Investment strategy is a little like religion in the financial advisor community. There are few situations that would get emotions boiling, fists flying, and require police action faster than putting a buy-and-hold advocate and a market timing zealot in a room and asking them to resolve their differences. The truth is that most strategies work some of the time, a few work most of the time, and only Bernie Madoff figured out how to make one work all the time, right up until he got caught. Investment strategies have two major parts: 1) what investments to buy, and 2) when to buy and sell. Because I'm an investment advisor and human, I have some built-in biases, but following is an attempt to objectively look at several common strategies with a minimum of sarcasm.

Traditional asset classes include stocks, bonds and cash. These classes are then divided into subcategories based on geographic location (U.S., developed foreign countries, emerging markets), company size (small-cap, mid-cap, large-cap), and bond style (treasuries, mortgage-backed, high-yield, etc). Real estate, commodities, and hedge funds are sometimes added as additional asset classes. The idea behind Strategic Asset Class Allocation is to come up with a portfolio of non-correlated assets that meets an acceptable risk profile, and then stick with that allocation as the market goes up and down. The portfolio is typically rebalanced periodically to maintain the percentages of each asset class, but mostly the portfolio is left alone.

Dividing the stock world by geographic location (U.S. & foreign) or by company size no longer results in a diversified portfolio. This has been a long-term trend developing and getting worse over the last 20 or so years. As an intuitive example, when oil drops from $150/barrel to $35/barrel, all energy companies get hurt, whether they are large or small, based in the U.S. or based in Brazil. However, it is true that an asset class allocation model is easy to implement with mutual funds, and the addition of non-correlated alternative investments can improve overall diversification.

As stated above, a major problem with Asset Class Allocation is that the major equity classes do not behave differently enough to do an effective job of diversification. Balanced Sector Allocation gets around this by diversifying across low-correlated sectors (Technology, Energy, Financials, Healthcare, etc). This is not a new concept. Just about any portfolio that uses individual stocks diversifies this way, and the strategy can be implemented using either individual stocks or sector-based Exchange Traded Funds (ETFs).

In addition to showing a significant performance improvement over the last 10-20 years, Sector Allocation passes the "this just makes sense" test. Intuitively, a Healthcare stock and an Energy stock will do a better job at diversification than a large-cap Energy stock and small-cap Energy stock. The manager of an actively-managed mutual fund is typically doing sector allocation within a particular Asset Class (e.g. Large Cap Value), but if you own several mutual funds, there is obviously no coordination between the managers.

These strategies are similar, with the difference being that one uses traditional asset classes and the other uses stock sectors. In both cases, the objective is to predict which stock class or area of the market will perform better in the near future, and overweight the portfolio to take advantage of that market segment or segments. The basis for determining which asset class or sector to invest in or stay out of can be based on a computer model, economic indicators, or (more commonly) an advisor's opinion or gut feel.

In order to significantly beat the market, you have to take some additional risk, and this strategy does that. When called correctly, this strategy can make huge gains. It can also lose a significant amount of money while everyone else is making money. By picking the right sectors or asset classes at the right time, it is possible to make money in practically any environment. However, similar to flipping a coin and trying to get "heads", I'm not sure past success is a great predictor of future success.

A pure buy-and-hold strategy involves buying a high-quality investment such as stocks or a mutual fund, and then holding the investment through highs and lows until either your investment objectives change or you find out the investment is not as high-quality as you thought it was. The rationale is that the overall market goes up over time, and you don't want to miss a big up day in the market by holding cash.

Buy-and-hold makes money when investments go up, and loses money when they go down. Therefore, it works well during bull markets and works poorly during bear markets. For this strategy to continue to work for the next 30 years like it did the last 30 years, you have to assume that investments will continue to go up like they have during a period of economic growth that was fueled by the Baby Boom generation, an Energy bubble, a Technology bubble, and a Real Estate bubble.

Market Timing is one of the most loosely-defined terms in the financial industry. There are many advisors who deride market timing, and yet routinely practice market timing themselves. Broadly-defined, market timing is a strategy that makes changes to a portfolio based on predicted market performance. These changes may involve selling all investments and moving to cash, or simply adjusting the percentage of stocks and bonds because of economic conditions or anticipated market behavior. Prediction-based market timing bases decisions on an advisor's assessment of future conditions. If high-inflation is anticipated, investments that hedge against inflation would be added. If economic contraction is anticipated, an advisor might move to a heavier cash position.

This method is highly dependent on the person or statistical model making the prediction. If the predictions are accurate, this strategy has a good chance of significantly outperforming other methods. If the predictions are wrong, the opposite is true. Because of the large number of advisors who make predictions, a certain number will get it right several times in a row, but statistically this will not indicate any greater likelihood that they will continue to be right in the future. As mentioned above, unanticipated news events or government action will instantly derail most statistical models.

Momentum-based market timing uses technical indicators (stock charts and current market behavior) to determine whether the market is in a downtrend or an uptrend. Downtrends occur when more people want to sell than want to buy, and uptrends occur when more people want to buy than want to sell. Price movement and trading volume can determine whether there is more buying pressure or more selling pressure at any given time, and the theory behind momentum is that once a trend is in place, it tends to stay in place. For how long? Until it stops.

There are some key components required for this system to work.

1) The method for determining trends must not be too early or too late. Stocks seldom move in a straight line. They typically make a strong move, and then rest or pullback. Assuming too early that a trend is being established or ending will result in jumping in or out during pullbacks or corrections. Waiting too long or for too many confirmation signals will result in missing a good portion of the trend.

2) Investments must be liquid. You must be able to act when your system tells you to buy or sell.

3) Whether you use Moving Averages, charting, or any other system to determine a trend, the trend will not always hold. Each system will break down under certain conditions, so the objective is to use a system that works under the widest set of conditions and/or breaks down under the narrowest set.

This is not a strategy that is typically planned for or entered into intentionally, and is the form of market timing most often practiced by those who swear they hate market timing. Many practitioners of this strategy consider themselves to be buy-and-hold investors, but they end up moving to cash when the pain gets too great or the market is too scary. Typically, this happens after a significant loss is already on the books, which actually makes this a form of momentum. The rationale is that if my investments have already lost money, they may continue to lose money. The problem is that if emotion or fear drives the sell decision, then the decision to get back in is typically based on "feeling better", which almost always happens at a higher price than the sell price.

This strategy seldom works, and is the reason that the vast majority of investors buy when the market is high and sell when the market is low. It doesn't matter which strategy you use; just about anything is better than basing investment decisions on emotion.

I use a Balanced Sector Allocation strategy using low-correlated ETFs, and momentum-based market timing. The objective is to participate as much as possible in uptrends, and avoid as much of the downtrends as possible. This requires a set of rules that makes the decision points unemotional. A Balanced Sector Allocation guarantees participation in the hottest trending sector at any given time, but with a mechanism to get out of a sector when it starts heading back down.

Because it takes a little while for a downtrend to show itself, sell decisions will never happen right at the top of a trend. The same holds true for uptrends and buy decisions. If the market gets indecisive and swings far enough that it keeps looking like uptrends and downtrends are forming but no follow-through happens, a condition could occur where losses are exaggerated. This would be a very specific and narrow set of conditions, and I have other checks that attempt to minimize this condition, but it still exists.