Friday 18 July 2008

Sucker Rally? Not According to COT Data... At Least Not Yet

Wow. That was surely one of the most nutso weeks in recent market history. My 200-percent leveraged UYG ProShares Financials Fund collapsed 22.9 percent by the intraday low Tuesday, after which it rose - wait a sec, is that right?! - yup it sure is - 46.8 percent. And what do you suppose the Commitments of Traders data was doing in the middle of all this circle-puking? I was really wondering about that considering the data was compiled Tuesday, which was the "panic day" of the week. Turns out that for the first time in a long while, there were no new signals from this afternoon's data for my trading setups based on the COT positioning. None. I love it. See my latest signals page table for some pending signals updates. Here are a few more highlights looking more closely into the data:

- Was this week's bounce for financials a sucker rally, like some people fear? Not yet, at least not according to my trading setup for the BKX U.S. Bank Index (based on the three-month Eurodollar COT data). That setup is still bullish, and as of Friday's data even more so. The "smart money" Eurodollar large speculators have increased their net long percentage-of-open-interest position and now stand 1.4 standard deviations above the moving average I use in this setup. However, also of note is that my setups for the S&P 500, NASDAQ 100 and Russell 2000 all remain in cash.

- My trading setup for Japan's Nikkei Stock Average goes to bearish on the open of trading, Monday, July 21. It flipped to bearish with the May 20 COT report, but the setup uses an eight-week trade delay, which means execution was held off until this coming Monday.

- Is crude oil done? That is definitely the consensus out there. The breakdown this week does seem pretty bad. By most technical standards, I think people would agree the uptrend is in trouble. Mind you, I don't think the technical evidence is completely clear just yet. The trendline since February for the USO U.S. Oil Fund has been violated, but the breakdown hasn't progressed to the place where I'd call it decisive just yet. My view is it would have to fall below that line by a notch beyond the Average True Range on the daily chart for me to start to feel crude is a goner (plus some other stuff, like fail on a rally attempt and make a lower low). USO is just sitting at that line right now. But for me, of course, the true arbiter is the COT data. My crude setup is still very much in bullish mode. It's based on trading opposite to the "dumb money" large speculators in crude futures and options. Their net percentage-of-open-interest position is presently 1.4 standard deviations below the moving average I use for this setup. Where are all these speculators who are supposedly driving up energy prices? They've actually been reducing their net position for three weeks now.

- My small long position in natural gas got pole-axed in this week's dramatic selloff, but my setup remains nowhere near going to bearish for the moment. Natural gas is one of the most volatile sectors you can own, so I'm in it knowing there will be ridiculous ups and downs. This week was a good example. I've also just found a new setup to use for this market and have posted the results on the latest signals page table. The setup went from starting equity of $100 to over $95,000 by the end of 2007 (when my test period ends). I think it scores quite nicely across all my robustness indicators, so I'm excited to start trading it. It combines the signals from two of my best setups, including a slight variation of the one I was trading before. Its largest past drawdown is a lot lower than that of my previous setup, and that means I can play it with a larger position. (See my how it works and glossary pages for more details on my risk-control rules - one of the most important parts of investing and trading, IMHO.)

- Also of note: commercial traders in U.S. dollar index futures have dramatically reduced their formerly massive net short position this past week. However, their move highlights the need to put this kind of development in some historic perspective. Their position still has an overall bearish tilt at 0.4 standard deviations below the moving average I use for this setup.

Hope you have a relaxing weekend, and see you early next week with my portfolio page update and hopefully another new setup or two.

TAGS: Bank Index, financials, BKX, UYG, ProShares, crude oil, U.S. dollar, Nikkei, natural gas, NASDAQ 100, NDX, S&P 500, Russell 2000, COT, Commitments of Traders, market timing, trading system development, CFTC, Commodity Futures Trading Commission, COTs Timer

8 comments:

Anonymous said...

Hi Alex,

I haven't been fortunate these two weeks, just getting chewed up in Nat.Gas :( The price of Oil & Gas seems to be so manipulated or controlled... worldwide consumption seems to have suddenly fall off a cliff in a week. Is it talked about correction or bubble bursting? I guess only time will tell.

As for financials bouncing, will Nasdaq follow; or Google and Microsoft earnings foreshadowing of what's to come... hence the no change in the cots report today.

Rgds.
Pete

Anonymous said...

hi Alex,

can you post a history of all signals that your system gave in the last one year?

johnny

Alex Roslin said...

Hi Johnny,

I will if you come mow my grass, weed the garden, plant the trees sitting in a bucket for the past three weeks, turn the compost, put up the swing for the kids and trap the raccoon that's been eating same compost for nearly a year. Let me know.

Thanks,
Alex

Anonymous said...

Alex,

What are you going to do tomorrow with UYG due to disappointing AMEX numbers?

Anonymous said...

Hi Alex,

have you ever tried letting the market confirm a COT-signal by breaking out to the up- or downside in the delay period instead of always delaying the buy?

If yes, what are your experiences?

Thanks,

Andreas

Anonymous said...

thought you kept track of this stuff? why not just keep the old version of the signals table every time you update it? or at least you must keep track of how much money you won or lost, if you really use this system to invest your savings.

johnny

Alex Roslin said...

Hi Anonymous,

Thanks for the comments from several readers this morning. (1) Disappointing or favourable news events have zero relevance to how this system generates signals.

(2) Andreas: I've done some testing of price action as a filter for the COT signals, but so far I've found that the COT data alone gives the best signals.

(3) Johnny: Yes, I do keep track of my own trades. I've been reporting my trade gains and losses for a few months now on the portfolio page. As well, you can see a basic version of how my spreadsheets work on the DIY S&P 500 sample spreadsheet page. If you were to download this yourself, you would be able to see all past signals, and you would be able to update it yourself. The specs for several other setups are available on my latest signals page table, which you can use to generate spreadsheets for those markets yourself if that's of interest.

Regards,
Alex

Anonymous said...

Hi Alex,

interesting to hear.
FWIW: I found this valid for seasonal trades too.

Andreas