The Realities Of Market Timing

Market timing systems are based on patterns ofwe were trying to put together an enhanced version
activity in the past. Every system that you are likelyof the Standard & Poor's 500 Index, based on
to hear about works well when it is applied tothe past 30 years. Based on hindsight, we could
historical data. If it didn't work historically, you wouldprobably significantly improve the performance of the
never hear about it. But patterns change, and theindex with only a few simple changes.
future is always the great unknown.For instance, we could conveniently *remove* the
A system developed for the market patterns of theworst-performing industry of stocks from the index
1970s, which included a major bear market thatalong with any companies that went bankrupt in the
lasted two years, would have saved investors frompast 30 years. That would remove a good chunk of
a big decline. But that wasn't what you needed in thethe *garbage* that dragged down performance in
1980s, which were characterized by a long bullthe past. And to add a dose of positive return, we
market. And a system developed to be ideal in thecould triple the weightings in the new index of a few
1980s would not have done well if it was back-testedselected stocks; say Microsoft, Intel and Dell. We'd
in the 1970s. So far in the 1990s, any defensiveget a new *index* that in the past would have
strategy at all has been more likely to hurt investorsproduced significantly better returns than the real
than help them.S&P 500. We might believe we have discovered
If your emotional security depends on understandingsomething valuable. But it doesn't take a rocket
what's happening with your investments at any givenscientist to figure out that this strategy has little
time, market timing will be tough. The performancechance of producing superior performance over the
and direction of market timing will often defy yournext 30 years.
best efforts to understand them. And they'll defyThis simple example makes it easy to see how you
common sense. Without timing, the movements ofcan tinker with past data to produce a *system*
the market may seem possible to understand. Everythat looks good on paper. This practice, called
day, innumerable explanations of every blip are*data-mining,* involves using the benefit of hindsight
published and broadcast on television, radio, into study historical data and extract bits and pieces of
magazines and newspapers and on the Internet.information that conveniently fit into some philosophy
Economic and market trends often persist, and thusor some notion of reality. Academic researchers
they seem at least slightly rational. But all thatwould be quick to tell you that any conclusions you
changes when you begin timing your investments.draw from data-mining are invalid and unreliable guides
Unless you developed your timing models yourselfto the future. But every market timing system is
and you understand them intimately, or unless youbased on some form of data-mining, or to use
are the one crunching the numbers every day, youanother term, some level of *optimization.* The only
won't know how those systems actually work. You'llway you can devise a timing model is to figure out
be asking yourself to buy and sell on faith. And thewhat would have worked in some past period, then
cause of your short-term results may remain aapply your findings to other periods.
mystery, because timing performance depends onNecessarily, every market timing model is based on
how your models interact with the patterns of theoptimization. The problem is that some systems, like
market. Your results from year to year, quarter tothe enhanced S&P 500 example, are
quarter and month to month may seem random.over-optimized to the point that they toss out the
Most of us are in the habit of thinking that whatever*garbage of the past* in a way that is unlikely to be
has just happened will continue happening. But withreliable in the future. For instance, we recently looked
market timing, that just isn't so. Performance in theat a system that had a few *rules* for when to
immediate future will not be influenced a bit by thatissue a buy signal, and then added a filter saying such
of the immediate past. That means you will nevera buy could be issued only during four specific
know what to expect next. To put yourself throughmonths each year. That system looks wonderful on
a *timing simulator* on this point, imagine you knowpaper because it throws out the unproductive buys in
all the monthly returns of a particular strategy over athe past from the other eight calendar months.
20-year period in which the strategy was successful.There's no ironclad rule for determining which
Many of those monthly returns, of course, will besystems are robust, or appropriately optimized, and
positive, and a significant number will represent losses.which are over-optimized. But in general terms, look
Now imagine that you write each return on a card,for simpler systems instead of more complex ones.
put all the cards in a hat and start drawing the cardsA simpler system is less likely than a very complex
at random. And imagine that you start with a pile ofone to produce extraordinary hypothetical returns.
poker chips. Whenever you draw a positive return,But the simpler system is more likely to behave as
you receive more chips. But when your return isyou would expect.
negative, you have to give up some of your chips toTo be a successful investor, you need a long-term
*the bank* in this game. If the first half-dozen cardsperspective and the ability to ignore short-term
you draw are all positive, you'll feel pretty confident.movements as essentially *noise.* This may be
And you'll expect the good times to continue. But ifrelatively easy for buy-and-hold investors. But market
you suddenly draw a card representing a loss, yourtiming will draw you into the process and require you
euphoria could vanish quickly.to focus on the short term. You'll not only have to
And if the very first card you draw is a significanttrack short-term movements, you'll have to act on
loss and you have to give up some of your chips,them. And then you'll have to immediately ignore
you'll probably start wondering how much you reallythem. Sometimes that's not easy, believe me. In real
want to play this game. And even though your brainlife, smart people often take a final *gut check* of
knows that the drawing is all random, if you drawtheir feelings before they make any major move. But
two negative cards in a row and see your pile ofwhen you're following a mechanical strategy, you
chips disappearing, you may start to feel as if you'rehave to eliminate this common-sense step and simply
on *a negative roll* and you may start to believetake action. This can be tough to do.
that the next quarter will be like the last one. Yet theYou will have long periods when you will
next card you draw won't be predictable at all. It'sunderperform the market or outperform it. You'll
easy to see all this when you're just playing a gameneed to widen your concept of normal, expected
with poker chips. But it's harder in real life.activity to include being in the market when it's going
For example, in the fourth quarter of 2002, ourdown and out of the market when it's going up.
Nasdaq portfolio strategy, with an objective toSometimes you'll earn less than money-market-fund
outperform the Nasdaq 100 Index, produced a returnrates. And if you use timing to take short positions,
of 5.9 percent, very satisfactory for a portfoliosometimes you will lose money when other people
invested in technology funds only. But that wasare making it. Can you accept that as part of the
followed by a loss of 7.8 percent in the first quarternormal course of events in your investing life? If not,
of 2003. Most investors in this strategy, at leastdon't invest in such a strategy.
those we know of, stuck with it. But theyEven a great timing system may give you bad
experienced significant anxiety at the loss and theresults. This should be obvious, but market timing
shock of a sharp reversal in what they had thoughtadds a layer of complication to investing, another
was a positive trend. The same phenomenonopportunity to be right or wrong. Your timing model
happened, with more dramatic numbers, in our moremay make all the proper calls about the market, but
aggressive strategies.if you apply that timing to a fund that does
Some investors entered those portfolios in thesomething other than the market, your results will be
winter of 2002, and then were shocked tobetter or worse than what you might expect. This is
experience big first-quarter losses so quickly aftera reason to use funds that correlate well you're your
they had invested. Some, believing the losses weresystem.
more likely to continue than to reverse, bailed out.The bottom line for me is that timing is very
Had they been willing to endure a little longer, theychallenging. I believe that for most investors, the best
would have experienced double-digit gains during theroute to success is to have somebody else make
remainder of 2003 that would have restored andthe actual timing moves for you. You can have it
exceeded all of their losses. But of course there wasdone by a professional. Or you can have a colleague,
no way to know that in advance.friend or family member actually make the trades for
Most timers won't tell you this, but all market timingyou. That way your emotions won't stop you from
systems are *optimized* to fit the past. That meansfollowing the discipline. You'll be able to go on
they are based on data that is carefully selected tovacation knowing your system will be followed. Most
*work* at getting in and out of the market at theimportant, you'll be one step removed from the
right times. Think of it through this analogy. Imagineemotional hurdles of getting in and out of the market.