Forex may be chaotic...but some things are still predictable

I was just reading this post over at Trader Rich's Forex Project about a study at MIT that concluded that "treasury bonds are random, the stock market is correlated, and forex is chaotic." Firstly, let me say I haven't actually read the study in question, and probably never will. What I have read is Rich's summary of the study author's summary of his findings in Currency Trader Magazine. Also I've never studied chaos theory, higher math or physics. So obviously I'm extremely well-qualified to comment on this research in a thoughtful, informed, and in-depth manner. So here goes...

While I agree that a lot of forex market activity appears chaotic, and as soon as you come up with a predictive rule the market breaks it, there are still a few things you can predict with some accuracy amid all the chaos. And these are the things that keep me trading forex. Off the top of my head, they are:

Trading range: I can say with reasonable confidence that the trading range of the EUR/USD tomorrow will be somewhere between 30 and 150 pips. Occasionally it may be more, occasionally less. But it will almost certainly not be 500 pips. Or 1000 pips. Nor will it flatline and refuse to move at all. Now a currency whose range varied from 10 to 1000 pips a day on a fairly unpredictable basis...that would be chaotic.

Reaction to certain news events: some events will move the market. Period. What direction, and how many pips, can be hard to prediect. But I can predict with a high degree of confidence that there will continue to be news events that shake things up periodically.

Periodic emergence of trends: very real, very tradeable trends will emerge from the chaos every so often, and in all likelihood will continue to do so. Just looking at a price chart without a single fancy indicator on it can tell you this.

If you place trades long enough you'll eventually get one right. This is the principle behind the Martingale strategy. I'm not saying it's a good strategy to use, but it's based on a statistically valid and predictable observation. Chaos or no chaos, the odds will eventually swing in your favor.

Buying some currency pairs pays you interest. Holding others costs you interest. This is what carry trading is all about.

Someone who routinely takes on too much risk when trading in chaotic conditions probably won't last as long as someone who knows exactly how much risk they can afford to take.

Someone with a clear head can combine observations like these into a trading strategy with decent odds of paying off in the long run. Someone whose outlook is clouded by wishful thinking, impatience, inconsistency, lack of discipline or impulse control, and any other problems of the compulsive gambler doesn't have a chance in the world.

Forex can be boring for long periods. Whether that's chaotic or not I can't really say. But it's certainly predictable.

I'm sure the chaos theorists wouldn't disagree with any of this, and would point out why chaos theory allows for all of these possibilities. But I'm not writing for them - I'm writing for the traders out there like me who see a statement like "forex is chaotic" and think they must be crazy to keep chasing the market if the MIT scientists say it's a giant chaotic whirlpool ready to suck your accounts dry. So if I've made any of you feel a little better, I've done my job. Enjoy the chaos!

You can't change the weather

I just came across this line in a good novel I'm reading, "Brooklyn Follies" by Paul Auster. And it struck me as a worthwhile axiom for forex trading as well as lots of other things in life. The best a trader can hope for is not to take the "weather" in the markets personally, and instead adapt tactics and strategy to the current conditions. Carry an umbrella. Avoid puddles. Try not to get splashed by passing cars. Don't get greedy. Use your leverage wisely. Don't trade when it's really weird and nasty out there. That kind of thing.

This quote also got me curious about weather forecasting and how it might compare with market forecasting. They have more in common than you might think:

"Components of a modern weather forecasting system include:

• Data collection
• Data assimilation
• Numerical weather prediction
• Model output post-processing
• Forecast presentation to end-user"

"During the data assimilation process, information gained from the observations is used in conjunction with a numerical model's most recent forecast for the time that observations were made (since this contains information from previous observations) to produce the meteorological analysis. This is the best estimate of the current state of the atmosphere."

All this is from the Wikipedia article on weather forecasting. Worth a read if you find the weather remotely interesting and want to compare/contrast how its forecasters try to predict the future with all the tools and techniques used by traders. I suspect the meteorologists get it right a lot more often than market analysts.

(OK, maybe you personally can change the weather and just aren't telling anyone. In which case, I'll just hope you stay in a good mood.)

**name that Simpsons episode**

Daily Forex Forecasts

I just came across a site called Forex Predictions which generates daily high/low price forecasts for a number of currencies. So if you're into trading based on daily ranges, an idea I've toyed with periodically, it might be worth a bookmark. My guess is it's applying the average daily range over a certain period (month, maybe?) to the previous day's closing price. Or something like that.

Here's their page with all the daily predictions for currencies like the Euro, Yen, Pound, Swiss Franc, Canadian Dollar, South African Rand, Thai Baht and various others, all calculated vs. the US Dollar:
Daily High/Low Forex Forecasts

Another handy feature is a little box you can install on your site with that day's high and low predictions. For instance, here's the Euro prediction box. Maybe I'll stick it in the sidebar one of these days.

My main quibble with the site is it doesn't provide data on the past performance of their predictions. So I definitely wouldn't take the forecasts at face value - I'd monitor them for a while to see how close they come to the real prices.

How My Forecasts Work

Since I've been tossing out forex predictions without much explanation of their parameters, I thought it was about time to clarify exactly what I'm specifying about the markets when I say "forecast" or "prediction".

My system runs off of daily market data for the EUR/USD and GBP/USD (or Cable) pairs. This data includes the opening, high, low, and closing prices for each pair. If I really set my mind to it I'm sure I could come up with systems for other pairs, but these keep me plenty busy for now. (Actually, I once tried to come up with one for the USD/JPY but it really didn't turn out well. At all.)

Of the Euro and the Pound I only really trade the EUR/USD, with the occasional foray into the Cable for kicks. Eventually I may trade both pairs simultaneously, but only after I build up a larger amount of principal. Cautious money management is the name of the game here.

I created this system because I wanted to be able to trade only once a day, using straightforward Yes or No signals that took under 10 minutes to generate. I did this in an effort to preserve my sanity, and my funds, after attempting to trade 5 minute, 30 minute, and hourly charts. I quickly found that these time intervals were causing me to lose sleep, money, and mental stability.

Anyway. The official time that each forecast goes into effect is 4:00 pm US Pacific Time; I live in the San Francisco Bay Area and so that's when my trading day starts (except on weekends, but I'm not likely to be posting many forecasts on weekends). If you want to convert this to your local time, here's a handy conversion tool. Each prediction is made for the coming 24 hours, and will hence be proved true or false at 4:00 PM the next day. The exception is on Fridays, when market makers close for the weekend at 1:00 or 2:00 PM Pacific Time. I know Global Forex Trading wraps up the week at 2:00 PM my time, and Forex Capital Markets closes at 1:00 PM. Other brokers' closing times may vary.

In practice, I generally post the forecast within a couple hours of making it, if I have time (which I didn't yesterday, as you may have noticed). Since there's usually not a huge amount of market activity in those couple hours, the prediction should remain valid -- however, I make no guarantees, and am only posting these signals for informational and entertainment purposes. Trade at your own risk!

So that's how it works. Yes, but how exactly are these signals constructed in the first place, you may wonder. The short answer is through trial, error, more error, then finally, identification of some promising indicators, followed by a process of layering these indicators while excluding counter-indicators (see my Process of Elimination post), and then, at the end of it all, a whole lot of hoping (often accompanied by some cursing).

That's the short answer. The long answer I'll save for another day.