Saturday 7 March 2009

S&P 500 May Pause Before Next Leg Down: Data

Wow. What an absolute disaster of a week. The Dow and S&P 500 are now well below their 2002 lows. The Nikkei has decisively broken below its 2003 low. The FTSE is now testing its lows of the last bear market. But Canada's TSX index is, oddly, well above its 2002 bottom. Does that mean there is still some hope for the market, with Canada's outperformance representing resilient commodity demand? Maybe the commodity bull really isn't over and is just seeing one hell of a huge correction? I don't know. What I can say is Friday's Commitments of Traders report doesn't give much hope for equity bulls. I've now updated my latest signals table. Some highlights from the latest weekly data:

- S&P 500: My trading setup for the S&P 500 has been bearish since November, but this coming week it goes briefly to cash. That's based on the commercial trader net futures and options positioning during the week of Feb. 10, when it abruptly rose to a bullish extreme. It quickly fell back into bearish territory the following week, so my setup goes back to bearish on the open of Monday, March 16, and will remain so for at least the ensuing three weeks. In the latest COT report, the commercial net position as a percentage of the total open interest dropped from 0.47 to 1.02 standard deviations below the moving average I use to evaluate their positioning. Meanwhile, the "dumb money" small traders - far from turning more bearish during the latest market cataclysms - have actually become even more bullish. Amazing! In Friday's data, their net percentage-of-open-interest position went from 1.48 to 1.60 standard deviations above the average. Not good at all if you happen to be a bull. They obviously are convinced the bottom is in. But their poor past record raises questions about whether that's the case.

- Gold: My trading setup for gold based on the COT data remains in cash again this week - the fifth in a row. The two signals that make up this setup don't agree with each other again. However, some possible light at the end of the tunnel: my signal based on the large speculator total open interest (long plus short positions) has flipped to bullish. That signal has a seven-week trade delay - which means the overall setup can't go to bullish for seven more weeks. Until now, the large spec total open interest has been persistently excessively bullish during the recent gold rally. That turned the setup to cash after a three-week bullish call. Now, the large specs have suddenly pulled back from being 1.07 standard deviations above the moving average to 0.72 standard deviations below. If my other signal - based on fading the large spec net percentage-of-open-interest position, but with no trade delay - is bullish in seven weeks' time, I'll be shopping for some bullion.

11 comments:

Oskar said...

Three questions:

1. From last post's comments: "... the gold stocks will go down while bullion goes up."

Are you talking about Gold Mine corporations here, or some different way of measuring the price of gold than just its daily price at the exchange?

2. I would be extremely interested in hearing your views on USD/Yen, and if you're thinking of developing a setup. I've read (and felt) more and more that Japan is close to a huge crash, and the Yen is currently historically very overpriced against the USD. More curious speculation here:

http://www.dailywealth.com/archive/2009/mar/2009_mar_04.asp

And the ProShares US Yen ETF, which would be the one to choose, no? Already some nice gains since jan.

3. What's your opinion on the recent BRK yearly report from Buffett on the bubble in long US Treasury bonds? It makes total sense after pondering it for a while, and Buffett is someone I highly respect for his reasonable and straightforward, indeed very American tone.

http://www.berkshirehathaway.com/letters/2008ltr.pdf

The ETF for this would probably be the ProShares US Treasury 20+, right?

http://www.proshares.com/funds/tbt.html

Your latest two trades (SP500 and Gold) were very nice. Missing the huge rise in gold from October to March, sadly, but of course anyone who claims to be able to predict peaks and bottoms is either delusional or a hypocrite with a motive.

Keep up the great work, and thanks for letting us share your thoughts and participate in a discussion with you.

Anonymous said...

While I habitually check your COTS data interpretations, something bugs me:

1. I would expect that once such an arbitrage is known, it would become ineffective.

2. Do you monitor for evidence of point #1?

Alex Roslin said...

Thanks for your message, Anonymous. This issue doesn't bother me. I just started using these two setups for SPX and gold in recent months and no more than a few hundred people visit my blog per day. The derivatives markets I'm studying are based on billions of dollars of positioning, so I think it's highly unlikely these readers could influence the outcomes of the signals.

Regards,
Alex

Alex Roslin said...

Hi Avocade,

Thanks for your message and kind words.

1) I don't see where I said anything about gold stocks going down while bullion goes up. My setup is for gold bullion. Gold stocks usually trade in the same direction as bullion, but not always.

2) I don't have views on the USDYEN. It would be an interesting setup certainly, but it's on my to-do list. Technically, it does look like it's breaking down, but there's some key support from a DeMark Setup Trend line at 100 on the daily chart for FXY.

3) I haven't read the BRK report and don't trade based on fundamentals, so I can't really comment on his views. On the other hand, TBT did break up above a TDST daily line around $44, but I've traded that one on a very short-term basis. I don't see a good way to have any long-term views on anything at this point in the markets. I have no idea how things will play out.

Regarding the gold setup, it's actually been on a buy signal for 12 weeks since October (with the remainder of the time in cash) - so it did participate in quite a bit of the rally. The current cash signal since Feb. 9 is now beating GLD as of today's trading. But as you rightly point out, it's impossible to catch every bottom and top, so I don't get too worried being in cash even if gold is up.

As someone once said, I'd rather be out of the market and wishing I was in than in the market wishing I was out.

Regards,
Alex

Oskar said...

"As someone once said, I'd rather be out of the market and wishing I was in than in the market wishing I was out."

Well said. That one goes straight into my quotes list.

Oskar said...

Anyone else thinking today's insane bull market was some major hedge funds buying back their shorts? I would have, if I had had them, like you yourself did, Alex.

I'm very impressed by how you timed this exit Monday. But I still can't fathom how anything the commercials did three weeks ago to make them jump suddenly to be temporarily net long (and then back a week after), would have an effect right now.

Alex Roslin said...

Hi again Avocade,

Thanks for your comments. I can't pretend to know why the trade delay works the way it does, but in my opinion it is a robust market effect. I can only speculate that it typically takes a specific period of time for trader positioning to make itself felt in the broader markets. To a large extent, if the effect passes all my tests of robustness, I don't really care why it works. That's a question for a fundamental analyst to worry about.

Regards,
Alex

Oskar said...

Very true. But count me in as one scared trader next monday when I'll follow you and shorting SP500 again. (ProShares US SP500 as usual?)

Could this rally today have been just a blip (hedge funds buying back short positions)?

Oskar said...

Interesting that the NASDAQ futures/options are only a bit shorted for the commercials (and long for others), but the "mini" version is _extremely_ shorted for non-reportable and quite long for commercials.

Why this huge discrepancy since the underlying asset is still the same? More of the very dumb money (small traders) on the much cheaper futures version, yes, but why are commercials so happy?

http://www.advancedfutures.com/eminnasd100.asp

Alex Roslin said...

Hi Avocade,

Thanks for your comment. The commercials and small traders on the mini contract might not be the same entities as those on the other contract. Check the reporting thresholds for both contracts.

Regards,
Alex

Tony said...

Hello Alex,

Thanks for posting your method on the web for people to review. You have an interesting method that I am going to watch for awhile. I like the heavy statistical approach. This week's signal to go cash on the S&P 500 has saved you a lot of losses compared to the previous short position. Good work and good luck!

Regards,
Tony