The Contrarian View - A case for Fundamental Analysis in Forex

[FX_Traders_Weekly]

Greetings!

I'm Scott Percival, the publisher of FX Trader's Weekly, and I'm happy to be one of Elite Trader's newest sponsors. I introduced myself and FXTW over in the announcements forum yesterday, so I won't go through all that here (ok, I guess I can't help but mention that FXTW just published its 100th issue - yay!).

So tonight I'm popping into the Forex forum to start a more substantial (and most likely controversial) conversation on Forex trading philosophy. I'm talking about the two perennial debates that pervade the trading world, Forex and otherwise. These are the Technical Analysis (TA) vs. Fundamental Analysis (FA) debate and the Trend Following vs. Contrarian debate.

I'm willing to bet that most of the members here are primarily users of TA who subscribe to the common wisdom that "the trend is your friend." But I'm here to (respectfully!) challenge both approaches. FX Trader's Weekly is built around a contrarian view of the FX market, and mainly uses fundamental information to implement that.

I'm going to put the TLDR bullet point summary of my argument for this approach here, but for a more detailed explanation you can check out two of my articles:

How to see what other FX traders don’t see…
Why the trend may not be your friend…

Those are a couple of fairly long pieces, and I know people are busy, so here's the argument in a nutshell:

1. Trend following works well in some environments, the equity markets being a prime example.
2. The broad equity market has an upward bias because it tracks economic growth driven by technological progress.
3. Individual equities trend as a result of how they're managed - for good or bad.
4. But Forex is different. There are forces that tend to cause reversion to the mean.
5. These are classical economic forces, political forces, and central bank interventions.
6. Trends take time to identify in any market, including Forex.
7. So in Forex, visible trends are often close to ending.
8. Thus, we're contrarian.
9. As contrarians, we must identify when price has deviated from value.
10. Price is the province of TA, but value is the province of FA.
11. Thus, our primary tool as contrarians is Fundamental Analysis.
12. Quod Erat Demonstrandum!

I'm currently only available to post for an hour or two each evening. So now that I've waltzed in here and thrown down the gauntlet, I'll bid you all adieu until tomorrow evening, and waltz right back out! (I can't actually waltz, btw.)

Keep pipping up!

 

[easymon1]

But Forex is different. There are forces that tend to cause reversion to the mean.

Does apply equally to currency futures like jpy, gbp, aud?

 

[MKTrader]

You can make it as fancy and complicated as you want...all I say is show me the money. Where are your audited account statements or third-party verified performance reports? That's worth more than 100 exec summaries and research reports.

 

[FX_Traders_Weekly]

Does apply equally to currency futures like jpy, gbp, aud?

Short answer: Insofar as the contract price action closely tracks the spot in higher time frames, then yes. The image below is the EUR/USD compared to the September contract (light blue). See how closely they correlate. So if something's been driving the EUR down for the past week or so, then the futures contracts are most likely falling as well.

Longer Answer:

That being said, I realize there are several additional factors when it comes to derivatives like futures (which I've never traded btw - only options). Derivative price movements are also affected by inputs like the the time remaining in the contract, contract price, volatility of the underlying, etc. Also, if you're trading intraday, or even with a somewhat longer horizon like a few days, then the price action you're concerned with is likely to be driven more by technical factors than fundies.

The reversion to the mean behavior I'm talking about in those articles is driven by fundamental forces which come into play over weeks to months.

Now, just to throw another wrench into the works, something I didn't mention in my post or in those articles are a bunch of research studies I've done over the past couple of decades into price behavior. These involved time frames from daily and weekly bars down to <1Min bars. I remember being struck by how often reversion to the mean seemed to win out over trending behavior in just about every time frame where I tested for each. So even down in the scalping world where TA is everything and FA is irrelevant, you still tend to get reversion.

There's a free ebook on my site (Weekly Price Action Statistics) that details one of those research projects using weekly bars, but that just scratches the surface. I don't know if I can still dig up my old notes on some of the other studies. If there's any interest in this subject out there, I'll give it a try.

Hope this helps!

 

[easymon1]

Short answer: Insofar as the contract price action closely tracks the spot in higher time frames, then yes. The image below is the EUR/USD compared to the September contract (light blue). See how closely they correlate. So if something's been driving the EUR down for the past week or so, then the futures contracts are most likely falling as well.

Longer Answer:

That being said, I realize there are several additional factors when it comes to derivatives like futures (which I've never traded btw - only options). Derivative price movements are also affected by inputs like the the time remaining in the contract, contract price, volatility of the underlying, etc. Also, if you're trading intraday, or even with a somewhat longer horizon like a few days, then the price action you're concerned with is likely to be driven more by technical factors than fundies.

The reversion to the mean behavior I'm talking about in those articles is driven by fundamental forces which come into play over weeks to months.

Now, just to throw another wrench into the works, something I didn't mention in my post or in those articles are a bunch of research studies I've done over the past couple of decades into price behavior. These involved time frames from daily and weekly bars down to <1Min bars. I remember being struck by how often reversion to the mean seemed to win out over trending behavior in just about every time frame where I tested for each. So even down in the scalping world where TA is everything and FA is irrelevant, you still tend to get reversion.

There's a free ebook on my site (Weekly Price Action Statistics) that details one of those research projects using weekly bars, but that just scratches the surface. I don't know if I can still dig up my old notes on some of the other studies. If there's any interest in this subject out there, I'll give it a try.

