Trading Plan Help Required

[lukeaye]

I recommend CFD's, you need leverage. but you must understand risk beforehand as well and proper money management. PROPER. There is alot more to it then people will lead you to believe.

If you go into CFD's with no real plan, you will get hit, and hit hard. but its your only chance with such a small capital base.

 

[AlterEgo]

i disagree, whilst it is harder, it can be done. if you are correct with your position sizing you can make 700 from 100. all it means is that your limited to the trades you can take.

I never said it couldn't be done. I said "most likely" end in failure, not saying it's impossible. The poster had admitted to being a newbie. I'm sure an experienced trader, with a proper plan, could succeed at it, but a newbie?? Possible, but unlikely.

 

[lukeaye]

Yes it is dificult for a newbie, but whether he has 10,000, 100,000 or 3,000 it doesnt matter, if he is going to fail he is going to lose money anyway. If he can practice good trading methods with positive expectancy and good risk reward ratios, then he can succeed, regardles of capital base.

I find those who say you need more money, are those who pratice bad habbits such as moving stop losses, or having large losing streaks due to poor methology. More money does not mean a higher expectancy, and thats what matters

 

[AlterEgo]

Yes it is dificult for a newbie, but whether he has 10,000, 100,000 or 3,000 it doesnt matter, if he is going to fail he is going to lose money anyway. If he can practice good trading methods with positive expectancy and good risk reward ratios, then he can succeed, regardles of capital base.

Well I agree with you to a point, but let’s look at an example:

Say you had a $3000 account (as you mentioned above), how many stocks would you divide this between? If you put it all on one stock, and the stock gaps down 20% on some news announcement, that’s 20% of your whole account gone on a single trade!

So instead, let’s say you divide it between 3 stocks, $1000 in each. Your brokerage is say $15 each way, so that means you’re already down 3% (plus the spread) as soon as you buy the stock – hardly a good start.

Now compare that to having a $100K account split equally between 10 stocks. Brokerage is the same, but now only represents 0.3% of the trade value, so Risk:Reward shifts further in your favour. And one stock gapping down 20% only affects your total portfolio by 2%, so risk is much reduced.

So although it may be possible to succeed with a $3K account, you have a much better chance with a larger account.

 

[lukeaye]

Well I agree with you to a point, but let’s look at an example:

Say you had a $3000 account (as you mentioned above), how many stocks would you divide this between? If you put it all on one stock, and the stock gaps down 20% on some news announcement, that’s 20% of your whole account gone on a single trade!

So instead, let’s say you divide it between 3 stocks, $1000 in each. Your brokerage is say $15 each way, so that means you’re already down 3% (plus the spread) as soon as you buy the stock – hardly a good start.

Now compare that to having a $100K account split equally between 10 stocks. Brokerage is the same, but now only represents 0.3% of the trade value, so Risk:Reward shifts further in your favour. And one stock gapping down 20% only affects your total portfolio by 2%, so risk is much reduced.

So although it may be possible to succeed with a $3K account, you have a much better chance with a larger account.

Im afraid your missing the point entireley, you would risk say $100 per trade, regardless of brokerage and the spread, it applies to both sized accounts, your expectancy of a profitable reutrn is no less or greater, the extra hurdle is only it is harder to make enough over $9 brokerage, but really its not a great amount. a trade i set up yesterday with amp, i only risked $200 and i have already made $650, the percent increase on the win is no greater or lesser regardless of account size. Have a look at the thread i started on my trading strategy where i show the amp trade.

Your chance can be no greater or worse, how does a larger capital base give you a greater expectancy? it doesnt? it can't? please explain to me how it can?

And if you are splitting 100,000 between 10 stocks, i am very concerned for you, because when the market turns around you will cop a 50% drawdown, regardless of your account size

 

[AlterEgo]

Your chance can be no greater or worse, how does a larger capital base give you a greater expectancy? it doesnt? it can't? please explain to me how it can?

Well in the example I posted, yes it does give greater expectancy. Let me expand on the above example to better explain:

A stock you’re looking at has resistance at $2.00 and support at $1.95, so you enter on breakout at $2.01 and set stop loss at $1.94. Lets suppose our target is $2.20, lets see what this works out to with the above position sizes:

Scenario 1
Buy 500 shares at $2.01 costs $1,005 + $15 brokerage = $1,020
Sell at target = 500 shares at $2.20 minus $15 brokerage = $1,085 (gain of $65)
Sell at stop loss = 500 at $1.94 minus $15 brokerage = $955 (loss of $65)
So you have risked $65 for possible gain of $65.
Reward to risk of 1:1

Scenario 2
Buy 5,000 shares at $2.01 costs $10,050 + $15 brokerage = $10,065
Sell at target = 5000 at $2.20 minus $15 brokerage = $10,985 (gain of $920)
Sell at stop loss = 5000 at $1.94 minus $15 brokerage = $9,685 (loss of $380)
So you have risked $380 for a possible gain of $920.
Reward to risk of 2.42:1

 

[wayneL]

Yes, that's what I'm saying. 2% vs 1% will see larger drawdowns more frequently, and newer traders might be surprised at how great a difference it makes. If someone flips coins 30 times a week, it wouldn't be unusual to see a 50% drawdown within a couple months. A 50% drawdown could happen using 1% trades, but since it will usually require a far greater sample, the trader doesn't have to be as prepared for it.

Ah, some content materializes.

This may be more probable in a system where numerous correlated trades are taken. Less probable, but not impossible where a less correlated trades are taken.

However, certainly something I've always designed into my systems is overall portfolio risk. As an extreme example, take 20 2% trades all in one day in the stock market and your risk is actually 40%.

I think a figure of risk should be selected for the entire system. This will trim down the risk (but not totally negate) of such high drawdowns.

 

[tech/a]

Yes, that's what I'm saying. 2% vs 1% will see larger drawdowns more frequently, and newer traders might be surprised at how great a difference it makes. If someone flips coins 30 times a week, it wouldn't be unusual to see a 50% drawdown within a couple months. A 50% drawdown could happen using 1% trades, but since it will usually require a far greater sample, the trader doesn't have to be as prepared for it.

Dont you mean string of losses rather than drawdown?
Do you mean initial drawdown or relative drawdown on a system?

Wayne you speak of Portfolio Heat.

 

[wayneL]

Wayne you speak of Portfolio Heat.

Ahh, it's got a name. good, thanks.

 

[Nick Radge]

Probability theory - losing streaks and therefore drawdown are directly related to winning percent of the trades taken. The larger the iterations the larger the drawdown.

A system with 50% winning trades and 10,000 iterations will suffer 16 losing trades in a row at some stage.

16 at 1% is -16% drawdown.

16 at 2% is -32% drawdown.

Nick

 

[tech/a]

Ahh, it's got a name. good, thanks.

Yeh I call it Frank.

 

[Wysiwyg]

I think this is correctly worded.

The Total Core Equity = Starting equity minus open trades.
The Total Equity = Starting equity plus open trades at current value.
The Total Reduced Equity = Starting equity minus open trades value at risk.

Which is best to work out % risk using any of above please.

 

[wayneL]

Probability theory - losing streaks and therefore drawdown are directly related to winning percent of the trades taken. The larger the iterations the larger the drawdown.

A system with 50% winning trades and 10,000 iterations will suffer 16 losing trades in a row at some stage.

16 at 1% is -16% drawdown.

16 at 2% is -32% drawdown.

Nick

Something the martingale guys at the casino find out soon enough.

 

[skc]

Probability theory - losing streaks and therefore drawdown are directly related to winning percent of the trades taken. The larger the iterations the larger the drawdown.

A system with 50% winning trades and 10,000 iterations will suffer 16 losing trades in a row at some stage.

16 at 1% is -16% drawdown.

16 at 2% is -32% drawdown.

Nick

Nick I read about this in your book and don't get me wrong I definitely think it is helpful way of putting relationship between risk% and drawdown into context.

However, this simple relationship doesn't show that a trader can fail (or reach the drawdown pain threshold) in other ways... 6 losses, 1 small win, 4 losses, 2 small wins, 6 more losses and 1 medium win... and next thing you know you are down 15-20% without ever needing to see 16 losses in a row.

BTW this is a 1yr old thread... who dug this out?

 

[Mr J]

This may be more probable in a system where numerous correlated trades are taken. Less probable, but not impossible where a less correlated trades are taken.

Yes, but trade size should be adjusted for correlation for this very reason.

However, certainly something I've always designed into my systems is overall portfolio risk. As an extreme example, take 20 2% trades all in one day in the stock market and your risk is actually 40%.

It depends how we define risk. Say we flip coins for $1. I win the first flip, and decide as soon as I lose the dollar that I won, I'll quit. We make 9 more flips as I never lose the original dollar I won. How much have I risked? After the first flip, I only risked $1, and while that dollar was placed at risk 9 times, it was still only that dollar that was at risk.

