Leverage for Spreadbetting

Discussions about the testing and simulation of mechanical trading systems using historical data and other methods. Trading Blox Customers should post Trading Blox specific questions in the Customer Support forum.
Post Reply
immiejay
Contributor
Contributor
Posts: 2
Joined: Wed Dec 31, 1969 7:00 pm

Leverage for Spreadbetting

Post by immiejay » Mon Jul 02, 2007 10:41 am

The software looks good and should help my trading.

When will ver 2.2 be available? And if the global currency is changed does that only affect equity or also the contract price?

Also please find attached the dealing guide for IG Index through who I trade spreadbets rather than future contracts.

Could you please tell me for example if I traded 1 Cocoa Contract and with the equivalent margin deposit what leverage factor I should use. (page 61)

Similarly on page 62 for NLG and Gold on page 63.

If the leverage factors differ for different futures then I could change the figure in the portfolio but a global change would be easier.
Attachments
dealing_handbook.pdf
(836.71 KiB) Downloaded 233 times

Tim Arnold
Site Admin
Site Admin
Posts: 8523
Joined: Tue Apr 06, 2004 1:41 pm
Location: Boston, MA
Contact:

Post by Tim Arnold » Mon Jul 02, 2007 1:23 pm

Version 2.2 for the Builder edition is available now. The 2.2 version for the Pro and Turtle editions will be available by the end of this month, if not sooner.

If you change the system wide base currency, all P&L transactions will be converted back from the denominated currency to the base currency.

I don't have any experience with Spread Betting, but perhaps others on this forum do.

Turtle40
Roundtable Knight
Roundtable Knight
Posts: 201
Joined: Wed Oct 19, 2005 1:53 pm
Location: Guernsey, Channel Islands

Post by Turtle40 » Mon Jul 02, 2007 2:27 pm

immiejay,

I use spreadbetting for my trades. The way I tackle the bet sizing is by using a simple excel spreadsheet to work out the £/point equivalent to the number of contracts.

All I do is put in the entry and stop levels then the number of contracts traded is calculated (fixed fractional). This will be the same number as TB calculates, then the number of £/point is calculated as well. If you want the formulas I use I can pm them to you.

I don't use IG any more but watch out for the different tick sizes. For example coffee (10 ticks/point), copper (20 ticks/point) and TY bonds can be anything from 1/64th to 1/32 but are quoted in decimal. These vary between SB companies.(Just to keep you on your toes!)

See example below:
Attachments
Excel Converter.jpg
Excel Converter.jpg (41.68 KiB) Viewed 6236 times

immiejay
Contributor
Contributor
Posts: 2
Joined: Wed Dec 31, 1969 7:00 pm

Post by immiejay » Tue Jul 03, 2007 1:21 am

Turtle40,

Thanks for the helpful response.

However, do you leave the leverage in TB unchanged?

Which SB company has good spreads and a range of futures that you can suggest?

Can you please email / upload the sample excel spreadsheet.

Has TB and the 'Way of the Turtle' trading method worked for you using spreadbetting?

Many thanks once again for your helpful remarks.

immiejay

Turtle40
Roundtable Knight
Roundtable Knight
Posts: 201
Joined: Wed Oct 19, 2005 1:53 pm
Location: Guernsey, Channel Islands

Post by Turtle40 » Wed Jul 04, 2007 3:30 pm

Hi immiejay,

To answer your questions:

I leave the leverage set to 1. All I am doing is "translating" the number of contracts into, in my case, £/point. I simply use the entry and stop price from TB which is entered into the excel sheet that I will shortly be pm-ing to you.

SB companies vary. What you may like to do is download other dealing guides and compare. You can normally download them from their sites without having an account. The two that I favour are CityIndex and Finspreads. Be aware though that when TB puts in a correction for slippage you will probably find it much worse with an SB quote.
For example the CityIndex NG spread is 100 ticks. CT is usually 50 ticks, S three whole points (12 ticks) and so on. You will need to make allowance for this in your testing.

We these sort of spreads I find that longer term systems work better. The entry price is not quite so critical if you are trying to stay in a trend for weeks or months.

TB has without doubt transformed my trading. Without it I would be nowhere near as competent as I am now. The software is simply amazing. How we all managed without it is a mystery and I'm sure many folk would agree.