Hope this helps!

View attachment 262944

Very refreshing to see such a well done response.
Thank You

For the chart above, which is the blue line - Futs?

 

[FX_Traders_Weekly]

You can make it as fancy and complicated as you want...all I say is show me the money. Where are your audited account statements or third-party verified performance reports? That's worth more than 100 exec summaries and research reports.

Sir? SIR???!!! This is a Wendy's.

Seriously though...sir? SIR??!! This is a newspaper. Except that we don't cut down a lot of trees or squeeze baby octopi for their ink. Also, it's a very highly specialized and kind of geeky newspaper. But it's a newspaper. FX Trader's Weekly takes a bunch of price and news data generated over the week, runs it through some algorithms, and publishes the results in easy to digest visual form.

Would you call up the Wall Street Journal and ask for their audited account statement? They'd be like, "Ummm...WHAT account? Do we know you? Are you an attorney? Is this a prank call?"

Ok, I'm being slightly dishonest with that last point. There actually IS an account - the FXTW Model Account, the equity curve of which is on page 2 each week. But as explained in that linked page, the Model Account uses (kind of a dumb) mechanical average of the five key metrics tracked by FXTW. This forms a sort of "index" of the performance of those metrics, but I wouldn't recommend anyone actually trade like the Model Account does. To quote myself:

"However, some indicators in the FX Trader’s Weekly may work better than others at various times, so traders should use them as a tool kit, tailoring their trades based on their own experience and knowledge.


Think of the model account as simply a “market index” of the average performance of the five key metrics. By doing that, you may find that you can outperform the mainly mechanical model account."

Anyway, I actually get where you're coming from, because I've been that guy on other forums that always challenges the new "guru" who shows up touting some holy grail or another. They gain a following with big promises and vague statements about teaching the "real secret to the markets" or some such nonsense, and then of course comes the pitch for the paid chat room or trading course or whatever.

This isn't that. This is a Wend... ahem, I mean...this is a newspaper. The "audited statements" are the almost two years of past issues that you can see for free here. Anyone can go look through those and decide if FX Trader's Weekly is for them or not.

Hope this helps, and as always...keep pipping up!

 

[FX_Traders_Weekly]

Very refreshing to see such a well done response.
Thank You

For the chart above, which is the blue line - Futs?

Thank you for your kind words! Yes the blue line is the September-21 futures contract (I think that's the one I chose). I just went to Bar Charts and plotted the two together, which is a cool feature they have.

 

[FX_Traders_Weekly]

...annnd, I just realized I haven't had dinner yet. I'm off for the night! Back tomorrow.

 

[Tradex]

Trend following works well in some environments, the equity markets being a prime example.

Famous systematic trader and author Perry Kaufman proved via backtests that the equity market is mean-reverting in nature (including individual stocks), and responds better to overbought/oversold indicators like RSI.

On the other hand the Forex and the interest rates markets are the best trending markets in the world. So moving averages, MACD and Donchian breakouts (for instance) are more suited for these markets.

And yet you are basically saying that the exact opposite is true.

Care to elaborate a little bit more on this, please?

Thanks.

 

[FX_Traders_Weekly]

Famous systematic trader and author Perry Kaufman proved via backtests that the equity market is mean-reverting in nature (including individual stocks), and responds better to overbought/oversold indicators like RSI.

On the other hand the Forex and the interest rates markets are the best trending markets in the world. So moving averages, MACD and Donchian breakouts (for instance) are more suited for these markets.

And yet you are basically saying that the exact opposite is true.

Care to elaborate a little bit more on this, please?

Thanks.

Fair enough!

I'm sure we could find studies and books supporting both views, as well as a third one - remember Burton Malkiel? He would disagree with both me and with Kaufman. He wrote the famous book, A Random Walk Down Wall Street, which introduced the Efficient Market Hypothesis. By that view, neither a mean reversion or trend following strategy can work in any market because prices are essentially random. Of course I disagree with that view, as I'm sure Kaufman and most traders do as well.

So we're all in our various camps with our various studies and books that purport to support our outlook. And it will always be that way. The cool thing is there are ways to succeed with either approach. Jack Schwager, the author of the Market Wizards series of books said, "There are a million ways to make money in the markets. The irony is that they are all very difficult to find."

I'm certainly not going to beat up on Mr. Kaufman and try to "debunk" his findings. For all I know, his research methods are perfectly sound, and probably show what he says they do. But different sets of data will produce different results.

So all I can do is point you to the work I did, with the data that I used. I'll also point out that my conclusions aren't just based on the data analysis, but are grounded in a set of logical arguments about what's fundamentally going on in each market. So I have not only the results, but an explanation for the results as well.

I'll deal with the broad equity markets very briefly. Look at any chart of the DJIA from 1790 (I know it wasn't a thing until the late 19th century, but analysts have been able to backdate it) until now. That's clearly an upwardly biased market. And as I said in the first post, there's a logical reason for that - the equity markets are actually charting the upward progress of human technological civilization and its creation of wealth.

As far as the Forex market goes, I'll point you to a couple of my books - Beyond The Big Mac and Weekly Price Action Statistics. Both describe research that supports the mean reversion view in Forex, and the first one provides a detailed explanation for why that should be the case. Each one provides full details on the datasets, so anyone could duplicate the research for themselves.

Hope this helps, and as always, keep pipping up!