I really like reading the papers that Bridgewater publishes. I read somewhere also from Ray Dalio that he aimed to have “Seven, high quality, uncorellated bets”. This has always stuck with me for some reason, and I think about it every now and then as I’m learning to trade forex.
I also think about my account in terms of “capacity”. For example, I have decided to never risk more than 1% per trade, and never exceed a true leverage of 5:1 . This gives my account a certain working capacity which is available whenever markets are open.
I am trying to exercise some forward thinking here, and its early days so maybe it doesn’t matter too much yet, but what I would like to do is expand my “playbook”, so that I have another style of uncorrelated bet I can make to keep my account operating at as close to capacity as possible for as much of the time as possible. So far example, my current playbook looks like this:
A forex system, swing trading majors and minors on the daily charts.
The same forex system, operating on the weekly charts, on other pairs.
nothing yet (trading the DAX or other index on a different time frame?)
nothing yet (trading a commodity on a different time frame?)
nothing yet
nothing yet
nothing yet
nothing yet
What I”m saying here is that I wish I could find a second, uncorrelated bet that was available to me from time to time. I know correlations are tricky, and if you told me the DAX was correlated to the Euro somehow I wouldn’t be surprised. But rules could be made for it: for example a DAX trade can’t be made if a EUR long/short is also open.
Does stepping up and down timeframes help to achieve decorellation? I feel intuitively it might.
I hopefully envision a future where I am up and running in my forex live account, while also backtesting and demo testing my next 1 or 2 complimentary techniques. Is this how other people do it? Please forgive me if I’m getting ahead of myself, I just like to plan.
Just out of curiosity, have you tried using a correlation calculator?
You might want to do a web search and try different ones out. I’m skeptical of their accuracy, but it may at least give you a baseline to follow.
By the way, there is no shame in trading one pair that you really like. In fact, I know of several professional traders that trade only one instrument. The grand daddy of course is the E-mini S&P500 (ES). But there a lot of traders who like just trading one currency pair. Me included.
I saw a Rayner Teo clip a while ago where he wanted to explore MA crossover strategies. He found they can work if you use a large number of markets and you might find some of his selections useful -
That was very interesting, thank you. I will go through and watch the rest of his videos also.
I read once of his that trade frequency is a variable people often overlook- that you might have a system with a great risk reward and great win/loss, but if you only get 1 signal from it a year, you might still be going nowhere. I think this video says a similar thing in a slightly different way.
Its interesting the groupings too. I suppose there is also a larger question, of what is the utility of uncorrelation (as opposed to correlation)? It seems to me twofold.
To stop you getting smashed by accidentally doubling down on one risk.
To allow you to find more trade opportunities and generate a high trade frequency.
Anyway anyone else who might be reading this you should know I am a newb demo trader, so careful if you listen to my musings! Thanks again @tommor
I haven’t no, but I will now that you have mentioned it.
TBH I found the correlation section of the babypips course sometimes crucial but often befuddling. Knowing about the CHF/GOLD relationship, and the CAD/OIL relationship, and relationships like AUD/NZD were just magnificent to find out about. But the correlation table itself I found deeply confusing. The timeframes over which everything shifts, and the shifting nature of the correlations, left me pretty uncertain about what to make of it all.
I came away from it having learned a bunch of things to watch out for and avoid, but not feeling like I had learned anything that I could use to actively press an advantage.
and in an effort to educate myself I’ve been going through it. I expected to at the very least see some kind of smooth and meaningful relationship between AUD/NZD. But far from it it seems.
I’m thinking it makes sense to filter out most of the tables that show correlations below one month timeframes, as it could just be coincidence/noise. I think I just need to spend more time with it, chipping away at my own ignorance.
Interestingly it seems alot of those european exotics are very tightly correlated across all timeframes with the euro. Makes sense I guess.
Also it seems like Silver doesn’t have very strong correlations with anything.