Money management

[Grey1]

Money managment

Lets say Joe, had £5000 and decided to trade the market to buy a private jet..

Also lets say he had a trading strategy with 50% success rate ( coin flipping strategy ) and he never had to pay any commission not even a penny ... Joe is a simple man and likes to keep things simple so he risks 10% of his capital in every trade and takes profit after 10% gain .. ( Reward / Risk =1 )

After 200 trades what is going to happen to Joe

1) He buys his jet
2) He can not afford to buy his jet but he makes profit
3) He goes bankrupt and gets a job in his local Pub

What you think is going to happen to Joe ?




Iraj

 

[ChartMan]

Sadly, he loses the lot

 

[Oldun]

Or lots & lots more in a margin account

 

[stevem12]

If he always uses 10% of his bank, adjusting it for each winner and loser he`ll never go broke. If he uses 10% of starting bank he`ll go broke on first losing run of ten, this WILL happen.

If he bets 10% and takes profit at 10% and has a winning % of 50% he will never ever make any money. 50% winners at even money = breakeven.

Steve

 

[bonsai]

if he risks 10% of his capital on every trade, he is a fool
and a fool and his money are soon parted.

And so if he applied for a job in my pub, he wouldnt get it.
It's a cash handling business and you don't want fools
handling your money.

Poor old Joe.

 

[Grey1]

Bonsai,
what if he was not a fool and only risked 2% of his capital in each trade and took profit at 2% ..

Would he eventually make money or lose money ?
Would you give him a job in your pub if he risked 2% instead of 10% ?


iraj

 

[stevem12]

grey1

Think about the risk/reward, how can you make any money with a ratio of1/1 with 50% winners. It doesn`t matter what % you risk. A lower % reduces chance of going bust thats all. 40% wins with a ratio of 2/1 or higher risking 1-2% shows profit!

Steve

 

[ChartMan]

Again, sadly, he lets his losers run until he can stand the pain no longer.

 

[Trader333]

Consecutive Losing Trades

The formula for probable consecutive losing (or winning) trades based on a known percentage of success is as follows:

C = Ln(X)/-Ln(Y)

Where C = Consecutive Losing trades
X = Number of trades made
Y = Losing percentage of trades

Attached is an Excel Spreadsheet that will do it automatically for you and as Chartman says with that level of risk per trade you will lose the lot.


Cheers


Paul

 

[jpwone]

He cannot lose his complete initial stake if the stake used for each trade is a fixed percentage of his current balance. So, he wont go bankrupt (option 3).

The probability of making a profit is the same as that for making a loss. He has an equal chance of making a profit as he has of making a loss for any one trade Whether he makes a profit overall will depend upon the length of the win/loss runs in the 200 trade sample and the actual randomness of the wins to losses. If the win/loss per trade is random with each trade having a 50/50 probabilty of a win then it is possible to have more wins than losses (and vice versa) in the 200 trade sample.

Only with exceptional luck would he buy a jet (Option 1).

Making a loss is as likely as making a profit (Option 2).

 

[bonsai]

Grey1
lol
sorry but now I suspect he is still a fool for thinking he can live long enough to buy a jet if he only takes 2 % profit on winning trades with a starting trade of £100.

How much do these things cost anyway?

I know plenty of 747's are sitting idle in the desert but even so ?
A second hand Lear would still be out of reach.

PS: Don't forget also that when he gets down to rock bottom
the minimum number of shares you can buy is 1.
On that losing trade, he cannot re-enter ?.

 

[Grey1]

Joe will evantually lose all his money even though his is betting equal amount of his toal capital

Conclusion :--

1) if your trading strategy is 50/50 or less do not trade irrespective of how much you betting ..

2) donot not close your position unless your reward/risk is > 1
( lets say ur stoploss is 10 points on DOW including the spread, then you must let your trade run to justify the risk... if you feel you can not get a greater than 1 ratio DO NOT TAKE the trade..

3) improve on you entry to reduce the risk ( many technicians say entry is not all that important it is the exit that counts... This is not correct at all..)


iraj

 

[ChartMan]

Grey.You are absolutely right about 3)

 

[FTSE Beater]

Hi all

Interesting discussion.

The reason Joe would lose all his money is this. If he lost, won, lost, won, lost and then won he would have £4851.50

£5000 - 10% = £4500
£4500 + 10% = £4950
£4950 -10% = £4455
£4455 +10% = £4900.50
£4900.50 -10% = £4410.45
£4410.45 +10% = £4851.495

So on 6 trades he would lose £148.50


Grey1 wrote

1) if your trading strategy is 50/50 or less do not trade irrespective of how much you betting ..

