How much to buy w/2% risk of $1k and w/50pip SL? - Page 5

MickMason · 30-12-2005, 03:37

Quote:

Originally Posted by kenazen

1) Why don't I have to account for leverage?

Why would you need to, what difference would it make? You're basing trade size on the value of a stop loss, and stop loss based on 2% of equity, leverage would be of little consequence or interest.

Mick

kenazen · 30-12-2005, 05:03

Quote:

Originally Posted by MickMason

Still number crunching kenazen

Ok, try this, I didn't realise you were trying to write a formula.....

USD/? : .0001 divided by the rate = pip value ie USD/CHF .0001/1.3100 = .000076 (or in the case of USD/JPY .01 as it only has two decimal places)

?/USD : .0001 divided by the rate times the rate ie EUR/USD .0001/1.1850 = .000084 * 1.1850 = .00001

?/? : .0001 divided by the rate = pip value, times the USD rate = USD pip value ie GBP/CHF .0001/2.2636 = .000044 * 1.7271 = .000076

Does that help?

Mick

Hmm.. What is this telling me? How can I use this forumla to determine how many units to buy?

I'm basically trying to create my own version of what Ampris2000 is doing. I'm not sure if he worked out certain things, or stumbled upon them, but there are several things he is doing that I think are requirements for a successful methodology. The primary being, you have to make a lot of little bets instead of a few big bets. He doesn't (formally) account for volatility or correlated pairs, but as a little guy, he/we probably don't have to.

He is also betting on continuation rather than direction, which is another thing I agree with. He uses stacked MA's, but his method of anticipating continuation probally doesn't even matter, that he cuts his losses and lets his winners run - does. He could probably make random entries and do okay.

The key for me, is to make same-sized bets, though I may add to a position after it becomes profitable. That these currencies are sometimes not expressed in USD, makes it tricky to figure out quickly how big a bet I should make. One thing I do not want to do, is make a larger bet than I should.

I want to be able to backtest these ideas, which is why I need to be able to calculate them exactly. I wouldn't bet the farm on backtesting, but it should turn up anything obviously out-of-whack.

I'm new to forex, so I'm using demo accounts right now. I still don't have a feel for Forex (though I'm practicing everyday). I'll consider trading real money after my demo account is consistently profitable.

Back to Ampris, I think he has pinned down the two most important questions: 1) How much money am I willing to lose, and 2) How many pips may I need to give up. I'm surprised the answers to these questions are not engraved somewhere. For example, the Aonda function of setting your SL based on % NAV you are willing to lose, is backwards. Your stoploss should be determined upfront, with position size based upon it, not the other way around.

As for correlations, I suspect Ampris is more than aware if he is making correlated bets. For example, he prolly wouldn't take 5 trades that are all the same bet of "the USD will go down" (and neither would I).

kenazen · 30-12-2005, 05:08

Quote:

Originally Posted by MickMason

Why would you need to, what difference would it make? You're basing trade size on the value of a stop loss, and stop loss based on 2% of equity, leverage would be of little consequence or interest.

Mick

If I have $1000, and 20:1 leverage, I thought that means I can buy $20,000 worth of currencies.

A $20,000 position that loses 1% if its value is down $200.

If I have 10:1 leverage, I can buy $10,000 worth of currencies. A 1% loss would be $100.

In one case, a 1% loss cost me $200, and the other it cost me $100.

If I am buying a number of units based how many dollars I am willing to lose, it would seem leverage would matter.

Actually, I don't understand why everyone says it doesn't matter. It doesn't make sense that it doesn't matter.

My formulas use a number called "Leverage", but I just realized it should be calling % Willing to Lose (2%). The amount should be $40 intead of the $100 in my examples. So not even my own formulas use Leverage. At least my math agrees with everyone, even if I don't. Beer is not helping either.

MickMason · 30-12-2005, 05:12

Quote:

Originally Posted by kenazen

Hmm.. What is this telling me?

Huh

It's telling you the value of a pip, it's what you said you wanted to know....

Mick

MickMason · 30-12-2005, 05:14

Quote:

Originally Posted by kenazen

Actually, I don't understand why everyone says it doesn't matter. It doesn't make sense that it doesn't matter.

Beer is not helping either.

Have a decaf or six and re-read the thread

Mick

kenazen · 30-12-2005, 05:25

Quote:

Originally Posted by MickMason

Have a decaf or six and re-read the thread

Mick

Then how does leverage effect how much money you win/lose?

kenazen · 30-12-2005, 05:28

Quote:

Originally Posted by MickMason

Why would you need to, what difference would it make? You're basing trade size on the value of a stop loss, and stop loss based on 2% of equity, leverage would be of little consequence or interest.

Mick

I thought the amount of equity risked would need to account for leverage. That doesn't make sense?

kenazen · 30-12-2005, 05:32

Quote:

Originally Posted by MickMason

Huh

It's telling you the value of a pip, it's what you said you wanted to know....

Mick

I'm trying to figure out how many units to buy, and I have been under the assumption that leverage would somehow effect how many untis I could buy.

It would be simple to say I a pip is worth X, so I am willing to lose 50x per trade, but if I am using leverage - I thoguht the leverage would multiply the risk.

If I buy stocks on margin, it changes the number of shares I can buy. It doesn't work the same way with Forex?