Newbie asks: How to trade futures with only $100K?

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.
Robwynge
Senior Member
Senior Member
Posts: 42
Joined: Wed May 25, 2011 10:33 pm

Post by Robwynge » Sat Aug 20, 2011 6:27 pm

fab1usa1 wrote:Oh Great Socrates (dig your handle!):

Me, I'm a bit stubborn. I like the full-sized contracts. Had a long talk with the wifey and we upped our investment to $165K. As we speak that allows us to trade 6 contracts of EuroDollar, 1 contract of Corn, 1 contract of Soybean, and 1 contract of Canadian Dollar. BTW, at this moment I decided on a "universe" of 8 markets including: EuroDollar, Canadian Dollar, Corn, Soybean, Cotton, Copper, Gold, and Heating Oil. Am planning on expanding that universe as equity grows. (Please consider sluggo's excellent recommendation; I have not yet implemeted but will.)

Good luck!
Hmm, unless your broker has very different margins than mine (Thinkorswim), some of those contract sizes are huge relative to the size of your account. For example, Cotton moves $500 per point and currently has an ATR of about 4. That's $2,000 per day in average price movement, which would be over 1% of your account. You would need some very tight stops to keep your risk in line. Heating Oil moves $42,000 per point, with an ATR of about 0.1 right now, so that's an average daily move of $4,200 per day. Copper is similar. My account size is in the ballpark of yours (a tad higher), and I don't touch those contracts. I will trade the full sized grains, livestock, sugar (pushing it), some currency futures and sometimes bonds (though lately the ATRs have gotten so large I would trade etfs like TLT and IEF instead). Otherwise I look for minis or default to efts (or spot currencies).

fab1usa1
Roundtable Knight
Roundtable Knight
Posts: 383
Joined: Sat May 21, 2011 8:30 pm
Location: New York

Post by fab1usa1 » Sat Aug 20, 2011 9:33 pm

Thanks for your reply, Robwynge.

A little more information before I proceed: I began this odyssey with a long-only stock system that had a nice CAGR of 20.1% but the DD was unbearable at 35.4% over 33.2 months. I decided that I needed to mix in futures to bring the DD down. The system that I am writing about is a mix of 66% full-sized futures, as mentioned, and 34% stocks, starting with $250,000 USD. Backtesting from 1/1/2000 to 6/30/2011 yields a CAGR of 20.42% and DD of 13.5% over 11.9 months. Now that's more like it!

Since that post I more closely examined the trade history and the filtered trade file. It turns out that the EuroDollar futures were the only contracts trading in the early years. The major factor in the growth of the equity curve was due to the stocks. As equity grew more futures met the 1 contract threshold and began contributing to further growth. At the conclusion of the backtest the equity had grown from $250,000 to $2,159,959. That teminal amount was more than sufficient to fund all of the full-sized futures.

I spent the good part of today taking heed of the advice offered to me by RHC: trade the mini's and micro's, and for all else trade ETFs and ETNs. To be honest the reason why I avoided those was the lack of history for backtesting.

So now my new, new system is a mix of 50% mini and micro futures, 25% commodity ETFs and ETNs, and 25% stocks. Backtesting from 1/1/2010 to 6/30/2011 yields a CAGR of 15.44% and DD of 7.3% over 4.7 months. That grows $250,000 to $309,731 in that year and a half.

By the way this system handled the recent stock market correction nicely: the terminal equity on 8/19/2011 is $305,720. Not bad.

Fab1

Chris67
Roundtable Knight
Roundtable Knight
Posts: 1046
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 » Sun Aug 21, 2011 2:24 am

How does it test out from 1980-2011 ?

fab1usa1
Roundtable Knight
Roundtable Knight
Posts: 383
Joined: Sat May 21, 2011 8:30 pm
Location: New York

Post by fab1usa1 » Sun Aug 21, 2011 9:13 am

As I said, lack of history puts minis, micros, ETFs and ETNs at a disadvantage. My CSI price history for full-sized futures goes back to 1992 and stocks to 1999. I worry about survivorship bias when going too far back with stocks since my stock system picks from a basket of companies that are currently constituents of the S&P500, DOW, and NASDAQ. It's a learning experience that's for sure. There's a phrase amongst computer engineers called "analysis paralysis", or a phrase I like better from my father: "shit or get off the pot".

