'. I also got hit once where the wide spread made me automatically enter. In the real markets the high wasn't taken...ugly thing. This could also work for a shorter term system which is very robust. But I decided not to not try
With (very) long term systems you have the advantage that now TB's order generation calculates the distance to the stops. I ignore all markets where the distance proposes that the stop won't be hit. That saves much work.
I only have a small number of stops in place and they don't change too often.
Ok, it gets quite boring sometimes...
Last tip: Always (!) compare TB performance with your betting performance. My worst nightmare was when some positions had really false risk/bet sizes. I took me some days to find out. It happens as you maybe didn't adjust bet size to TB equity on entry day. I would recommend to check your bets often. I note the TB equity and the betting equity each morning and graph it with excel. If I have a doubt a double check the bets.
I hope this helps...
The key is to use wiiide stops. IMHO don't try to trade systems with 0.5 ATR stop or the like with spreadbetting. I use 6 ATR and this works fine. The drawback is that you have to go really long term. These systems don't tend to depend on the wider spreads of spreadbetting firms.
Also, take care when choosing markets. Calculate rollover costs in advance and choose markets then...there are some markets where rollover would cost really much ($1000 and more for me a year). Other markets are quite cheap.
Next thing: Place your 'betting stop' at the 'TB stop + betting spread'. I also got hit once where the wide spread made me automatically enter. In the real markets the high wasn't taken...ugly thing. This could also work for a shorter term system which is very robust. But I decided not to not try
With (very) long term systems you have the advantage that now TB's order generation calculates the distance to the stops. I ignore all markets where the distance proposes that the stop won't be hit. That saves much work.
I only have a small number of stops in place and they don't change too often.
Ok, it gets quite boring sometimes...
Last tip: Always (!) compare TB performance with your betting performance. My worst nightmare was when some positions had really false risk/bet sizes. I took me some days to find out. It happens as you maybe didn't adjust bet size to TB equity on entry day. I would recommend to check your bets often. I note the TB equity and the betting equity each morning and graph it with excel. If I have a doubt a double check the bets.
I hope this helps...
Hi Medius
Thanks for the reply.
I have TB Pro. I thought that would be enough for me as i don't have a clue about coding but the more research i do the more i think i might have to learn!
I did find that part of the reason that i was being stopped out was using a stop of less than 1 ATR. Doesn't work well with EOD. Oh the learning curve!!
I can see the merit in 'not advertising' your stops, but i think i would prefer to do that than have my stops blown to pieces. I have a very small starting stake so staying in the game is paramount.
Trading is a constant battle against the gremlins!!
Regards
Matt
Thanks for the reply.
I have TB Pro. I thought that would be enough for me as i don't have a clue about coding but the more research i do the more i think i might have to learn!
I did find that part of the reason that i was being stopped out was using a stop of less than 1 ATR. Doesn't work well with EOD. Oh the learning curve!!
I can see the merit in 'not advertising' your stops, but i think i would prefer to do that than have my stops blown to pieces. I have a very small starting stake so staying in the game is paramount.
Trading is a constant battle against the gremlins!!
Regards
Matt
Hi Jens
Thanks for the reply. Your thoughts are very helpful.
I am working on short term systems at the moment and have been wondering whether to add the spreadbetting spread or include it in the TB calculated S.L.
Would you mind exlaining this a little further? I think i understand but am not quite sure.
'With (very) long term systems you have the advantage that now TB's order generation calculates the distance to the stops. I ignore all markets where the distance proposes that the stop won't be hit. That saves much work'.
Thanks for you time
Matt
Thanks for the reply. Your thoughts are very helpful.
I am working on short term systems at the moment and have been wondering whether to add the spreadbetting spread or include it in the TB calculated S.L.
Would you mind exlaining this a little further? I think i understand but am not quite sure.
'With (very) long term systems you have the advantage that now TB's order generation calculates the distance to the stops. I ignore all markets where the distance proposes that the stop won't be hit. That saves much work'.
Thanks for you time
Matt
If the distance between today's price and tomorrow's stop is very very large, then you might conclude it is very very unlikely that your stop order will be executed tomorrow. Then you might conclude that you needn't bother placing the stop order -- it isn't going to be filled anyway. Not bothering to place the stop order saves time, which you might or might not consider beneficial.
Human judgment is involved: you, the human trader, get to choose your own definition of "very very large distance to tomorrow's stop". You have an infinite variety of choices. For example you could decide you are not going to bother placing tomorrow's stop order, if the distance between today's price and tomorrow's stop is greater than:
Or you could decide it's a Human Judgment after all, so why have a rigid rule and why do a mathematical calculation? You could decide to simply eyeball the chart (or the Order Generation report) and make a snap judgment, based on your intuition and experience. Can you or can't you live without a stop order tomorrow?
