Are futures suitable for exploiting trends that last several months to years? The major issue will be settlement dates. Is there a mechanism that will automatically roll over all my positions to the next contract or do I have to do this by hand? Selling the old contracts and immediately buy the new one. I'm at IB.
If you use a "spread order" to roll over, the "exit old" and the "enter new" transactions will occur at the same instant. For example, today happens to be the day I am rolling over my long term position in the index futures known as "CAC-40" (the French stock index futures), info HERE. I submitted the following order: Bonjour, this is a spread order, in the CAC-40, traded at MATIF in Paris: Buy 25 contracts Oct-2008 CAC-40, AND Sell 25 contracts Nov-2008 CAC-40, as a spread trade, at the market price for the spread. This order simultaneously exits my existing short position in the October CAC, and enters a new short position in the November CAC. As a previous writer remarked, not rocket science.
Any advice on how far out to open a trade? If I want to take a long term futures position in an equity index, what are the pros and cons to rolling over to the contract 6 to 9 months out versus doing it every 3 month?
Why don't you try it both ways and gather some data for yourself? Put on half of your position in the near contract and roll it to the next-near, then the next-near, then the next-near, ... , as needed. Put on the other half of your position in a distant contract and roll less often. Then see which half you like best, when the trade is complete. The beauty of this method is, you aren't relying upon the say-so of complete strangers. You're comparing two KNOWN FACTS, in your own account, and deciding for yourself which of them is best for you personally.
MGJ, I think that is a good idea, and know real lessons can be learned by doing things yourself. But, I also don't believe in re-inventing the wheel and DO believe in learning from other's mistakes. I was looking for any glaring mistakes that might have stood out to a veteran and would be missed by me. Is it just a matter of preference? If so, it would seem that saving commissions and slippage by rolling over less often would be smarter. I figured there is more to it that I was missing and that either liquidity or contango or something else I couldn't even fathom might make one strategy better than another.
Personally, I haven't done your trade both ways and tracked the results for lots and lots of trades, to see which method I prefer. So I can't really help you. But perhaps it is worth wondering, if you're trying to capture big moves by holding onto positions for a long time, whether it's silly to worry about slippage and commissions. Perhaps for big-move trades held a loooong time, S+C represent a negligible fraction of expected average profit. Even if the trade is rolled over numerous times, incurring commission each roll and incurring (small) spread slippage each roll. But I'm sure you've already thought about that and looked at the dollar size of some examples, either by-hand on charts, or in computer "backtests" of trading systems. Maybe, of all the factors to consider when deciding between the two methods, S+C is one of the least important (?)
Good point MGJ. In a long term position, S and C are smaller percentages of your costs. I guess what I'm trying to figure out is how the different contracts are priced. Sometimes the further out contracts trade at a higher price. Sometimes they trade at a lower price. As expiration approaches, they trade closer to spot price. What I'm wondering is, could you get a far out contract that is trading at a higher price than the closest month and then as its expiration approaches, the contract loses value as it trades closer to the spot price. If this is a potential, the cost of that could be a lot higher and it might be worth the C+S of rolling over the current month contract which trades closer to the spot price of the good.