Global Macro Trading Journal

[Ghost of Cutten]

Quote from Daal:

This is not to say that playing a greater fool game right now won't work. In fact, using the Hussman model to the long side plus a good stop strategy probably makes money. But if I were to do that, the first thing I would do is to admit to myself that I'm playing greater fool not to pretend that somehow the fundamentals are ok(Then dump everything as soon as the market has a few bad days in a row and start to argue completely opposite things as reasoning, ala Gartman)

Or, you and Hussman might be totally wrong, and it might not be a greater fool game at all - it could easily be a totally justified and long-overdue move. There aren't any actual *facts* that say it's a greater fool game - just one person's model.

Even if it is a 'greater fool game', as a trader you have to ask which do you value more - sticking to a view, or making a profit?

 

[darkhorse]

Quote from Ghost of Cutten:

Let's say AMZN is a great short. What's the best way to play it?

1. Short outright immediately, as a hedge against other longs in your portfolio. If so, what % of total capital should be short AMZN?
2. Wait for a bear trend in the stock, then place a trading short (with a stop etc).
3. Buy some long-dated puts.
4. Write bear call spreads each month.

Depends on a few things, but I like a combination of 3 and 4 -- finance the purchase of long-dated puts with bear credit spreads, thus taking a limited amount of upside risk, but having a zero carry position that can't be stopped out with open-ended potential if AMZN freefalls.

 

[Daal]

Quote from Ghost of Cutten:

This is pure data-mining. Data-mining doesn't work reliably in complex systems, and financial markets are a complex system. His sample size is incredibly small and based only on one stock market - the US market. He has not tested this data on the dozens (probably hundreds now) of stock markets in the rest of the world. He has done no out-of-sample testing either. It's worthless unverified bunk.

And what about you?How you done this supposed extensive scientific work in hundreds of markets to base your long on?

 

[Daal]

Quote from Ghost of Cutten:

Or, you and Hussman might be totally wrong, and it might not be a greater fool game at all - it could easily be a totally justified and long-overdue move. There aren't any actual *facts* that say it's a greater fool game - just one person's model.

Thats why trading isn't science, its making educated guesses based on limited information

 

[darkhorse]

"After a certain high level of technical skill is achieved, science and art tend to coalesce in esthetics, plasticity, and form. The greatest scientists are always artists as well."

 

[Specterx]

Quote from Ghost of Cutten:

This is pure data-mining. Data-mining doesn't work reliably in complex systems, and financial markets are a complex system. His sample size is incredibly small and based only on one stock market - the US market. He has not tested this data on the dozens (probably hundreds now) of stock markets in the rest of the world. He has done no out-of-sample testing either. It's worthless unverified bunk.

So, was he just lucky in 2000, 2007, 2010 and 2011?

The chart here is rather dramatic: http://www.hussman.net/wmc/wmc120305.htm

Of all the angles one could take to criticize Hussman's results or methodology, going after his "risk is extremely high" calls is by far the weakest. They've been consistently spot on, certainly within the tolerances prescribed by his investment horizon (years to decades). Other aspects of his investment strategy have proven manifestly detrimental to his returns, but not this one.

His "syndrome" described above is merely saying that SPX has gone up really far, really fast, in the midst of heavily bullish sentiment and overvalued conditions, and it's therefore not the best time to buy. It can't actually hurt that historical backtesting supports this conclusion. Nobody can be certain of the future and nobody can make you change your opinion, but to me it seems irresponsible to so easily dismiss Hussman's arguments as fluff without a more rigorous investigation.

As always, this is all hugely irrelevant to whether going long ES next week on a pullback is a good idea.

 

[Specterx]

Quote from Ghost of Cutten:

Let's say AMZN is a great short. What's the best way to play it?

1. Short outright immediately, as a hedge against other longs in your portfolio. If so, what % of total capital should be short AMZN?
2. Wait for a bear trend in the stock, then place a trading short (with a stop etc).
3. Buy some long-dated puts.
4. Write bear call spreads each month.

AMZN could be a decent candidate for an outright short.

Short on a decisive break of the green trendline with a trailing stop above the red. $130 is a reasonable technical target, figure you get the short no lower than $170 and the stop will be a maximum of $188; 2.3:1 R:R.

 

[Butterball]

Quote from Ghost of Cutten:

This is pure data-mining. Data-mining doesn't work reliably in complex systems, and financial markets are a complex system.

What about your very own 'bond yields are low so investors are forced to bid up equities' or 'companies are beating analyst expectations so that's bullish' theories? That's no data mining?

Have you tested them over hundreds of international equity markets? What about walk-forward testing?

If you're trying to piss on Hussy you'd look a lot better if you brought at least a fraction of his logic rigor to the game.

 

[ralph00]

Treasurys finally getting the news that the world isn't ending. Ugly.

TLT might be down 10% YTD before week's end.

 

[Debaser82]

Quote from ralph00:

Treasurys finally getting the news that the world isn't ending. Ugly.

TLT might be down 10% YTD before week's end.

Not 3 months ago it was all private banking clients wanted to buy...