Short-term Outlook (1-5 Days):  25% Bullish  Since June 23, 1085 SPX

Go to:  Top  |  Short-term Outlook  |  Int-term Outlook  |  Equity Updates  |  Indicator Summary  |  Commodity Updates

Go to:  Top  |  Short-term Outlook  |  Int-term Outlook  |  Equity Updates  |  Indicator Summary  |  Commodity Updates

Equity Indicators - Updates and Extremes

S&P 500 Moving Average Crossover

Long-time analyst Peter Eliades mentioned something in his daily update last night that bears testing.

Quoting another respected long-time analyst, Tom DeMark, "[when] the 10 week moving average of an index moves above the 30 week moving average and both averages are slanted upward, it is a good buy signal."

Peter wondered whether the opposite was true - a valid point, since the S&P 500 is about to print one of those bearish crossovers.

So let's check.

We'll go back to 1928 and go long the S&P whenever its rising 10-week moving average crosses above its rising 30-week average, and then sell when the 10-week falls back below the 30-week.  Then we'll reverse those conditions for short trades.

Here are the results of going long on a bullish crossover:

Net Profit/Loss:  +1,253 points (versus +1,074 for buy-and-hold)

Winning %:  58%

Average Winner:  11.2%

Average Loser:  -4.6%

Maximum Drawdown:  -236 points (versus -878 for buy-and-hold)

Most trend-following strategies, especially using moving averages, have a very low winning percentage.  They can still be profitable since they catch the really big trends, but often the numerous whipsaws just aren't worth the trouble.

This strategy has been better than most others.  The winning percentage is actually above 50%, which is unusual, and the average winner is more than twice as large as the average loser.

Overall, it beat buy-and-hold handily, which is hard to do, and all while cutting the maximum drawdown one would have suffered by more than 60%, which is excellent.

For reference, here are the past 30 years' worth of buy signals.

None of that does us a whole lot of good right now, though, since we're about to trigger a sell signal instead of a buy signal.

So now let's look at the reverse, and sell short the S&P on a bearish crossover.

Net Profit/Loss:  +522 points

Winning %:  43%

Average Winner:  16.2%

Average Loser:  -9.1%

Maximum Drawdown:  -323 points

It's extremely difficult to find simple trend-following strategies that are profitable on the short side when looking at equities.  Due to the long-term upward drift, the whipsaws almost always swamp the few times the strategies catch major bear markets.

This one, however, was responsive enough to not get whipsawed too badly.  While the winning percentage was only 42%, that's actually not too bad for a trend-following system, and it's excellent for a trend-following system that sells short.

The average winner was an impressive +16.2%, which reflects all of the dramatic bear markets that it caught.  The average loser, though, was an uncomfortable -9.1%, which happens when the market rebounds hard from short-term selloffs.

Here are the past 30 years' worth of short sales (remember, these are short trades, so a positive return means that the market declined):

Four of the past five signals were nicely profitable, as the market has been relatively cooperative with trend-following strategies over the past decade, moving in long, multi-year waves up and down.  During the '80s and '90s it wasn't quite so smooth, thus the numerous whipsaws during that time.

Overall, the fact that we seem to be on the verge of getting a sell signal is not a good sign, but neither is it a reason to panic.  Almost 60% of the time, the market is higher by the time the 10-week average crosses back above the 30-week.

Over the past decade, though, the market has moved in long-term cycles that have benefited strategies like this, and if that continues then obviously it's a worse sign than it was prior to the past 10 years.

Go to:  Top  |  Short-term Outlook  |  Int-term Outlook  |  Equity Updates  |  Indicator Summary  |  Commodity Updates