Human judgment is involved: you, the human trader, get to choose your own definition of "very very large distance to tomorrow's stop". You have an infinite variety of choices. For example you could decide you are not going to bother placing tomorrow's stop order, if the distance between today's price and tomorrow's stop is greater than:
- (Param1) times the (Param2)-day ATR
- (Param1) times the (Param2)-day Standard Deviation of closes
- (Param1) times the maximum value of TrueRange over the last (Param2) days, i.e., some multiple of the biggest 1-day move seen in the most recent Param2 days
- (Param1) times (HighestHigh(Param2) minus LowestLow(Param2)), i.e., some multiple of the Param2-day Donchian Channel width
Or you could decide it's a Human Judgment after all, so why have a rigid rule and why do a mathematical calculation? You could decide to simply eyeball the chart (or the Order Generation report) and make a snap judgment, based on your intuition and experience. Can you or can't you live without a stop order tomorrow?
-
LeviF
- Roundtable Knight
- Posts: 1436
- Joined: Mon Dec 22, 2003 12:24 pm
- Location: Des Moines, IA
- Contact:
I never use stop orders. I only trade on the open. Sometimes this results in worse fills, sometimes better. The historical performance is slightly worse with my method than using in-the-market stops. However, that is the price I am happy to pay to lighten my daily workload.
In agreement with: this.
In agreement with: this.
-
Jez Liberty
- Roundtable Knight
- Posts: 123
- Joined: Tue Nov 03, 2009 8:49 am
- Location: London
- Contact:
In the light of the above (ie putting stops in or not)
I know that OANDA is not really a spread betting firm (and they only deal in currencies) but it seems very similar (to me anyway). And one thing that strikes me is why there does not seem to be an equivalent to them in the Futures/Commodities space, ie: competitive spreads, fair funding rates (ie their funding spread is not 250 bps but closer to 50) AND most importantly (to me): with a retail API access (which would allow for a program to be written and not skip any SL orders)...
While doing a first google research, I came across Futuresbetting.com that allowed API access via NinjaTrader, but since then the company has been renamed to ProSpreads and does not seem to advertise API/NinjaTrader on their website ( http://www.prospreads.com/ ). As a side-note they seem less biased against their users (their marketing spiel implies this by selling the idea of Direct Market Access-like spread better, whatever that is...)
I also found:
http://www.cmcmarkets.com.cn/en/content ... ection=api (but this info is found in the Institution solution / partner program webpage so I guess it is not accessible to retail traders)
I might try to get in touch with these companies to find out a bit more but I would be interested if anybody has direct real-life experience with them (or of spread betting API in general).
maybe a gap in the market... Or a dead-in-the-water business plan?
I know that OANDA is not really a spread betting firm (and they only deal in currencies) but it seems very similar (to me anyway). And one thing that strikes me is why there does not seem to be an equivalent to them in the Futures/Commodities space, ie: competitive spreads, fair funding rates (ie their funding spread is not 250 bps but closer to 50) AND most importantly (to me): with a retail API access (which would allow for a program to be written and not skip any SL orders)...
While doing a first google research, I came across Futuresbetting.com that allowed API access via NinjaTrader, but since then the company has been renamed to ProSpreads and does not seem to advertise API/NinjaTrader on their website ( http://www.prospreads.com/ ). As a side-note they seem less biased against their users (their marketing spiel implies this by selling the idea of Direct Market Access-like spread better, whatever that is...)
I also found:
http://www.cmcmarkets.com.cn/en/content ... ection=api (but this info is found in the Institution solution / partner program webpage so I guess it is not accessible to retail traders)
I might try to get in touch with these companies to find out a bit more but I would be interested if anybody has direct real-life experience with them (or of spread betting API in general).
maybe a gap in the market... Or a dead-in-the-water business plan?
Hey Jez
I have looked at Prospreads and on the surface they look very good, only problem is their minimum bet. They are indeed aimed at 'pros', and this little black duck ain't quite there yet! I have spoken to them several times with rudimentary questions and they were very helpful. My aim is to have an account size that would allow me to use Prospreads.
I haven't looked at CMC but will do so.
Regards
Matt
I have looked at Prospreads and on the surface they look very good, only problem is their minimum bet. They are indeed aimed at 'pros', and this little black duck ain't quite there yet! I have spoken to them several times with rudimentary questions and they were very helpful. My aim is to have an account size that would allow me to use Prospreads.
I haven't looked at CMC but will do so.