Way of The Turtle is also an excellent book. You must get a copy. The information contained will give you a greater understanding of how to test than anything else I have read. We all owe a huge debt of gratitude to c.f. and his team.

Good Luck!

Jens Albrecht
Senior Member
Senior Member
Posts: 40
Joined: Thu Feb 08, 2007 3:19 pm
Location: Dortmund, Germany

Post by Jens Albrecht » Thu Jul 12, 2007 1:17 pm

I can only give some experience from trading some months with Finspread. It is really easy and works good for me trading a long term system. Though there are some caveeats which will cost some money to learn.
First thing, they raised the minimum bet to 1 euro (assume it's then 1 pound too) for internet transaction. I had to re-work my trading system as I didn't get the neccessary bet size within tolerable risk.

Then you will have the same problems as trading real contracts. Energy bets are way too large sized. Using e.g. 1% risk gives you a 4€ bet for Soybeans and a 0.2€ bet for crude oil. You really must have a moderate account size to get a bet size > 1€ with crude oil. Or you find a way to set risk to something like 5% (you should read c.f. book :D).

So my tip is to first make a excel sheet like turtle40 did, evaluate the resulting bets for all markets you want to trade (or that you afford...) and check that the bet is > 1€ for all...don't forget a possible drawdown.

Apart from all that...I can recommend Finspreads (excluding any opinion about size of spreads. I don't know the ones of other spreadbetters).

Imperium
Roundtable Fellow
Roundtable Fellow
Posts: 52
Joined: Tue May 17, 2005 1:04 pm
Location: London

Post by Imperium » Fri Jul 13, 2007 2:09 am

Capital Spreads are good too, usually tighter spreads than other SB companies. Also their CEO answers questions on t2w.com which I think is a really nice touch as far as customer service goes (just like this forum!).

Turtle40
Roundtable Knight
Roundtable Knight
Posts: 201
Joined: Wed Oct 19, 2005 1:53 pm
Location: Guernsey, Channel Islands

Post by Turtle40 » Fri Jul 13, 2007 9:43 am

Probably the most important factor with SB companies is to use at least two if not three at the same time. That way you can shop around for the best spread for each instrument-they can and do vary tremendously.

The main players remain CityIndex, Finspreads (owned by CityIndex), IG Index and CMC Markets.

Jens: I don't know how you are calculating your bet sizes, but I find Crude (Nymex) easy to size.

If you take a 1.7ATR stop and 1.8% risk/trade, you only need an account size of £12000 to trade at £1/point. I find NG works fine as well (fixed fractional approach)

This is one of the major advantages of SB firms, the bet size can be scaled right down. In the above example you are actually trading about 1/5th of a contract..obviously not possible with a "normal" broker

White Cube
Roundtable Fellow
Roundtable Fellow
Posts: 98
Joined: Fri Oct 20, 2006 2:25 pm
Location: London, UK

Post by White Cube » Wed Aug 22, 2007 12:16 pm

Hi Turtle40

It looks like Spread betting can be a better approach than paper trading before trading a real future account as the starting equity required for SB is much lower.

But how do you test a Spread betting Strategy in Trading Blox®?

Do you run simulations using the futures setting (as if you were testing a portfolio of futures contracts) and then tackle the bet size with your excel spreadsheet? But in this case are the results of any test valid as there seems to be many big differences between trading futures contracts with a futures broker and betting with spread betting houses.

For example Are the bets at spread betting houses quoted in £ BP when futures are quoted mostly in $ US ? if both are quoted in $ US how do you handle the difference between the future and spread bet price? Can you roll like you would do with a futures broker and then enter the spread as the slippage figure?
As for the parameters in a test do you enter 0 (zero) for commissions and slippage? But how do you simulate the spread (buying at the ask and selling at the bid) (Do you enter the spread figure for the slippage parameter?)

Thanks to share your valuable experience

PS: Can you please email me or upload your sample spreadsheet to convert the bet size?

Raphael

Turtle40
Roundtable Knight
Roundtable Knight
Posts: 201
Joined: Wed Oct 19, 2005 1:53 pm
Location: Guernsey, Channel Islands

Post by Turtle40 » Wed Aug 22, 2007 4:33 pm

Hi Raphael,

Will PM you in the next couple of days. Bit busy at the mo.

