Total Capital Requirements

Discussions about Money Management and Risk Control.
Post Reply
Roundtable Knight
Roundtable Knight
Posts: 1052
Joined: Tue Dec 16, 2003 2:12 pm
Location: London

Total Capital Requirements

Post by Chris67 »

Happy Christmas to Everyone

Please note that Ive never been a great believer that the best value from forum discussions comes when everyone agrees , but more so when frank exchanges of opinion and experience are able to take place in a adult and respectful manner. With that said :

What I have found to date , is what I obviously new from the start - that the demo default of Veritrader is obviously set up to absolutely maximise returns. As soon as you put the same data in over a much longer period the results fall away dramatically. One factor that also affects performance dynamically is start capital. I am reasonably fortunate in that I can start with a capital base in excess of 300 , 000 usd but I am finding that again performance dramatically tails off below 500 , 000. With 1 million usd then you get something reasonably acceptable.

I have seen many conversations on this forum whereby people have been told that its ok to start with smaller amounts of capital as long as they foresee volatile results.
I beg to diifer in that I believe your selection of markets , and future ability to trade these markets during a heavy drawdown period , collapses on amounts of starting capital below 500,000.
You simply cannot trade many markets and therefore ypou cannot achieve the necessary mix in the portfolio.
If we talk about mini futures we have to be honest and accept that most mini commodities , such as coffee and soybeans , yen , etc are really a very dangerous arena due to volumes and slippage - yes this applies to 1 lot trades as well as I have experienced on many occasions.

I will keep testing but my experience in trendfollowing over the years and my testing on veritrader demonstrate to me that you better be well capitalised and by that I mean in excess of 500,000usd.
Before anyone talks about spreadbetting please realise that it is totally unrealistic to trade commodities on spreadbetting because at present the spread are between 3 and 20 times as wide as the futures markets .. at present.

Any thoughts would be warmly appreciated and I know I'm probably setting myself up for some criticism but I would love to learn the errors of my above statement

Roundtable Knight
Roundtable Knight
Posts: 513
Joined: Wed Apr 16, 2003 1:18 am
Location: Nowhere near

Post by Kiwi »


I will just address one exageration which reflects a misconception that I think others have as well.

That is that the larger spread on a spread bet makes it unattractive for some/all/any strategy (depending on whos statement it is).

The facts are:
- given a strategy you need to ask can you make money with a given vehicle.
- if at first appearances you can't is there some way you can apply your intelligence to solve that problem.

Here is an example that directly addresses what you said. Lets say that the normal spread on CL is 2pts. On finspreads it is currently 10pts. So an entry and an exit would cost me 16 pts more with spreadbetting than my broker (ignoring brokers fees).

How can u resolve that?

Well u could well say that 50% of trades retrace a bit before they take off. And that this doesnt mean that they will be bad. So, lets just say we wont take a trade unless we can get a 16pt advantage. So if our buy order is issued for 2900 we will buy at 2884.

If we miss 50% of trades well so what ... and u can test so what by examining all CL trades over the last 20 years and seeing what happens if you miss all trades that dont retrace 16 pts before continuing.

I have a spreadbetting strategy where the spread takes over 30% of the average trade. But I dont care because it is still outperforming my long term strategies (by a huge factor) and has much lower volatility.

Forum Mgmnt
Roundtable Knight
Roundtable Knight
Posts: 1842
Joined: Tue Apr 15, 2003 11:02 am

Post by Forum Mgmnt »


The demo wasn't set up to maximize returns. This is a byproduct of the decision that the demo wouldn't allow testing of more than a few years of data.

What you are seeing is simply the direct result of this insufficient data. Simulating a system over a short period (three or four years), one can almost always get the kinds of returns you won't encounter over the long run. This is one of the ways that the scammers rip people off. It's too easy to optimize the short term and come up with unrealistic results.

Everything is more sensitive with less data. The portfolio, the parameters of the system; each change can affect results with two or three years of data to a degree you won't begin to approach looking at 10 year or 20 year tests. You don't even need to commit the serious errors of optimizing each market individually to get unrealistically rosy results.

The starting capital requirements are so dependent on the type of system that I don't believe one can make general statements. However, I strongly concur that there is a big loss of efficiency and much higher risk when trading smaller accounts.

One can build smaller portfolios an bring the minimum trading capital requirements much lower than $500,000 with most systems, but there are costs for this both on the return and risk side.

