Rationale

There should be a rationale behind any system you use, especially why it provides superior results.

For instance, in The Right Stocks at the Right Time, Larry Williams shows that the stock market has never lost money in years ending with 5 (1895, 1905,..., 1995 and probably 2005). However, he does not provide any explanation and I don't see any, therefore I'm not comfortable using such signal in a Market Timing system.

Simple Rules

Rules should not be too complex. If they are, they're likely the results of Data mining (adapting indicators so that they fit past stock market behaviors).

Favor systems that have simple rules.

Quite often, complex rules come from simpler ones: some people torture the data enough to come up with an optimized set of rules that supposedly provide superior performances than the original system.

As an illustration, in All About Market Timing by Leslie N. Masonson, starting from the simple Presidential Cycle timing strategy, the author then shows an "optimized" system that invest in optimum months of the Presidential Cycle. The strategy handsomely beat the original Presidential Cycle system, however the rules are just crazy to implement.

On the other hand, don't systematically reject any enhancement.

Backtesting

• Any system's performances should be checked against both Bull and Bear markets.

• Outperformance should not come from few exceptional years but be somewhat consistent.

• Check how the system fared over various 5 to 10 years periods rather than how it fared over 30-50 years. The latter may show impressive numbers but very few investors have 30-50 years timeframe.

Self-Adaptive Indicators

It is preferable to use indicators or systems that are self-adaptive to any (at least most) market conditions.

A typical example is Valuation: avoid using absolute Valuation such as "Buy when Market PE < 20" but use relative Valuation such as "Buy when Market PE < 1/(10 years Government Bond)". Valuation can be high or low for extended period so relative criteria is best.

Use Multiple Market Timing Systems

Use several systems from different groups (Trend following, trend anticipation...). Here are simple Market Timing systems that you can easily implement:

• Trend anticipation: Monetary (Fed Fund Rate), Valuation (Beating the Dow with Bonds),

• Trend following: Technical Analysis (Moving Averages),

• Neither anticipation nor following: Calendar (Best 6 months, Presidential Cycle or both combined),

Think Tactical Asset Allocation

Many people think of Market Timing as a binary system, that is 0% in the Market or 100% in it.

It does not have to be that way. You can design a system that progressively gets you in and out of the market as conditions become favorable or unfavorable. It is like Tactical Asset Allocation but as opposed to Tactical Asset allocators who try to predict the future stock market move and adapt their positions accordingly, you adapt your position automatically, unemotionally.

For instance, you can built a Market Timing strategy with 3 indicators and decide to progressively increase your stock allocation for each indicator that turns positive. You can even set minimum and maximum permitted percentage allocation to stocks.

Conclusion

• Use indicators that have a rationale, are simple, provide consistent results, are preferably self-adaptative.

• You can build a successful Market Timing system by combining several simple indicators from different groups (Trend Anticipation, Trend following, Calendar).

• For extra safety, design your system such that it progressively gets you in and out of the market as conditions become favorable or unfavorable.