Increasing lot size vs. trading entire basket

One strategy posted in this forum, Comdoc's Heatmap, is to buy or sell a basket of 10 jpy pairs in one direction and enter opposite trade for 2 usd pairs based on a combination of using the Orest indicator and some technical indicators. Seems to me that if you pick one pair with a low spread that is highly correlated to the jpy pairs and trade 12 times the number of lots, that would be better than entering a trade for each of the 12 pairs because the total spread of 1 pair would be lower and also it would be easier to control the trade of 1 pair vs. 12 pairs. For example, trading 1.2 lots of eurjpy vs. .01 lots for each of the 12 pairs in the basket. If anyone thinks the risk is lower trading the 12 pairs, explain why.

It is not lower, in fact it is higher. What people fail to understand is when you have 12 highly correlated positions (basket of yen pairs) then you really have ONE position with 12 times the risk.(plus 12x the spread) So basically you are risking your normal risk% X 12 for one idea. So instead of trading the basket find one pair that you know well, and if you know them all then pick the currency that you feel should go the opposite way. I.E I think yen will go up in price and I think the Australian dollar should drop, so trade AUD/JPY .
Try to be long one currency and short another. But this is more for trading on fundamentals.

And as far as increasing lot size I wouldn't do it. Anything more than 3% on a single trade and your a gunslinger. Also try to never have more than 10% in the market at the same time. But the last rule is more of a guideline and can be broken because you should never miss an opportunity.

See my post# 92:

http://www.forexfactory.com/showthread.php?p=2350263

The total risk depends on the degree of correlation.

As an example, if you have 10 simultaneous trades, risking 2% of your account on each:
-- if the positions are 100% correlated, the total risk = 20%
-- if the positions are 0% correlated, the total risk = 2%

Read chapter 8 of Dr Ryan Jones' book for a full explanation.

David

Take the T101 strategy, where 14 pairs are traded in one direction. My understanding is that the net effect of them is a hedge - so they would be 0% correlated correct? If so, then are you saying that the risk is 90% greater to trade 14 times the lot size for 1 pair vs. .014 lot size for 14 pairs?

Eagle, if I understand correctly, correlation is the degree of dependency that one instrument appears to exert on another. For example if every time A moves 1 pip upward then B is guaranteed to also move upward 1 pip 100% of the time, then we have 100% correlation. If B will move upward 1 pip 90% of the time, then we have 90% correlation. If B will move downward 1 pip 100% of the time, the the correlation is -100%. If B moves completely independently of A then the correlation is 0%.

I'm not exactly sure what you're asking, but if T101's 14 pairs are a perfectly correlated hedge, then the net movement across all 14 pairs will always be zero, at all times. Hence the return will always be zero and the risk zero, no matter when you enter or exit all 14 positions, provided that you do so simultaneously (assuming that is the suggested strategy). But in the real world of forex I suspect that nothing is perfectly correlated. The fact that correlation must inevitably break down at some point is the reason why hedging schemes like Freedom Rocks are eventually doomed to fail, how quickly depends on the extent to which the trader is over-leveraged.

David