White Cube
Roundtable Fellow
Roundtable Fellow
Posts: 98
Joined: Fri Oct 20, 2006 2:25 pm
Location: London, UK

Post by White Cube » Sat Aug 25, 2007 12:56 pm

Turtle40,

I have been thinking of how you calculate the amount you can bet on each spread betting trade and have worked on an excel spreadsheet like yours. You will find my worksheet attached to this email.
I have in fact considered a future account with a starting equity of $100,000.00 and a spread bet account with a starting equity of £100,000.00 Indeed in your version your calculation of the number of contract for the future account gives the BPV in US dollars so the calculation should be done IMO with a starting equity also in US$.

This point is in fact I think important as I still wonder how you can with confidence trade real a system with spread betting when the system has been tested previously on a future configuration. For example if the system has been tested in a future configuration with a starting equity of $ 1,000,000.00 or even 10,000,000.00 and with specific parameters like slippage, commission how can you then trade your system on a £ 15,000.00 spread betting account? IMO the two accounts are not equivalent. What is indeed the equivalent of a $ 1,000,000.00 future account in spread bet? I read somewhere on the forum a post I think by c.f. who said we should trade a system that has been tested and tested on the very same platform. How do we test a system for spread betting?

Raphael
Attachments
Spread Betting size.xls
(15 KiB) Downloaded 182 times

Turtle40
Roundtable Knight
Roundtable Knight
Posts: 201
Joined: Wed Oct 19, 2005 1:53 pm
Location: Guernsey, Channel Islands

Post by Turtle40 » Mon Aug 27, 2007 10:35 am

Hi Raphael,

Sorry for the delay in replying. You have asked a lot of questions so I will try to answer as best I can!

For the actual mechanics of SB, I suggest you look at these links and read the downloadable dealing guides:

www.cityindex.co.uk
www.finspreads.com

(Yes you can roll contracts in the normal way).

The reason I use a SB account is purely one of scale. If I had the account size that would allow me to trade with a "normal broker" then that it what I would do. In the meantime I can scale down my trading whilst at the same time keeping the same risk/trade that my system demands. This might not possible otherwise, as the contract size could be less than one.

I don't see any disparity in testing in $ and trading in £. My reasoning is thus:

TB lets me test and design (hopefully) tradeable systems. Brokers allow me to action the system. If I have an entry at say 50 with a stop loss at 45 I could be risking 500 points, depending on the instrument. Using the fixed fractional position sizing in TB it might tell me to buy 10 contracts. That will be worked out as per my 1% risk/trade.

By choosing to use an SB company my excel sheet says buy (bet) £25/point. My risk/trade is exactly the same. Entry and stop levels are unchanged.
As far as TB settings are concerned I leave slippage and commissions unchanged. You might argue that one should try to equate with SB (which doesn't charge commis. but has a larger price spread) with a normal broker but my main concern is the execution of the price levels and risk within the portfolio. Yes there is a bit of a fudge factor but that's what I do.

You could decide to set commissions and slippage to zero giving better results but I like to test in a "worst case" situation as the SB price is likely to give you worse fills offsetting the slippage and commission figure.

(Taking out commissions and slippage made a 7% "improvement" to my system's results).

So I am happy to trade a system tested on a 1mil. a/c as I know that trading it with £50,000 will allow me to position size and make trades that would be below the one contract size, if using a normal broker.

Also don't forget any profits are tax free in this part of the world. Hard to equate that in TB's settings.

Trading involves compromises at many levels I tend to treat this as another one.

It looks as if you have my excel sheet already. Let me know if you still want me to send it.

Cheers.

White Cube
Roundtable Fellow
Roundtable Fellow
Posts: 98
Joined: Fri Oct 20, 2006 2:25 pm
Location: London, UK

Post by White Cube » Tue Aug 28, 2007 1:56 pm

Turtle40

Thanks for the links. I find indeed Spread betting a good way to start with a smaller account or to paper trade. I have opened a demo account to see how this all works. Doing a bit of simulation with the excel bet size worksheet it appears that if we want to bet the minimum of £ 1 per point we roughly need a starting Equity of £ 12000 - £ 15000. Are there spread betting houses that allow betting les than £1?

