Friday, 5 November 2010

Why I'm (Kind of) Bearish Despite the Rally

Nice breakouts after the U.S. congressional election Tuesday. But with some exceptions, the Commitments of Traders data remains generally bearish, according to my take. I've just updated my latest signals table based on Friday's weekly COT data from the Commodity Futures Trading Commission.

So what should I do? Ignore my signals? Well, no. The fact is signals are sometimes wrong. Actually, quite often. If you look at my backtesting results table, you'll see that my trading setups were profitable only 64 percent of the time on average in historical testing. That's actually not a bad success rate for a mechanical trading system. What's important for me is making money over the long run and cutting losses when the signal goes against me. Also, some signals are wrong for a while, then turn out to show a gain. You never know what's going to happen in the markets. And if I don't use my system with discipline and the right risk control - even when signals don't work - what's the point of having a system in the first place?

Check my FAQs page for more on the question of losing signals, risk management and how I trade my system. All that said, I am actually net long when I include my discretionary trades, so this ramp-up doesn't bother me too much anyway. I like having the COT short positions as a hedge and to help diversify my strategies.

Some highlights from my signals page:

- S&P 500: Still bearish, and no signs of changing its mind. The "smart money" commercial hedgers remain totally unimpressed with the market, while the wrong-way small traders just dig it.

- Nikkei: My trading setup for the Nikkei was bearish for eight weeks, and now it goes bullish for a single week on Monday's open of trading, then it heads to cash on the open of Nov. 15.

- Crude oil: My setup for oil is bearish for a fourth straight week and shows no signs of changing direction either.

- 30-year Treasury bond: My signal was in cash last week, but it goes to bearish for Monday's open of trading (meaning bullish for the yield). The setup will remain bearish for at least the next three weeks.

Hope you did well this past week and that you have a good weekend. Check back here early next week for a portfolio update.

TAGS: SPX, S&P 500, gold, BKX, Bank Index, natural gas, Nikkei, crude oil,Treasury, bond, COT, Commitments of Traders,derivatives, Black Swans,market timing, trading system development, CFTC,Commodity Futures Trading Commission,COTs Timer, out-of-sample testing, walk-around testing


willie25 said...

Why are you not using the system for trading the NASDAQ that you described in your article in TASC? Has the system stopped working? Thanks.

Mitu said...
This comment has been removed by a blog administrator.
Alex Roslin said...

Hi Willie25,

That was a very early generation of my system, which has evolved considerably based on more research into the best ways of backtesting trading systems. The system is the same in terms of parameters, but my backtesting has evolved a lot. See my FAQs page to read how I backtest now.

Since developing my current backtesting methods and trading setups, I haven't had a chance to develop setups in all the markets I'd like to be trading with the COT data, including the NASDAQ. When I update the setups based on new data since the end of 2007 (which was the end of my backtest period), I'll look at including new markets.