Dont you mean string of losses rather than drawdown?

I assumed drawdown meant loss on account. If not, yes, I'm referring to a string of losses.

Probability theory - losing streaks and therefore drawdown are directly related to winning percent of the trades taken.

I think it's better to think of losing stretches rather than losing streaks. Also, I'll throw in that R:R is the other factor in drawdowns. Stating the obvious, but many people do overlook R:R in favour of win% (and by not look at both, they lose).

16 at 1% is -16% drawdown.

16 at 2% is -32% drawdown.

If we adjusted after each result, it's more like 15% and 28%.

Something the martingale guys at the casino find out soon enough.

And unfortunately for them, they can usually double up just 6-7 times.

and next thing you know you are down 15-20% without ever needing to see 16 losses in a row.

If we're flipping coins, 20-30 will do that, and that kind of result will occur all the time over a sample of 50. Like I said before, anyone who trades 2% better be prepared for some wild fluctuations in capital.

 

[Nick Radge]

However, this simple relationship doesn't show that a trader can fail (or reach the drawdown pain threshold) in other ways... 6 losses, 1 small win, 4 losses, 2 small wins, 6 more losses and 1 medium win... and next thing you know you are down 15-20% without ever needing to see 16 losses in a row.

skc,
Yes, you're absolutely correct and I'm not sure of a way around it. I don't have the mathematical brain power that extends that far.

I guess my point really extends to those that using system testing stats to quantify their maxDD. For many years I trades a SPI/HangSeng system that in testing shows 14 losses and x% maxDD. However, using the formula above it suggest that 27 losses were more likely to occur. Funnily enough early last year it did have 27 losses...doh.

So in summary, what is usually seen on the surface will in reality be a lot worse, which is why I tell people to think of a maxDD they think they can handle, then halve their risk.

That said most people can't handle what they think they can anyway, so its best to halve it from the outset.

Remember the idea of successful trading is to never get taken out of the game, monetary wise or psychologically.

Nick

 

[Mr J]

I don't think of max drawdowns, since it is limited only by what we have in the account. All I can do is minimise the chance of experiencing one, and hope that my account has grown enough by that stage to cushion the blow.

 

[wayneL]

This may be more probable in a system where numerous correlated trades are taken. Less probable, but not impossible where a less correlated trades are taken.

Yes, but trade size should be adjusted for correlation for this very reason.
[/QUOTE]
No argument there. I like my risk @ < 2%, for better or worse. But I also think "portfolio heat" is under-emphasized.

However, certainly something I've always designed into my systems is overall portfolio risk. As an extreme example, take 20 2% trades all in one day in the stock market and your risk is actually 40%.

It depends how we define risk. Say we flip coins for $1. I win the first flip, and decide as soon as I lose the dollar that I won, I'll quit. We make 9 more flips as I never lose the original dollar I won. How much have I risked? After the first flip, I only risked $1, and while that dollar was placed at risk 9 times, it was still only that dollar that was at risk.

Two completely different situations... not relevant to my comments at all.

You are talking about sequential risks. I am talking about concurrent risks.

 

[johenmo]

So although it may be possible to succeed with a $3K account, you have a much better chance with a larger account.

I agree but I'd rephrase this to say with a larger account, the brokerage has less effect.

Do the maths with the same trades and with varying account sizes (So larger positions for each trade due to the larger account. Brokerage requiores a larger % return just to cover brokerage (e.g. a $500 vs$1000 vs $10,000 position). Repeat for different brokerage levels.

A larger account with no skills means you may take a bit longer to lose it all (i.e. success has nothing to do with account size but how you handle it). I have a small account and cut the losses (eventually) and am ok now. Not trading until I built a bit more - which I have now done. But I am a little more wise now than I was before.

 

[AlterEgo]

I agree but I'd rephrase this to say with a larger account, the brokerage has less effect.

Yes, that’s what I’m getting at. But to be entirely correct I’d rephrase to say that the larger the position size (which naturally will require a larger account), the less brokerage has an effect (up to a point).

But not only that, a larger account also allows you to spread the risk over a larger number of positions (eg. having 10 positions on a $5,000 account isn’t practical, as the brokerage would kill you), and therefore allows you to use a smaller percentage risk per position if using fixed fractional position sizing. eg. With a larger account you could use say 1% risk per postion, but may have to use say 5% risk per position with a smaller account. For example, you couldn't use 1% risk with a $2,000 account, as 1% of $2,000 is $20, which is probably less than your brokerage costs (total of buy and sell brokerage).