Chris67
Roundtable Knight
Roundtable Knight
Posts: 1046
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Post by Chris67 » Sun Aug 21, 2011 9:30 am

take your point that its hard when data doesnt exist / survivorship bias , etc - gotta balance that with the need to test back a long way though I guess - Mind you 11 years is definately better than nothing and a few stock market moves this century have almost corresponded to 87 ?
Good luck

Mike S
Contributing Member
Contributing Member
Posts: 8
Joined: Thu Dec 16, 2004 6:25 pm

Haven't checked Micros at the CME

Post by Mike S » Wed Aug 24, 2011 2:12 pm

But shouldn't the prices be extremely similar? The MM's for these products are usually quite good.

The contracts need to converge to the same price at expiry. Too large of a daily divergence and they will get picked off by other MM's

The differences in price will mainly impact the fills, not the overall trend.

rhc
Roundtable Knight
Roundtable Knight
Posts: 464
Joined: Tue Nov 18, 2008 8:46 pm
Location: Oz

Post by rhc » Mon Aug 29, 2011 3:53 am

Further to the previous post above;

Upon completion of my weekly orders review, I noticed a new stop-entry signal for platinum but unfortunately the risk-to-protective stop was just too large for my account. I could grumble about it and whine about how life is unfair that I can’t afford it or I could look for an alternative.
A quick google search revealed 2 ETN’s (PGM & PTM) and 1 ETF (PPLT) for the metal Platinum.
I will always prefer the ETF over the ETN to avoid counterparty risk and I will generally prefer the one with the highest liquidity & the one that actually holds the physical.
A bit of reading about the providers & their product and a review of each chart shows that PPLT is an ETF and trades with the most volume (approx. 500K) and seems to track the metal Platinum the best AND holds the physical.
So I will consider placing my entry stops here as a proxy for the entry stops in the futures.
Of course you pay an expense ratio of 0.6%, but what can we do. Thems the breaks

Moral of the story:
The small account need not be limited in trading a diversified portfolio when there are ETF/ETN products available. . . . .but we all knew that anyway.
(again, see the good Mr Garners book mentioned above)

HighSierraTrader
Roundtable Fellow
Roundtable Fellow
Posts: 87
Joined: Sat May 02, 2009 8:09 pm
Location: Truckee, California

Post by HighSierraTrader » Sun Sep 18, 2011 11:09 pm

This is a very interesting and useful thread to read, having just found it this evening. Thanks to all for your input on the difficult subject of instrument selection and position sizing in smaller accounts.

This may be old news so most readers, but specific to rounding up in portfolio manager, I find that it makes a slight improvement to backtested results on almost any system, the amount depending upon the test time span. Why? Because rounding up increases the risk per trade by as much as 100% for those 0.5 contract theoretical positions that tend to be concentrated near the start of the backtest. As equity grows the number of sub-1 contract positions decreases as does the overall effect of rounding up.

Using position size rounding, one's loss on stops is disproportionally multiplied at the start in relation to the overall backtest statistics, even though the system may be performing as designed. If the start of trading occurs in a choppy market there may be a higher than average number of stops per month or quarter or year that get hit. This generates unwelcome emotions not to mention trading losses beyond what was expected (but not out of spec for the system). The end result may be to abandon pefectly good trades or even the system.

The lesson I take from personal experience trading a system using position size rounding is that if one wants to take more risk, do so explicitly by increasing risk per trade, not by hiding that excess risk in the computations.

AceofAce
Senior Member
Senior Member
Posts: 49
Joined: Thu Sep 30, 2010 5:06 am

Post by AceofAce » Sat Nov 12, 2011 6:19 pm

To trade LTTF for a small account using mini/micro futures maybe, but using ETF/ETNs, i don't think so. You'll need to trade these on margin or as CFDs / spreadbets. Either way, the broker will change financing on their notional value adding their own financing spread of 100 possibly 200 basis points. If you get charged so much financing on the notional value of an ETF/ETN (which in turn has embeded the financing cost in the basis of the underluing futures) it would be impossible to make any money. To make things worse, the lower the volatility of the ETN/ETF, the bigger the position so the bigger the financing charge as a % of your equity.

Micro/mini futures would be the only way out but then again they are not that liquid and not sure they track the price of the benchmark future accurately.