I would disagree with this. If you have a 3:1 Reward / Risk ratio every time, then you would make money. Trading and money management comes down to 3 factors. Win / Loss Ratio, AVG Win and AVG Loss

2) donot not close your position unless your reward/risk is > 1
( lets say ur stoploss is 10 points on DOW including the spread, then you must let your trade run to justify the risk... if you feel you can not get a greater than 1 ratio DO NOT TAKE the trade..

I have to disagree with this. You could quite easily enter a trade on a good basis, only to find the market turns against you - YOU SHOULD CLOSE, and not let your stop-loss get hit just because your target hasn't been hit. Trading always requires some flexibility.

3) improve on you entry to reduce the risk ( many technicians say entry is not all that important it is the exit that counts... This is not correct at all..)

I agree with you on this - although the exit and entry both have to be good. One cannot work without the other.

Just my thoughts

 

[fowkesp]

If you have the trading stats (win% and win:loss ratio), this site is always worth a visit to give you a picture of the likely outcome http://www.hquotes.com/tradehard/simulator.html

If you put a figure of 100 in the Lines Qty box, you get a good idea of the likely range of equity curves (it uses a Monte Carlo simulation).

This will show that it is quite possible to have a winning system with very low win%, as long as the win:loss ratio is high enough to compensate. The drawdowns get ugly though

 

[jpwone]

To demonstate the points made by all, some of which I agree with and some which I disagree with , I have put together the attached spreadsheet.

The purpose is to allow you enter a starting capital sum, a win/loss ratio, an amount risked as a percentage of capital on each losing trade and an amount returned as a percentage for each winning trade.

It calculates the return for the 200 trades based on the win/loss ratio for each trade ie. each trade is treated as a new trade and the likelihood of a win or loss is based on the win/loss ratio. By doing this it is possible to have losing and winning runs which will affect the result.

Pressing 'F9' will force a recalculation of the trades.

Things to note.

1. If you are lucky even with the high stake values used in the original example it is possible to make a positive return. it is unusual but it is possible In one run it actually generated £151k, obviously rare and unlikely to occur very often.

2. Your capital never gets down to zero unless you set the risk to 100 percent of capital. Edit: Should have been clearer on this bit - it gets down to zero if the the risk is significnatly greater than the return. Capital can get down into the pennies very quickly depending on risk and reward setting.

As a general tool the spreadsheet would be useful for demonstarting to the novice why low starting capital rarely generates large returns and that a significant loss is the expected result when losses are allowed to run.

 

[FTSE Beater]

Hiya JP

A superb spreadsheet

 

[ChartMan]

The Hquotes is excellent, JP's is better.
How about doing a DOW specific one JP?
Bet size per point starting at £1
Maximum loss in points
Average gain in points.
Trades required to reach 300%, at which point one should be ready to double their size...
Now that would be a classic. I'd do it, but I'm a dunce when it comes to things like that

 

[Grey1]

FTSE beater..

Both arguments 1,2 was carefully worded... The risk reward in the example above was 1 ( this was clearly mentioned) and I did not mention other alternatives of Reward/Risk.. if you used higher risk reward, things would be totally different..
I am continuously seeing traders to take a position and close a trade with much less than their risk level).. This is totally un acceptable and mathematically risk of ruin is increased to an intolerable levels.. ONE MUST NOT INTITALE A TRADE UNLESS THERE IS AT LEAST HIGHER THAN 1 REWARD/RISK...

2) i said, Do not initiate a trade unless your Reward level is more than you risk level to be profitable in long term .. ( in 50/50 situation ).. Again there is a mathematical truth in this statement and i have simulated this argument using Monte Carlo technique ( see the attachments)

The above arguments are mathematically sound...

IRAJ

 

[jpwone]

Thanks FB and CM

Not too difficult to modify it to use points rather than percentage of capital and base the return on a fixed stake size. However, as the number of trades would vary dependent upon the win/lose history to reach 300% capital gain it is not ideally suited to a spreadsheet. It can be done in a spreadsheet and the easiest way is just to put enough lines in to get to 300% (or a user defined target) to cover this. It is probably easier to do it as a Javascript or VBscript application embedded in an HTML wrapper or as a standalone Visual Basic application. I'll have a look at doing it as a HTML wrapped script tomorrow as this will probably be easier for people to use as there will be no dependencies as there would be if done in Visual Basic.