Page 1 of 1

Gold futures: which contract (near or far)?

Posted: Tue Apr 17, 2007 7:37 am
by John Macgregor
I'm a share trader who's switching to gold futures.

From the charts it's obvious that short-term trading is
difficult because of gold's volatility.

For me that means holding for at least a few months - fine by
me.

That could mean buying a far-out contract like Feb 08.
That's fine too in principle - except that the Feb 08 is at a
~$30 premium to the nearby. By the time I sell (early 2008)
that $30 will have evaporated, will it not?

The alternative is buying a closer month (e.g. June 07)
and rolling over every contract month. That has the
obvious downside of more time & brokerage. But there is also a likely 'slippage' in price between the old and new contracts, I think?

That is, every time I sell the expiring month to buy the
next month, am I not likely to sacrifice a dollar or 3? And
(after about 10 months) see my $30 evaporate anyway?

I guess my basic question is this:

Is it in the nature of futures to sacrifice money in one or the other of the above ways; and does it not make much difference in the end which way you trade, as either will cost you about the same?

I'd be most grateful if someone could enlighten the
newbie.

Posted: Tue Apr 17, 2007 1:08 pm
by John Macgregor
Thanks for that.

I don't trade GLD, as I want the leverage of futures.

Can you expand on:

"Futures offer tremendous margin, which, when used over an extended period of time, can be quite detrimental, unless your trade is dead on." ?

I'm interested to hear you're a short-term trader - one hears there aren't many.

Thanks

Posted: Tue Apr 17, 2007 3:19 pm
by eudamonia
John,

Welcome to a website of short-term traders (well most of us). And welcome to the world of futures.

What TapeRead is saying is that when leveraged (as you naturally will be in Futures since all contracts are bought on margin - essentially you put up a good faith amount but not the full value of the contract) you have the power to do tremendously good and tremendously terrible things to your trading account. Often in a very short period of time. You choose :)

For myself I trade the stock index futures (eMini Russell, and eMini Dow). I scalp this market several times a day (each trade lasts no more than 5 minutes) and look to make about $1 or 10 ticks worth each day (that's $100 per contract in the futures). The advantage of this method for me is that I limit my risk of exposure and drawdown while allowing me tremendous access to leverage that simply wouldn't be available for those who hold trades overnight or longer. Thus it's pretty easy to hold 10 contracts with a $20k account - although it can be a bit of a rollercoaster if you aren't careful.

Edward

Posted: Wed Apr 18, 2007 6:55 am
by John Macgregor
Thanks for your welcome Edward.

When I said "not many short-term traders" I meant "not many short-term gold traders". At least one gold guru (Enrico Orlandi) says that's the case, tho I haven't confirmed.

I have a black box trading the eMini Russell for me: it makes about 5% a month so far; tho it's meant to make about 10% a month.

In terms of risk, I suppose an advantage gold has for me is that I don't believe its downside is great. It could go down to $630, even dip below $600 - but it wouldn't be for long based on the last 5 years of this bull market - in which it's rarely gone below its 200dma, and never for long.

Of course nothing is risk-free; but most things that could go wrong in the world from here on forward (terrorist attacks; US economy taking a dive; US$ doing same) are all bullish for gold. There isn't that much that's bearish for it - tho history is an imperfect guide to the future.

I am interested in your day trading activities, and indeed have been generously tutored by The Rumpled One in his method (and also the Sheiner method) - and will try the former in due course.

But my idea of diversifying is having numerous trading methods to use, and I'm presently working my way toward TRO thru a long list.

Thanks again.

Posted: Wed Apr 18, 2007 7:54 am
by John Macgregor
PS: I'm still keen for someone to explain to me the difference between nearby and far-off futures contracts. The practical and profit/loss differences, when you trade them.

For instance in gold right now you would buy a June 07 contract at something like the price of spot gold. However you'd buy a Feb 08 contract at $30 above the price of spot gold.

But it's not like you get a $30 higher profit when you sell the contract next February. By then the price of your Feb 08 contract has come down to something close to the price of gold. Your $30 premium has gone.

In other words far-off contracts lose value against spot, over time.

Of course I still might make a good profit - i.e. the gold price may be a lot higher by next Feb. But it would still be minus that $30 premium I paid against the spot price 8 months earlier.

