The History Books Turn Another Page

11/07/08 4:20 PM EST

As of: SPX 1045 HELP  ARCHIVE

Another week, another mark in history.

We began the week with some disturbing short-term signs.   On Monday and Tuesday, we took a look at how Rydex mutual fund traders and those trading small odd lots had become quite enamored with the idea of a rally, something that usually precedes some trouble for equities.  Individual investors monitored by AAII also showed a big uptick in relief heading into this week, not an especially good sign either.

That led to Wednesday's rout, which continued right through into Thursday.  On Thursday morning, we looked at how performance in the overnight futures market since September has given us good clues as to how the regular trading session might unfold, and that proved to be a good guide the past two days.

By Thursday's close, the S&P 500 had suffered two consecutive 5% drops for only the second time in history.  The DJIA didn't drop quite as much, but the back-to-back declines were enough to place this week in the record books anyway.  Previous such one-two punches over the past 110 years have proved to be a good contrary signal in the intermediate-term, even during the very worst of markets, as the Dow was higher a month later 12 out of 13 times with a healthy average return.

Earlier today we discussed a "smart money" guide that had finally started to turn more positive.  Corporate insider purchases have picked up over the past couple of weeks, pushing InsiderScore.com's measure for S&P 500 stocks to a new multi-year record by a large margin.

So we have those positives, along with some good news from seasonality (when October has been bad and led to a poor first week of November, the rest of the year behaved admirably).  That's on top of all the other intermediate-term positives we've looked at over the past month (such as the multitude of historic oversold readings in mid-October and subsequent signs of explosive buying interest - two 7% rallies within a month of each other, an exceptionally low Arms Index reading and the quick flip in breadth).

Because of that, when the short-term negatives we discussed earlier in the week dissipated and our shortest-term guides fell well into oversold territory by late Thursday afternoon, I removed the short-side hedges that were protecting an (admittedly small) core intermediate-term long position.  I don't see much reason to change that at the moment, as we're currently in a kind of no-man's land technically and sentiment-wise, at least in the short-term.  I'm giving all of those longer-term positives a chance to work, and will continue to do so especially if the S&P 500 is able to hold above 890ish.

Have a safe and relaxing weekend and we'll see you here next week.

Insider Buying Picking Up Steam

11/07/08 9:10 AM EST

Good Friday morning...We begin the day with mixed trading in the pre-market futures after another volatile session.  The indices were showing large gains early this morning, but those have been erased following the jobs report.

Yesterday we took a look at the behavior of the S&P 500 during regular trading hours when the index futures made extreme moves during the overnight session.  Based on overnight trading heading into yesterday's regular session, we should have expected a bad day and we ended up getting it.

Today is different, as the futures have been quite positive.  They've given up most of their gains heading into and after the jobs report, but still the S&P was up more than +2% and hasn't lost more than -1% at any point (that may change in the next half-hour, but so far...).

So let's look at the exact opposite scenario from yesterday - other times since September when the futures were up at least +2% and never traded more than -1% overnight.  In those cases, during regular trading hours the index closed higher than the previous close 7 out of 8 times, averaging +5.2%, with the one loser being -1.4%.  So just like yesterday, we see that when the futures were extremely lopsided to one side or the other, that tended to bleed over into the regular trading session.

One of the missing pieces of smart money buying has been corporate insiders.  Both of the services we track, InsiderScore.com and InsiderInsights.com, have seen their data become more positive over the past couple of weeks, but haven't given buy signals yet.

For the InsiderScore.com data, the market-wide measure we track on the site shows that buying intensity has been picking up, and is now on a par with what we saw near the August and November 2007 market bounces, but still shy of the record set in March of this year.

Scrolling through their charts for indices and industries, I was struck by the data for the S&P 500.  For companies in that index specifically, buying interest has been heated, as the chart below shows.

Insiders in those firms tend to be net sellers, which isn't surprising given much of their compensation comes in the form of stock and stock options.  When their net buys and sells have turned positive in the past, showing more buyers than sellers, it has generally been a good buy signal.  The fact that it just spiked up to a new high certainly seems to be a good sign, though our history is unfortunately very limited.

On Friday, we discussed seasonality during the first week of November.  Over the past 10 - 15 years, it has been exceptionally positive.  However, we also took a look at those times when the previous month had been negative, and in those cases the results weren't nearly as good.

For the Dow in particular, there were three times it lost 10% or more during October, and the first week of November was down each time.  On the positive side, buying after the first week of November and holding through the first three trading days of the new year resulted in gains of +7.9%, +12.9% and +4.0%.

That dovetails fairly well with something we looked at yesterday afternoon, with the Dow suffering back-to-back down days of -4% or more.  In those cases, the Dow was up a month later 12 out of 13 times, averaging +10.8% (two months later, the stats were nearly identical).

The market this morning is showing a lukewarm response to the worse-than-expected jobs report.  It was showing a gap up afterwards, but historically a positive initial reaction hasn't been all that great going forward, when the previous day was down more than -1%.  When the S&P gapped up in those cases, three days later it was positive only 33% of the time (6 out of 18 occurrences).

If the S&P gapped down at least -0.5% on a jobs report day after losing -1% or more the day before, then buying the open and holding for three days resulted in 6 winning trades out of 7 attempts.  That is entirely in line with the other studies we've done over the years showing that extreme reactions to major economic reports tend to be a decent short-term contrary indicator, so ironically those hoping for higher prices probably don't want to see the market react well right out of the gate.

We've discussed quite a few intermediate-term positives over the past month, from a plethora of historic oversold readings in mid-October through the signs of powerful buying interest (such as two 7% rallies within a month of each other, the exceptionally low Arms Index reading and the quick flip in breadth from oversold to overbought).  When we see those signs of buying pressure coming on the heels of historically oversold levels, it has led to intermediate-term rallies with great consistency.

Given the short-term negatives we went over on Monday and Tuesday, I had hedged my (modest) intermediate-term long positions.  I removed them yesterday afternoon, and am back to about a 10% net long position in the S&P.  Our shortest-term guides are well into oversold territory for the first time since mid-October (except the Cumulative TICK), and so far potential support around 890 - 910 in holding on the S&P, for what that's worth.  If this can't hold, then it's a clear shot down to the October lows around 850, so today should hold some good clues.  We should be able to bounce back from early selling pressure, but I do not want to see a move much below 890ish.