Regards
Matt
-
Jez Liberty
- Roundtable Knight
- Posts: 123
- Joined: Tue Nov 03, 2009 8:49 am
- Location: London
- Contact:
Thanks matey/matti
I just got off the phone with ProSpreads and:
- API: they used to offer API access to their trading engine via NinjaTrader but this has been discontinued. They told me they were looking at MetaTrader - possibly in the first quarter
- They are a spread-better based in Gibraltar with "Direct Market Access". Apparently WYSIWYG: ie no requotes, instant fills, prices always track the underlying market, etc... The way they do this is by actually buying off the exchange every time you place an order and sell it back to you straight away at the same price (+their spread). In effect, they seem to be a spreadbetter-wrapper around a standard Direct Access Futures broker, solely for tax-purposes (ie you benefit from the spread betting tax breaks - which might apply to you).
There are therefore 2 major drawbacks:
- Size is identical to standard Futures trading
I was looking at this spread-betting API concept as a possible way of implementing an automatic LTTF with a "small-ish" starting equity size) - so obviously this makes it much less attractive.
- Introduces another intermediary (=Credit Risk)
I might start a thread on spreadbetting API to collate info if it's worth it?
Edited: forgot to add the link to ProSpreads CEO interview: http://www.financial-spread-betting.com ... tting.html
I just got off the phone with ProSpreads and:
- API: they used to offer API access to their trading engine via NinjaTrader but this has been discontinued. They told me they were looking at MetaTrader - possibly in the first quarter
- They are a spread-better based in Gibraltar with "Direct Market Access". Apparently WYSIWYG: ie no requotes, instant fills, prices always track the underlying market, etc... The way they do this is by actually buying off the exchange every time you place an order and sell it back to you straight away at the same price (+their spread). In effect, they seem to be a spreadbetter-wrapper around a standard Direct Access Futures broker, solely for tax-purposes (ie you benefit from the spread betting tax breaks - which might apply to you).
There are therefore 2 major drawbacks:
- Size is identical to standard Futures trading
I was looking at this spread-betting API concept as a possible way of implementing an automatic LTTF with a "small-ish" starting equity size) - so obviously this makes it much less attractive.- Introduces another intermediary (=Credit Risk)
I might start a thread on spreadbetting API to collate info if it's worth it?
Edited: forgot to add the link to ProSpreads CEO interview: http://www.financial-spread-betting.com ... tting.html
Jez
Excellent summary of Prospreads! Good interview with the CEO too! Thanks for that.
In your listed drawbacks you say that Prospreads 'Introduces another intermediary (=Credit Risk)'. Any different to any other spread betting company?
An API thread will be useful for some people now i am sure. Personally it will be a little time before i am concerned with it. I am still deciding on which company/ies to use, agonising over spreads (both quoted and actual, and boy is there a difference a lot of the time!), looking at carry costs, minimum bet sizes etc.
Regards
matti/matey/Matt
Excellent summary of Prospreads! Good interview with the CEO too! Thanks for that.
In your listed drawbacks you say that Prospreads 'Introduces another intermediary (=Credit Risk)'. Any different to any other spread betting company?
An API thread will be useful for some people now i am sure. Personally it will be a little time before i am concerned with it. I am still deciding on which company/ies to use, agonising over spreads (both quoted and actual, and boy is there a difference a lot of the time!), looking at carry costs, minimum bet sizes etc.
Regards
matti/matey/Matt
-
Jez Liberty
- Roundtable Knight
- Posts: 123
- Joined: Tue Nov 03, 2009 8:49 am
- Location: London
- Contact:
Matt,matti wrote:Jez
In your listed drawbacks you say that Prospreads 'Introduces another intermediary (=Credit Risk)'. Any different to any other spread betting company?
An API thread will be useful for some people now i am sure. Personally it will be a little time before i am concerned with it. I am still deciding on which company/ies to use, agonising over spreads (both quoted and actual, and boy is there a difference a lot of the time!), looking at carry costs, minimum bet sizes etc.
Regards
matti/matey/Matt
Completely right.. I was making this remark more in the context of broker vs spreadbetter comparison. Using ProSpreads vs Interactive Brokers (or any other Futures broker) has the drawback of adding the spreadbetter in the "Credit Risk chain" but does not provide any benefits other than potential tax savings (
ie ProSpreads does not have the "smaller trading size" advantage of other spreadbetters) - and I am not too sure tax advantages can apply to any serious venture (ie running it as a business or as main source of income). See this post for more discussion on tax and Spread betting: viewtopic.php?p=40424
-
Jens Albrecht
- Senior Member
- Posts: 40
- Joined: Thu Feb 08, 2007 3:19 pm
- Location: Dortmund, Germany
Matti,
>>>
With (very) long term systems you have the advantage that now TB's order generation calculates the distance to the stops. I ignore all markets where the distance proposes that the stop won't be hit. That saves much work'.
<<<
Sluggo explained completively what I wanted to say with the above. But don't underestimate the effort to adjust the stops for all positions each day. Also if the stop doesn't change the number of contracts can vary for entry. Ok I am part time trading and this was my personal solution.