A strategy for a long-term trader to overcome this is to always buy the nearby month - and roll over to the nearby contract every time the present one expires. But that loses you a dollar or 3 in premium each time too. Plus brokerage every month instead of just once.

The question is: Has anyone analysed which method is cheaper or better?

Posted: Wed Apr 18, 2007 3:11 pm
by sulli
Unfortunately, I believe that is going to be the cost of doing business for you.

I am about to test one of my swing systems on Comex Gold, once I receive my historical data from tickdata.com. If this system appears to work well on GC (as it has on other markets), I'll probably trade the contract with the largest open interest (e.g. GCM7). For this system, the holding time is ~ 5 days so, for me it will be easy to cycle through contracts.

For you: I would test your method/system and determine how those roll/overs, slippage, commissions, premium will affect your results? It may not be worth trading or you should look at different products.

Good luck.

Posted: Mon Apr 23, 2007 7:18 pm
by rosenskier
hi john,
from what i know about futures- here's the thing. i think everyone has made it clear the risk elements involved. simply put- if you were planning on buying 1000 shares of GLD at 68 and stopping at 63, you're just as well off buying 1 contract of gold at 680 and stopping at 630. either way, you're gonna lose $5k and your profit/loss should be thought out in either case. so at least you only have to use half of a 10k account to make that trade instead of coming up with like 40k to cover margin and interest in a stock account.
if you're really good, trading futures will allow you to scale up very quickly which stock trading wont. i know someone who had a 100k account and now has made over a million trading gold and silver futures. he'll have 20 contracts easily on a trade. there is no way he could have grown that quickly (or failed so miserably if that is your fate) trading GLD, as 20,000 shares will cost over $1 million to own GLD.

about the time frames. gold trades pretty standard since it is based on currencies, politics, and economic outlooks which has less of a cyclical element. therefore, the rollovers (and longer term contracts) are based on convenience, volume, and expected time for your trade to work itself out. and if you get close to expiration, the cost of rollover is basically a commission since everyone else pays the spread too. (i've heard that spread is a 'storage fee' in metals). its not like we're gonna need more gold in february than in july since february is golden idol season or something.
when you trade grains, natural gas, or fuel- the months you trade matter. with grains, the months reflect different crops, weather, and cyclical demand for that crop. fuel- summer driving season for instance, natural gas- the october hurricane contract can be all over the place while the june can be moving no where.

so in short, knowing the cyclical elements to the particular instrument you are trading is important when looking at which month you'll trade. but i think in gold its nominal- so if you're trade is based on a 6 month outlook, you can either buy the far out contract or keep buying the front month and roll it over if expiration nears.

hope that helps,

jr

Posted: Tue Apr 24, 2007 7:30 am
by John Macgregor
Thanks very much to both of you: yes, it helped a lot.

I've gone long 2 Feb 08 cts and margined up my account with 14k so I'm covered if they fall from the 690s to the 630s. (I'm using the gold price equivalent - actually the Feb 08 cts are around 720.)

If they did that, I actually think I would top up the account and wait for them to come up again. I'm pretty bullish gold.

If they don't come up again, and stay at (say) 600, or fall below that, I will lose about a third of my trading capital, and I will have been wrong about something on a very fundamental level. Which will probably mean a year off trading (so I don't lose the other two-thirds) and back to the books.

It's a punt I'm prepared to take: I've studied gold long and hard in the last year, and if I don't jump in on the basis of that knowledge, and present conditions, it won't be good for my psychology.

rosenskier: Your comments about margin/scalability are appreciated. As are the stuff on cycles - I had no idea 'seasonal' stocks were so variable.

(Gold does have seasons BTW - e.g. May is usually bullish - but they're a secondary factor, and as you point out, not based on actual meteorological seasons.)

I think the spread from ct month to ct month - i.e. the gradually rising 'cost of carry' - is interest on the money you have borrowed to buy the un-margined portion of your ct.

Finally, I did some calculations, and found that - if I am going to liquidate in Jan 08 - buying a Feb 08 now and sticking with it till then is about 50% cheaper than buying the front month and rolling it every month or two.

The Feb 08 will 'decline' in value against the (ever-changing) front month, sure - but you pay a premium anyway every time you roll - and these add up; plus there is the increased commissions. Plus my time, and the stress of having to think about it every rollover period.

Even tho the buy/sell spread on the Feb 08 is one dollar tonight (30c in day sessions), it's a better deal from my perspective given these factors.

Thanks again.