[BlowFish]

Actually spread betting could negate the need of trading spreads. You can simply reduce your risk by betting a small amount. Pick an instrument with suitable volatility for your purposes and then bet a size that fulfils your requirements for potential loss or gain. Still kind of interested to see how you can wring juice out of two highly correlated instruments. Also won't transaction costs be pretty high? (as the spread isn't going to move as much larger positions will be requires). As an aside have you considered calendar spreads?

[jfutures]

That's a valid point Blowfish. In fact, I've tried to locate a decent online execution only broker that provides spread betting on treasuries (with the option of doing stock indexes also) but haven't had much luck as yet. Would you be able to recommend any???

When you say calendar spreads, are you referring to intramarket contract spreading?? (i.e. newer contract to older contract). Can you please give some advice on the subject?? Would really appreciate it if so.

To spread effectively over time, yes you would need a large account in order to leverage your positions. Comms can be manageable, as long as you are prepared to use a low cost execution-only service.

The best way I've found to wring out the juice would be to try to identify which bond for treasuries, or which market in stocks (NB still investigating stocks at time of writing) is setup to move 'the most' in your direction. This trade qualification may be based on a number of things, say trend lines, channeling, OS/OB, market profile, or whatever tools you can find to identify the setup. Then once you trade the spread, assuming you're correct, the trade can show profit. The key (and this is the real key) is to not overtrade, but rather concentrate on picking the cherries, even if they occur 4 or 5 times a month, which is the number of trades I place on average in a month.

For anyone interested in spreading treasuries, I found a good article here which sums it up: -

Current Issue

[BlowFish]

To be honest I don't know an awful lot about spreads. A calendar spread would indeed be buying the front month and selling one further out (or vice versa). It tends to be used on commodities for fundamental reasons (which I don't profess to understand!)

Joe Ross has a book "trading spreads and seasonals" his books are quite expensive but he tends to tell it how it is. One of these days I'll get round to reading it and then might be able to give some more helpful answers.

[coval]

Hi guys,

actually it is my first post on this forum, but as I can see you are talking here about something what I have been doing for around a year and I would be happy to exchange some ideas about spread trading.
I usually trade DAX vs CAC as well as some forex instruments.

So far, to trade in such way I have always measured correlation of the securities, to make sure that there is a chance to have the spread closed up....

Recently I have learnt that much more reliable method would be to calculate cointegration of the instruments, because 2 charts might be almost perfectly correlated but keep constantly getting away one from another - at least when you measure correlation with Pearson's method...

Than, another very important issue comes - when to take a position - the easiest way to answer is: " when the spread is higher than usually". For this I created a very simple indicator which shifts me CAC chart on DAX chart with one point taken as a mutual neutral point, i.e. Jan 1. And than you can see a nice spread between instruments.
Unfortunately, we never know whether we took a really "neutral" neutral point or was it in reality an extreme point and we need to wait months before the spread would come back and let make us earn some money...

If this is interesting to anyone please let me know so I will post in some pictures to show clearly the problem...

greetings,

CoVal

[kannondo]

Coval can you post your picture? I am interested to see what you are talking about.