- Forum Mgmnt
Erwin Dicker
Senior Member
Senior Member
Posts: 25
Joined: Mon Dec 22, 2003 3:11 pm
Location: Moelingen - Belgium

Re: Total Capital Requirements + BREAKOUTS

Post by Erwin Dicker »

I disagree with you about capital needed = 1000.000 usd.

You can pick up trends with 100.000 usd. Even less, in my vision and experience.

Which trendfollower is NOT LONG IN EURO/DOLLAR? It is impossible to not be in a long position. The maximum risk is appr. 2000 - 3000 usd per position. And this is a reasonably wide stop. Now we are at 1.24, the trend started around 90-95.

You can take signals with appr 1000 usd risk, so that would be 2 % of 50.000 usd.

Even the mini's are oké, although less liquidity. Never get out in panic, take some time to get out.

I feel you had bad experiences and want to get confidence from testing. This is oké. I would like to know what you test exactly in Veritrader.


One remark about 20 day breakouts. Why they are not so good is that the breakout is not so often a breakout of another top (or bottom). The # of days are simply too less to set. Therefore other rules have to help.

Why the 100 day breakout is better than 55? I think that there are already a few losers..... previous to the 100 day breakout. The longer it takes the greater the chance that again a trend starts. I tested this with 10 year end of day data. But 55 days are oké too.

Perhaps the second part of the story is not so relevant, but anyway.

Senior Member
Senior Member
Posts: 44
Joined: Mon Apr 05, 2004 11:16 pm
Location: San Diego

Post by Asterix »

I'm curious to hear what others have to say about what I might call the Novice Trader's Dilemna. Here is how I would describe it. Most new traders start with small capital - maybe $5000 to $15000 is a good guess. As I've seen discussed in several different places on this forum, a good trend following system requires a much higher level of starting capital in order for the trader to have a decent chance at success. Consequently, the new trader will need a different system other than a long term trend following system in order to participate.

Now here is the dilemna. If you are starting with only $5k to $15k, you will obviously need another source of income. (Assuming you are living in the US, of course. :wink: ) Most of the time, this means working for an employer who will require you to be physically present at the workplace during normal working hours. This requirement will conflict with many trading systems that require watching the market during trading hours.

Consequently, the novice trader has a conflicting set of circumstances that make getting started very difficult.
Roundtable Knight
Roundtable Knight
Posts: 208
Joined: Fri Jan 23, 2004 1:39 am
Location: San Diego

Getting Started

Post by ksberg »


It's been said that you want to match a trading system to personal characteristics, but I would add that you need to also match personal constraints. I believe there are a whole range of options (I employ several of the ones mentioned).

I've found that many traders have a desire to focus on systems where they must watch the market, even when they are in situations like having a day job. There are decent end-of-day systems where you place orders before and after market. There are also systems that only involve checking prices at open, then placing orders. Some end-of-day systems will trade market-on-close. None of these require watching the market during the day. For most, you can setup some communication on fills (phone, email, pager) that allow you to handle placing protective stop orders, or manually manage synthetic OCO (e.g. bracket stops).

Another option is broker-assist-trading, offered by numerous firms (Angus Jackson, TradeCenter, and Striker come quickly to mind, and there are others). With broker-assist-trading, you're paying someone else to trade a mechanical system for you. You have to ask "what is my time worth?" I find this is a pretty good trade-off.

If you're inclined to daytrading, programs like Strategy Runner can be packaged with daytrade systems, or customized with your own, and either run automatically or handed off to broker assist.

Another option is private funds, although placement qualifications might be out of range, depending on the fund. It's certainly possible for several people to pool monies together and trade under NAIC guidelines, for example. It takes some effort to coordinate (and reach consensus), but again, you can leverage OPT (other people's time). For me, it's worked out pretty good so far.

It's probably worth examining the reasons behind asking the question. I find most traders to be fiercely independent, as in needing to do everything themselves; from placing the orders to tweaking system parameters. If you examine every other business discipline, independence is not the norm. What works well is to focus on what unique values you can bring to the equation and build out a balanced team. I believe this principle holds in trading as well: for instance, some folks are better at execution than analysis, some do better at organization, others with accounting. If the aim is to become an independent trader and treat trading as a business, why not start applying business principles that leverage your personal USP?


Post Reply