Why has the turtle trend-following system stop working?

[Trend Following]

Long term trend following has lousy metrics (e.g. sharpe ratios...

Great example of a misunderstanding. David Harding had nice insight on Sharpe confusion:

 

[Mtrader]

Computing has assisted trend following traders as far back as the 1970s.

I agree, but in the 70's, only the happy few had computers, the mass did not have one. So it was easy for these happy few to make profits and make the "computerless" lose money. Today even traders who have only a 100$ account have more powerful computers then the happy few had in 1970. The game changed completely.

It is like a war. In 1970 a few soldiers had guns and the mass had only their bear hands. It was easy to win a war against the mass. Today everybody can buy more powerful weapons, so today it is much more difficult to win a war. There is much more (exponentially more) resistances from the mass who were the weak ones in 1970.

 

[Trend Following]

I agree, but in the 70's, only the happy few had computers, the mass did not have one. So it was easy for these happy few to make profits and make the "computerless" lose money. Today even traders who have only a 100$ account have more powerful computers then the happy few had in 1970. The game changed completely.

It is like a war. In 1970 a few soldiers had guns and the mass had only their bear hands. It was easy to win a war against the mass. Today everybody can buy more powerful weapons, so today it is much more difficult to win a war. There is much more (exponentially more) resistances from the mass who were the weak ones in 1970.

If you are talking HFT, options models, very short term, point taken.

But TF is not intensive. It's not about the latest and greatest software/hardware. David Harding spoke at event in the last few years and was talking EXCEL. TF is more about organization.

Additionally, all of the software/hardware is out there, it's known, and evidence that it has refuted Daniel Kahneman's work is non-existent. Markets are not efficient. People are people.

 

[Mtrader]

Long term trend following has lousy metrics (e.g. sharpe ratios, sortino ratio, calmar ratio, Schwager's Pain/Gain ratio). Paltry returns and huge drawdowns.

Exactly.
Some examples:

• Between 2000 and 2013 some trendfollowers made no money at all. So they were invested 13 years for no return and still losing inflation. And for that they had to take risk.
• Between 2007 and 2013 some trendfollowers made no money at all. So they were invested 5 years for no return and still losing inflation. And for that they had to take risk.

I know that the examples are a bit extreme, but 13 years or 5 years only losing money in trendfollowing seems to me extreme too.

These are the maths:

What I see too is that what happened since 2000 confirmed my opinion that it is not impossible that we will have in future a different (not so favourable evolution for the Buffet type of investing) behaviour of the market. Before 2000 nobody would have believed that you could stay for years in the market without making any money. But it happend, and not once but twice in a very short time period. So the strategy of buy and hold because eventually the market will go to new highs is not for 100% garanteed. Since 2000 we also see drawdowns that we never ever experienced before. It has been proven twice now that even if the markets goes only up for 50 years, it will stay like that forever. And this proof came in a time period of only 10 years.
As a good soldier I protect myself against this. Never underestimate the enemy, which is in this case the market.

I belief in trendfollowing, but I would add the word "dynamical". Trendfollowing needs some management, not just passive investing. You should constantly monitor your position, and get out before huge drawdowns. If not, it is not impossible that one day you will meet the mother of all losses.

 

[Mtrader]

Markets are not efficient.

That's why you need a computer. If markets would be efficient you would not need a computer. But there would be not much trading left.

 

[countercountertrend]

Exactly.
Some examples:

• Between 2000 and 2013 some trendfollowers made no money at all. So they were invested 13 years for no return and still losing inflation. And for that they had to take risk.
• Between 2007 and 2013 some trendfollowers made no money at all. So they were invested 5 years for no return and still losing inflation. And for that they had to take risk.

I know that the examples are a bit extreme, but 13 years or 5 years only losing money in trendfollowing seems to me extreme too.

These are the maths:
View attachment 155720

What I see too is that what happened since 2000 confirmed my opinion that it is not impossible that we will have in future a different (not so favourable evolution for the Buffet type of investing) behaviour of the market. Before 2000 nobody would have believed that you could stay for years in the market without making any money. But it happend, and not once but twice in a very short time period. So the strategy of buy and hold because eventually the market will go to new highs is not for 100% garanteed. Since 2000 we also see drawdowns that we never ever experienced before. It has been proven twice now that even if the markets goes only up for 50 years, it will stay like that forever. And this proof came in a time period of only 10 years.
As a good soldier I protect myself against this. Never underestimate the enemy, which is in this case the market.

I belief in trendfollowing, but I would add the word "dynamical". Trendfollowing needs some management, not just passive investing. You should constantly monitor your position, and get out before huge drawdowns. If not, it is not impossible that one day you will meet the mother of all losses.

Markets trend around 15% -20% of the time. Applying a trend strategy to a rangebound market is stupid.

 

[Trend Following]

You should constantly monitor your position, and get out before huge drawdowns. If not, it is not impossible that one day you will meet the mother of all losses.

Trend following DDs result from many small losses across many markets adding up to a larger loss or DD. The DD is a direct result of taking many small losses. It's risk management. So in TF you are constantly getting out before the mother of all losses.

 

[Trend Following]

Markets trend around 15% -20% of the time. Applying a trend strategy to a rangebound market is stupid.

Ah, but who has that perfect crystal ball that allows the timing of the back and forth between trend and flat? Never seen that evidence. If you are talking another strategy, ok. TF, no way. TF has losses. It has DDs.

 

[Trend Following]

That's why you need a computer. If markets would be efficient you would not need a computer. But there would be not much trading left.

No one said you don't need one, but it's just a tool. A rather dumb, precise tool that follows human instruction. Daniel Dennett in his work 'Intuition Pumps And Other Tools for Thinking' makes the point:

The idea that an algorithm is a foolproof and somehow “mechanical” procedure has been present for centuries, but it was the pioneering work of Alan Turing, Kurt Gödel, and Alonzo Church in the 1930s that more or less fixed our current understanding of the term. Three key features of algorithms will be important to us, and each is somewhat difficult to define. (1) Substrate neutrality: The procedure for long division works equally well with pencil or pen, paper or parchment, neon lights or skywriting, using any symbol system you like. The power of the procedure is due to its logical structure, not the causal powers of the materials used in the instantiation, just so long as those causal powers permit the prescribed steps to be followed exactly. (2) Underlying mindlessness: Although the overall design of the procedure may be brilliant, or yield brilliant results, each constituent step, and the transition between steps, is utterly simple. How simple? Simple enough for a dutiful idiot— or for a straightforward mechanical device— to perform. The standard textbook analogy notes that algorithms are recipes of sorts, designed to be followed by novice cooks. A recipe book written for great chefs might include the phrase “poach the fish in a suitable wine until almost done,” but an algorithm for the same process might begin “choose a white wine that says ‘dry’ on the label; take a corkscrew and open the bottle; pour an inch of wine in the bottom of a pan; turn the burner under the pan on high...”--a tedious breakdown of the process into dead-simple steps, requiring no wise decisions or delicate judgments or intuitions on the part of the recipe-reader. (3) Guaranteed results: Whatever it is that an algorithm does, it always does it, if it is executed without misstep. An algorithm is a foolproof recipe. It is easy to see how these features made the computer possible. Every computer program is an algorithm, ultimately composed of simple steps that can be executed with stupendous reliability by one simple mechanism or another.

Argue the latest and greatest software/hardware/speed for some other strategy, not TF.

 

[Mtrader]

No one said you don't need one, but it's just a tool. A rather dumb, precise tool that follows human instruction.

I agree. But the point is that some had in the 70's a tool and most did not have one. Made a hell of a difference.
But a tool is only useful in capable hands....