To carry on with the discussion of starting equity, I think it is possible to design a system that trades futures in TB that would give good results with a $ 1,000,000.00 but bad results with a £50,000.00. In both cases we would use the same fixed fractional risk management risking 1% per trade. It is still not clear for me how you know that when you start spread betting with a £15,000.00 you will be in the $ 1 million case and not in the £ 50,000.00 case? In fact how do you relate your BS starting equity with the starting equity you have used in your simulation?

I have tried to understand how you made your spreadsheet from you previous post on the forum and come with my own version. But I am not sure that it is correct. So yes please I would appreciate that you send me your own spreadsheet so that I can compare you formula with mines.
Thanks
Raphael

Stephen Newton
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Wed Dec 31, 1969 7:00 pm
Location: London, England and Cape Town, South Africa

Post by Stephen Newton » Fri Aug 31, 2007 4:30 pm

A very interesting thread here guys. I think I agree with Turtle 40. There does not seem to be a need to change Trading Blox to test for the specifics of spread betting if you assume that what you loose on spreads you gain through no commission.

To me it is quite simple. There are only 2 constraints to bet/trade sizing:
1) The your total trading capital coupled with the amount of risk you wish to take on each trade; and
2) The minimum trade lot size set by the broker

The principles of capital and risk management are fully covered by Trading Blox, are scalable and therefore do not differ in the case of spread betting as opposed to a normal broker.

Without a minimum trade lot size (ie 2 above), in theory, you should be able to exactly mirror the trades made when both 1 and 2 apply, no matter how small your total capital is. The exception is when the numbers get so small that Zeno’s paradox and the convergence of an infinite sequence of numbers applies!

In back testing therefore you should get identical trades and returns in % terms (ignoring the wider spreads and other differing costs) until you reach the minimum trade lot size of the spread better if my logic is correct.

Turtle40 I would also be interested in looking at the formulas in your spreadsheet if you are prepared to share it on this forum…

Turtle40
Roundtable Knight
Roundtable Knight
Posts: 201
Joined: Wed Oct 19, 2005 1:53 pm
Location: Guernsey, Channel Islands

Post by Turtle40 » Fri Aug 31, 2007 4:40 pm

Thanks for your interest in my simple excel spreadsheet I will post it here over the weekend

Turtle40
Roundtable Knight
Roundtable Knight
Posts: 201
Joined: Wed Oct 19, 2005 1:53 pm
Location: Guernsey, Channel Islands

Post by Turtle40 » Sat Sep 01, 2007 3:43 am

Hi Stephen,

Thanks for your comments. I agree with you 100% that the whole procedure is indeed quite simple. My take is to use the power of TB to build a good system in the first place. I normally test with a far larger account size that I actually have (1mil. for example) because I want all signals to be taken from square one. If you start with less it can distort the results in the early stages as position sizing will filter out trades that would demand less that one contract to be traded.

So the good news is by rescaling the whole operation down to "realistic" account size (for me) I can trade a 1mil. account with say 100k by using spread betting instead of a normal broker. This is an advantage that we have over the US, where I believe it falls foul of their stricter gambling rules hence it is mainly us Europeans who may find this of use.

All that my spreadsheet does is take the signals from TB in the form of entry price and stop and perform the fixed fractional calculation. It generates the number of contracts (which should be the same as TB's with the same starting capital plugged in) and also gives the number of £s/per point to bet (this could be any currency of course).
As mentioned before though, SB companies vary in how they quote instruments, so if you use it make sure you check the appropriate dealing guide.
Attachments
SpreadBet Converter.xls
(30.5 KiB) Downloaded 201 times

Stephen Newton
Roundtable Knight
Roundtable Knight
Posts: 127
Joined: Wed Dec 31, 1969 7:00 pm
Location: London, England and Cape Town, South Africa

Post by Stephen Newton » Sat Sep 01, 2007 6:58 am

Turtle40 - for once an advantage to being in Europe and Trading a 'small' account!

Thanks for the sheet.

Stephen

White Cube
Roundtable Fellow
Roundtable Fellow
Posts: 98
Joined: Fri Oct 20, 2006 2:25 pm
Location: London, UK

Post by White Cube » Sat Sep 01, 2007 9:11 am

turtle40

Thanks for the spreadsheet and for sharing your experience with spread betting. It appears indeed to be a very interesting approach for smaller accounts.

Raphael

Post Reply