HOME

Thomas Nogales Financial's

Market Alert

Market Timing Works

Even A Simple Moving Average System can Beat Buy & Hold

As part of our ongoing market timing research, we've done an analysis on moving average systems to prove dead wrong the market "experts" who continually claim market timing doesn't work.  Thomas Nogales Market Alert doesn't use moving averages for timing the S&P500 but many investors and advisory services do. Our results confirm that many moving average systems can easily beat buy and hold.  There are, however, significant differences in performance over various time periods. The best moving average system we devised is based on a 40/10 week two average model but we'll discuss the results from other intervals too. The conclusion from our research:

Market Timing works. Simple moving average systems beat Buy & Hold and with less market risk.

The data below shows the results of systems we created that uses moving averages to time trades on the S&P500 index over a 43-year period from 1960 to 2003. The various rows show the result of changing the time intervals and using one and two moving averages. We used the weekly Friday closing prices rather than daily prices. The moving average results include the most recent Buy trade even if still open.

The heading captions are as follows:
B&H = Simple buy and hold aggregate returns
Model = Results of the moving average system
%Better = The percent by which the model beat buy and hold
Trades = The number of round trip trades
%Success = The percent of trades that made money
AvgGain = Total earnings divided by trades
AvgLoss = Total dollars lost divided by trades 
MedGain = The median average of all winning trades
MedLoss = The median average of all losing trades
WieghtedGain = The weighted median average of gains and losses per trade
Risk = The model's time in the market divided by the time in market of buy and hold

Test Results

The best performance on a $10,000 investment over the 43-year period came from a 40 week and a 10 week combination (200 day/50 day). But, there were big differences when we broke the periods into two sections: 1960-1980 and 1980-2003. The exact year ranges aren't critical but clearly show that moving average systems worked much better 20 years ago.

Here’s how the two moving average system works. 

Buy when the closing price is over the longer moving average
Sell when the shorter moving average goes below the longer moving average.

So, when the closing price goes over the 200 day moving average we buy.  Sell when the 50 day goes below the 200 day.  Don’t buy again until the price goes over the 200. Note: This may create a situation where the price is above the 40w MA but the 10w MA is below the 40w. In that case, buy back in when the 50 day goes over the 200 day.

In the chart below, on line four, the 40/10 is the best combination and beat buy and hold by 2.37 times. 68% of the trades were successful and it made about two trades per year. It was in the market 69% of the time. When not in stocks we gave it 5% per month/year for being in short term bonds. The 44/10 came in second.  The different totals in the B&H column occur because of different start and end dates.

Now, let’s break the 43 years down into two sections.  From 1960 to 1980 the 40/10 was the winner.

From 1980 to 2003, the 40/10 beat buy and hold by 20%. It was in the market only 73% of the time (risk) and was successful on 73% of the 33 trades.

Some people use a simple 200-day moving average (40 week) for market timing. It didn't do nearly as well over the 43 years as the 40/10. The 65 trades work out to 1.5 per year and only 45% were successful. It was in the market 66% of the time.

Now, let’s break that into two periods. The 200 day moving average performed well in the earlier years from 1960 to 1980.

It hasn't done nearly as well over the last 23 years but the risk level is still quite good.

Conclusion

The most important point of our analysis is: A simple, mechanical moving average system beats the Buy & Hold of an index fund over 20 year periods and with 30% less risk.

Moving Average systems will have periods of poor performance but hold up quite well over time.

So, why do so many commentators and so-called experts continually bash market timing when it obviously works?

First, most investors don't have the patience or the confidence to trade the stock market. They panic out of trades when the market moves against them rather than wait for the system's signal. They are probably better off in a mutual fund at least making some money. Second, uninformed investors are the source of profits for mutual funds. It's not the funds job to educate investors about market strategies. Besides, they get a percentage of the assets even if the investor loses money. Many mutual funds have over 100% portfolio turnover each year as they frantically buy and sell stocks. Ironically, they're lousy market timers trading on investor money.

Fact: Most mutual funds underperform index funds in the short term and virtually all mutual funds underperform over any 10 year period. They're held back by trading costs and expenses. The obvious conclusion: Place your assets in an index fund. Optionally, buy some individual stocks or manage a portion of your portfolio with a proven timing strategy.

If your're willing to manage your own money and have self discipline, shouldn' t you consider managing the market with some of your money to get much better returns?

A Much Better System

Moving averages are simplistic. Couldn't an intelligent model do much better?

A moving average system by definition always lags the market on the way up and on the way down. So, it always buys a bit late and sells late. The Nogales Market Alert is real-time and quite a bit smarter. It has about the same risk level but four times the return of the best moving average system. In 2003, Nogales Market Alert bought into the S&P500 in February at 829 whereas the moving average bought in May at 944. Nogales Market Alert will likely sell sooner too. Since markets generally fall much faster than they rise, getting out early is very important for producing superior returns.

Fact: A smart market timing strategy can greatly outperform a moving average system. Very few timing systems combine high returns with few losing trades.

Let's compare the best moving average combination (40/10) to the Nogales Market Alert. On a 25 year back test, 1979 to 2003, the moving average model turned $10,000 into $125,000 on 35 trades. Market Alert turned $10,000 into $450,000 on 12 trades.

The difference in capital growth between Nogales Market Alert and Buy & Hold is the result of money growing at a higher annual return. Market Alert earned 16% per year over the 25 years and Buy & Hold earned 10.2%. Many major corporations and independent investors seek a 15% return on equity per year and you should too. This is not unrealistic.

For more informantion about Nogales Market Alert, read our FAQ. It explains what the model does and how to use it for improved investment returns.

What does the 40/10 moving average show for the S&P500 as of January 2004?  Here's the chart.  It's still in the market and so is Market Alert. Which do you think will get a higher price when it's time to sell?

Check out our web site at WWW.ThomasNogales.COM

Copyright 2003, Thomas Nogales Financial, LLC