Bounty
Senior Member
Senior Member
Posts: 33
Joined: Tue Jun 19, 2007 8:59 am

Post by Bounty » Wed Nov 16, 2011 8:51 am

I am taking a different approach and am testing systems on accounts at the $30k and $50k levels. Initially I built portfolios of markets with margins of less than $5,000. From there I tested various parameters across the whole group and then started weeding out those markets that had lower profit factors.

For instance, in a $50k portfolio, it trades BO, FC, PA and TY. In the $25k - $30k portfolio I'm using just FC and TY.

I haven't gone live yet, but step testing across the range of time periods has shown consistent results (I look at MAR frequency distributions of all the time period results).

fab1usa1
Roundtable Knight
Roundtable Knight
Posts: 383
Joined: Sat May 21, 2011 8:30 pm
Location: New York

Post by fab1usa1 » Wed Nov 16, 2011 9:21 am

Bounty, Interactive Brokers recognizes PA as Paladium, and TY as 10y US Treasury Notes. What is BO and FC?

Moderator's edit: BO = Bean Oil, FC = Feeder Cattle. LINK

Toosday
Roundtable Fellow
Roundtable Fellow
Posts: 70
Joined: Fri Jun 03, 2011 11:21 am
Location: Austin, TX

Post by Toosday » Wed Nov 16, 2011 9:32 am

BO is soybean oil and FC is Cattle Feeder.

fab1usa1
Roundtable Knight
Roundtable Knight
Posts: 383
Joined: Sat May 21, 2011 8:30 pm
Location: New York

Post by fab1usa1 » Wed Nov 16, 2011 9:43 am

Thanks for the clarification.

Interactive Brokers symbol for Soybean Oil is ZL, and Feeder Cattle is GF.

stopsareforwimps
Roundtable Knight
Roundtable Knight
Posts: 199
Joined: Sun Oct 10, 2010 1:47 am
Location: Melbourne Australia

Post by stopsareforwimps » Wed Nov 16, 2011 5:42 pm

Bounty wrote:I am taking a different approach and am testing systems on accounts at the $30k and $50k levels. Initially I built portfolios of markets with margins of less than $5,000. From there I tested various parameters across the whole group and then started weeding out those markets that had lower profit factors....
Make sure you do out of sample testing, or the results when you put your money on the line might disappoint.

By out of sample testing, I mean you develop your strategies on one set of data and then do a final test on another set of data that played no part at all in developing those strategies.

In the machine learning community, they often use three sets of data. One set is used to optimize the system parameters. The second set is used to pick which system to use eg how many parameters to include in your system. The third set is used to get an idea how it is going to perform in real life.

Remember, the minute you go back and change your system in response to the results using the third set you lose.

More on what I have been learning in Andrew Ng's machine learning course soon ml-class.org.

DPH
Roundtable Knight
Roundtable Knight
Posts: 113
Joined: Tue Nov 17, 2009 1:07 am
Location: Central Pennsylvania
Contact:

Post by DPH » Thu Nov 17, 2011 12:04 am

fab1usa1,

One suggestion is to come up with some kind of screen that allows you to monitor a large universe of markets, but only "permissions" a few of them at any given time. (similar to Sluggos suggestion)

This is basically what I have done for the smaller accounts I manage (50-100k)..... My systems consider over 70 markets everyday, but I have screened them down hard enough so that on average those accounts are only in one or two trades at a time.

It is not unusual for me to have a 50k account "trading" over 70 markets, yet still have a margin to equity ratio under 10%.

So far this is the best solution I have come up with. It is far superior (in my opinion) to just simply trading a small portfolio of only 10 markets (or so)...

stopsareforwimps
Roundtable Knight
Roundtable Knight
Posts: 199
Joined: Sun Oct 10, 2010 1:47 am
Location: Melbourne Australia

Re: Newbie asks: How to trade futures with only $100K?

Post by stopsareforwimps » Sun Jan 15, 2012 2:11 am

fab1usa1 wrote:I read with great interest the old thread entitled: How to begin trading futures with account < $100K ? and all of the great contributions by members of this forum. I certainly see the advantage of having a large portfolio of 100+ markets and $1,000,000 startup capital for LTTF
Nothing profound here, just a graph I produced of return versus starting wealth. This is of a simple triple moving average system with a stop on a mixture of US and international markets. Returns are in USD. I am assuming you are not prepared to take a significant chance of blowing up. I ignore size-related changes in market impact and fixed costs such as data.

