 |
In
This Edition... |
|
While the 'Sell
in May and go away' advice has been brilliant as far as the month so
far is concerned, the long-term numbers behind the axiom don't necessarily
add up to tepidness for the entire spring/summer period. More importantly
though, now that the bears have gotten the ball rolling, when - and
where - might this pullback stop? We've discussed some initial possibilities
as well as the misleading 'sell in May' adage below.
|
 |
Do
You Really Want to Sell in May? |
|
|
We
delivered a similar message about this time last year, but considering
it's relevant again (and always will be), we're going to update the actual
numbers behind the whole 'sell in May and go away' thing again. Let's just
say the advice is misleading at best.
 The
principle of selling in May and just staying on the market's sidelines
until the end of September is one built on the long-term averages of monthly
returns. Fair enough. But, the actual May-September span hasn't been nearly
as bad (usually) over the last four decades as you might want to believe.
And, it doesn't
matter how you look at the data.... one month at a time, or all the May-through-September
gains or losses - the advice just doesn't make good sense if you're playing
the odds.
Let's walk through
the monthly numbers first, to illustrate what we mean. [The calculations
below are based on the S&P 500's results since 1971.]
-
May - The
average gain has been 0.9% in May; the market made gains in 61% of them.
-
June - The
average gain has been 0.5% in June; the market made gains in 59% of them.
-
July - The
average July has been a break-even; the market made gains in 41% of them.
-
August - The
average gain has been 0.2% in August; the market made gains in 59% of them.
-
September -
The
average 'gain' has been a loss of 1.0% in September; the market only made
gains in 43% of them. [Taking the worst four Septembers out of the equation
though, the average September return turns positive.]
Those numbers alone
should have investors scratching their heads about how the axiom came to
be in the first place, but a little more number-crunching casts even more
invalidity on the idea of 'sell in May and go away'.
While it is
true that the middle portion of the year has a few more than its fair share
of tepid months, it's also true - and this is the unrecognized part - that
when investors do see a bad month at some point between May and September,
the
vast majority of the time it's counteracted by a positive month (or two)
in the very same year.
As evidence,
chew on this - since 1971, the market has made gains between May and
September 2/3 of the time, meaning it only lost ground in the middle
of the year 1/3 of the time.
So how is the
'average' May-through-September move actually an 0.8% loss going back the
last 39 years? A few very, very bad years have greatly skewed the
averages. Specifically, the May/September span saw big double-digit losses
in '74, '01, '02, and '08. If you take those catastrophic and highly
unusual years out of the calculation though, then the typical May/September
span turns positive again.
In other words,
the 'averages' may look ugly, but the actual odds of seeing losses over
the five month period are actually quite low.
Bottom line:
May through September may well be disastrous for stocks in 2010, but it
will have nothing to do with the calendar. It will have almost everything
to
do with the fact that the market is overbought following the ridiculous
11% runup over the last two months. Far more often than not, you're better
off just sticking with stocks betwwwn May and September, unless we're in
the most dire of bear markets. The middle of the year may not be explosive
for
the market, but progress to any degree is still progress.
|
|
A
Roadmap For the Pullback |
|
|
Ironically,
though we've just explained how selling in May and going away is generally
bad advice, in this particular May, it may make sense.... it's
got little to do with the calendar though.
Tuesday was
a rout - the biggest in months. Wednesday's and today's selling isn't quite
as harsh, but it's not undecided either; the sellers are coming out of
the woodwork. One daily dip can be chalked up to volatility... maybe even
two. But three? And still no sign of a bounce? It's the first time in a
long time the bulls didn't quickly step up to the plate again in search
of bargains.
While odds are
good the bulls will push back (to some degree) after today, at this point,
the damage is too great to mend quickly. Most likely, it's going to take
a hard landing to stop the decline. The question is, where might that
be?
 While
we're still advocates of the 'take it one day at a time' strategy
(for maximum flexibility in this unpredictable environment), there are
a couple of short-term technical floors for the S&P 500 that would
make ideal levels to at least reassess this new trend, since they may well
act as the rebound points.
The first
support area is currently around 1100. That's where we find the lower 50-day
Bollinger band (2 SDs) as well as the 200-day moving average line (black).
Both are important technical indications, and the Bollinger bands have
been especially persistent reversal points for the market over the
last several months. Keep in mind, however, that both are on the rise,
and may be at levels above 1100 by the time the S&P 500 can get there.
The second
possible floor for this pullback is more of a range as of right now....
the 1005/1035 area, framed in green. Though not shown on the chart, that's
where we'll find more than a few key Fibonacci retracement levels from
several high-low spans seen over the last fourteen months.
As the new trend
takes shape (or doesn't), we'll be able to add more details as to where
the precise bottom is apt to be. In the meantime, we remind you to not
jump to conclusions after the first bullish day - or even bullish days
- following the decimation of the last three days. It's going to take a
lot
to undo this damage and new trend, and the bulls just might not have what
it takes to do so on a permanent basis yet.
Anyway, be sure
to check the blog later today as well as tomorrow.... we'll be updating
our breadth and depth charts, as well as adding some sector-based charts.
|
Updated: 8:15 pm PDT (03:15 GMT), November 2, 2010