But be really warned to try short term trading with spreadbetters like IG Index, Finspreads etc. with $120 transaction cost per contract (in TB) I got a good approximation of the two equity curves over long periods. With Oanda I maybe would it a try but I would have lost diversification.
>>>
With (very) long term systems you have the advantage that now TB's order generation calculates the distance to the stops. I ignore all markets where the distance proposes that the stop won't be hit. That saves much work'.
<<<
Sluggo explained completively what I wanted to say with the above. But don't underestimate the effort to adjust the stops for all positions each day. Also if the stop doesn't change the number of contracts can vary for entry. Ok I am part time trading and this was my personal solution.
But be really warned to try short term trading with spreadbetters like IG Index, Finspreads etc. with $120 transaction cost per contract (in TB) I got a good approximation of the two equity curves over long periods. With Oanda I maybe would it a try but I would have lost diversification.
Jens
You mention you operate using a 6 ATR stop may I ask how much you risk per trade in terms of percentage of your account?
If I adjust the stops on my simulation to 6 ATR I go from a simulated 29% CAGR to 10% and TE DD of 22% down to 15%!
Using data from 1996 to today.
Regards
Medius
Sluggo and all
Is it more normal with TB to see a conservative system outperform when it goes live or to see a high CAGR% high TEDD % system fail?
Regards
Medius
You mention you operate using a 6 ATR stop may I ask how much you risk per trade in terms of percentage of your account?
If I adjust the stops on my simulation to 6 ATR I go from a simulated 29% CAGR to 10% and TE DD of 22% down to 15%!
Using data from 1996 to today.
Regards
Medius
Sluggo and all
Is it more normal with TB to see a conservative system outperform when it goes live or to see a high CAGR% high TEDD % system fail?
Regards
Medius
Last edited by Medius on Mon Jan 25, 2010 6:53 am, edited 1 time in total.
-
Jens Albrecht
- Senior Member
- Posts: 40
- Joined: Thu Feb 08, 2007 3:19 pm
- Location: Dortmund, Germany
What you observe is IMHO the most difficult part and dilemma for small traders: Transferring the TB simulations to real world. You simply don't have the capital to trade the system with a broker without very serious contraints.Medius wrote:Jens
You mention your operate using a 6 ATR stop may I ask how much you risk per trade in terms of percentage of your account?
If I adjust the stops on my simulation to 6 ATR I go from a simulated 29% CAGR to 10% and TE DD of from 22% down to 15%!
Using data from 1996 to today.
Regards
Medius
It took me much time to find the combination of system and markets which I could profitable trade with spread betting and my amount of capital. I finally found that I needed to risk 5% per position. But it's just a number...and just valid for my chosen system/markets and spread better.
Here (spread betting) you have the problem of the minimum bet. I checked each market and took care that at all time my bet would be above the minimum bet of bettting firm. And this alone took me days of fiddling and I still live with 2 comprises which I didn't get solved.
Jens
Sounds like i am going through exactly the same process that you have been through, and boy is it time consuming ha?!!?
You warn against shorter term systems. Did you research them to the same extent as LTTF? I have started off with swing systems as CF (and others) have said they are the best way to cut your teeth in trading. Having said that, if i can't find a system that i can work with my equity in the confines of spreadbetting i may well be forced into LTTF.
Reassuring to know that this is a time consuming process, was starting to think i was too thick for this!!
Cheers
Matt
Sounds like i am going through exactly the same process that you have been through, and boy is it time consuming ha?!!?
You warn against shorter term systems. Did you research them to the same extent as LTTF? I have started off with swing systems as CF (and others) have said they are the best way to cut your teeth in trading. Having said that, if i can't find a system that i can work with my equity in the confines of spreadbetting i may well be forced into LTTF.
Reassuring to know that this is a time consuming process, was starting to think i was too thick for this!!
Cheers
Matt
-
Jez Liberty
- Roundtable Knight
- Posts: 123
- Joined: Tue Nov 03, 2009 8:49 am
- Location: London
- Contact:
Is it not possible to create an alternative Futures dictionary to be used in Trading Blox and that would reflect/simulate the margin requirement and min move of each equivalent instrument at the spread better? The simulation would just run these figures instead - or am I over-simplifying?...Jens Albrecht wrote: Here (spread betting) you have the problem of the minimum bet. I checked each market and took care that at all time my bet would be above the minimum bet of bettting firm. And this alone took me days of fiddling and I still live with 2 comprises which I didn't get solved.
Also, what is the rough ratio of minimum starting capital for a given strategy in Futures vs spread-betting (I am realising that even a high five figures account will not allow to implement a lttfs using diversification and granular position sizing in Futures so spread-betting might be an avenue to explore...)