Below about $300k (2012 dollars - I adjusted the starting wealth back to the start of my data using CPI) you are seriously impacted in your ability to trade standard futures contracts and make money. From there to $500k things get better, and from $500k to $1m there is some further improvement. Beyond that, not much benefit is evident.

If you are young and have the ability to bounce back, it may be rational to take a serious risk of blowing up early on. Let's say your strategy is such that you have a 50% chance of "taking off" and a 50% chance of "blowing up". It may be better to take the chance than to wait patiently for your savings to accumulate sufficiently to allow a negligible chance of blowing up. This depends on your age, earning capacity, financial commitments, and the shape of your utility curve for money.

Quite a few of the Market Wizards followed this strategy without realizing it at the time. After the first blowup they tended to work a lot harder at their vocation of trading. and to take risk a lot more seriously.

Before trading ETFs ETNs etc, check the fidelity to the underlying asset, costs, spreads etc. In trading you can assume nothing.

One downside of starting with $1m is that it would be very distressing to lose that amount of money, and that may well be the most likely outcome for a new trader.
Attachments
sharpe-ratio-by-starting-wealth.png
sharpe-ratio-by-starting-wealth.png (23.28 KiB) Viewed 4942 times

fab1usa1
Roundtable Knight
Roundtable Knight
Posts: 383
Joined: Sat May 21, 2011 8:30 pm
Location: New York

Post by fab1usa1 » Sun Jan 15, 2012 8:38 am

Thanks. I did decide at the new year to abandon trading futures (for now) and replace them with commodity etfs and etns as per 'rhc'.

With so little capital I was compelled to trade a basket of only about a dozen mini and micro markets at a gut-wretching 2% risk-per-trade just to play the game! I grew fatigued by the intraday swings and decided on a change.

Early in my TB career, about six months ago, I devised a long-only stock system that backtests well, meets by goals, and has proved to make money. The only argument against it was its max te dd of 50%. I needed to find something to help attentuate that pain, hence futures. If I was better capitalized that dual system would have gotten the max te dd down to 15%. Now having abandoned futures trading (for now) and replacing it with commodity etfs and etns I am down to 25% max te dd. I can live with that.

Thanks for everyone's contribution to this post!

Robwynge
Senior Member
Senior Member
Posts: 42
Joined: Wed May 25, 2011 10:33 pm

Post by Robwynge » Sun Jan 15, 2012 4:53 pm

I also trade ETF's and ETN's when futures contract sizes are too large. One trick to avoid soaking up a large chunk of capital is to use a "synthetic stock position" with options. To go long, buy a call and sell a put at the same strike price, and to go short, buy a put and sell a call at the same strike. Bingo, you have a delta of one with much smaller margin outlay. There is the little problem that you can't easily use stops, so you'll be taking extra risk to have close-only stops. As a result, I make sure to keep my position sizes smaller than I might otherwise. Of course, not all commodity ETF's have options, so this isn't a perfect solution in cases. But it's something I'll use to make sure that I don't have to pass on valid trades because I am running low on available capital.

rhc
Roundtable Knight
Roundtable Knight
Posts: 464
Joined: Tue Nov 18, 2008 8:46 pm
Location: Oz

Post by rhc » Sun Jan 15, 2012 11:33 pm

When trading your system on ETF’s as a proxy to Futures you might, at some stage, be required to answer this question;

“o.k. what should I do now?, I just got filled on my ETF order but this order would not have been filled on the futures (it just missed!).
As far as my futures system is concerned I’m flat, yet I have an ETF position!!??
Should I exit the ETF with a loss to reflect the status of my system or should I simply hold the position thus contradicting my systemâ€
Attachments
RBOB_Futures.jpg
RBOB_Futures.jpg (52.96 KiB) Viewed 4852 times
RBOB_ETF.jpg
RBOB_ETF.jpg (64.77 KiB) Viewed 4853 times

rhc
Roundtable Knight
Roundtable Knight
Posts: 464
Joined: Tue Nov 18, 2008 8:46 pm
Location: Oz

Post by rhc » Sun Jan 15, 2012 11:39 pm

Further to the above charts;

You might notice that during early October the ETF hit its lows.
Not so for the Futures, the lows were hit in August.
This too may have caused a 'what-should-I-do' moment for the RBOB ETF trader who bases his/her decisions on the futures.

Post Reply