Wednesday, February 3, 2010

The Best Pair To Trade

What pairs? I never focused exclusively on specific pairs. I traded what I perceived at that time to be "trending" pairs. I remember doing some work on gbpjpy eurjpy eurgbp nzdusd and also on the emerging pair usdzar. If you ask my opinion right now I would favour USD crosses as the rest of the world seems to be lagging the last few rate hikes in the US and the US is becoming a bit stagnant at the moment. I think emerging market cross US offer some sustainable possible trends at the moment. (Disclaimer: trade whatever you like this is not intended as trading advice but merely an observation).

This question of "what pairs" is funny because I have seen q's on this forum asking exactly that on so many "systems" it becomes laughable same with the question "what timeframe".

The truth is - it doesn't matter. Today the gbp will have a personality the same as what the jpy had 10 years ago or 1 year ago. Things change over time you cannot ever build a system around a currency and then say - "but it doesn't work on the usdjpy" that is not true - it is not that it doesn't work on the usdjpy the truth is that it doesn't work on the market characteristics exhibited by the usdjpy at that moment and as soon as your pair that it does work on changes its personality to usdjpy "characteristics" your system is worthless. Thats why so many systems start out great and have many followers in the beginning but over time it fails. The markets change and thats the simple truth.

If you think about some of your profitable trades - no matter what currency or timeframe - how was your money made? You entered and the market moved in your direction - the more the market moved in your direction the more you made - if you trade in the direction of the trend you make money NO MATTER WHAT CURRENCY OR TIMEFRAME. That is also one of the reasons why the shorter timeframes are so difficult to trade - nobody really knows what will happen from morning to evening but if the current expectations persist then the gbp will close higher on 31 December than it is now. So it is easier to go long the dips on a larger timeframe because you back your decision with a long term trend.

If it was possible to accurately predict an up or down day then and only then something shorter than the daily can become profitable on a sustainable level. I am excluding scalping a few pips on the 5minute chart from this assessment because I believe the market becomes more random the shorter the timeframe and even in randomness there is order that can be exploited. The discipline and "luck" needed on the 5minute chart is however not worth the health risk - but it can be done.

Also having the long term trends at your back is much less stressfull than trying to buck the trend with a clever reversal trade - I have had many of these clever trades and they just cost me money. The market makes you think that this is the real reversal but just as soon as you take the trade she turns around and gives you a hiding.

Maybe the easier way to explain the strength of a longer timeframe trend is to explain about Paris Hilton, she is a celebrity at the moment and many people know her name today. How many people will know her name in 50 years from today? She is the short term trend. How many people will know who is Johann Sebastian Bach in 50 years from today? He is the long term trend! If you have to take a bet on who the most people in the world will know in 50 years from today who will you bet on? Yes, my friend, the long term trend will win every time so think about that next time you want to go in the direction of a short term pullback from a new high...

Of course during war (and other extended hardships) people will lose the trend completely. Go and ask someone in Angola who is Bach and not many will know and due to the media coverage of Paris Hilton more will know who she is, so the short term trend may persist in the short term but even in Angola in 50 years time more people will know who Bach is than people knowing who Paris Hilton WAS.

No comments: