It looks like we'll begin the day with a gap down that has ranged between "big" and "huge".  Even after a mild recovery, the S&P futures are still down 7 points from yesterday's close as I type.

I went back and checked for any time breadth has been this oversold, followed by a gap down more than -0.5% on the open.  As a proxy for "breadth", I'm using the 10-day average of the Up Issues Ratio.  That ratio is currently at 36%, meaning that during the past two weeks, only 36% of all issues closed up on the day on a daily basis.

The precedents were rather astounding: 10/19/87, 10/22/87, 9/28/90, 4/4/94, 9/21/01 and 7/24/02.

If you pull those up on a chart, they mark some of the very best buying opportunities of the past 20 years.  The one exception is 10/19/87, of course, which was Black Monday, when prices kept dropping dramatically throughout the day.

This is make or break time here.  Extremes like we saw last week led to "very" returns going forward - either very good or very bad...there was no in-between.  This morning is the trigger, and we should either see a reversal sometime during the day that leads to an intermediate-term rally...or we don't, in which case I would be significantly discouraged.

The key for me will be the morning low.  Assuming we continue the gap down, we will form a low sometime this morning.  Whether it occurs on time (somewhere within about 15 minutes of 10:30am EST), I don't know in this kind of situation, but at some point the buyers will attempt to give it a run.  If that bounce fails and we set lower intraday lows, then I will not be trying to step in front of the train by buying.  I would only do that if we saw a heavy reversal going into the afternoon hours.

One of the other big doomsday patterns being tossed around now is the "outside bar" on the monthly chart of the S&P 500 (^SPX).  The index hit a new high this month, with a higher high than we saw in June, and now it's going to close the month lower than June's low.  Lower than May's low, too, for that matter.

So just to see how much of a doomsday pattern this is, I checked for other times the index hit a new yearly high, then suffered so much of a reversal that it closed below the low of the prior two months.

Since 1950, it has happened three times (October 1979, January 1990 and November 1991).  The following month, the S&P was positive all three times by an average of +5.4%.  Four months later, the sweet spot for these occurrences, also showed a positive return all three times for an average of +9.7%.  All of them were positive by at least +7.5% over that time frame.  The average amount that the S&P declined during those four months was only -1.8% compared to an average maximum gain of +13.5%.  Wow.

Three instances isn't much of a sample, so I went back and checked the last 108 years of the Dow Jones Industrial Average (^INDU).  I show eight occurrences for that index, and it bounced back the next month seven of those times by an average of +2.7% (the one negative return was -1.5%).

Using the same four-month window as the S&P, I show six positive instances out of the 8, with an average return of +6.2%.  On average, the most the Dow went against us during those four months was -3.8% compared to an average maximum gain of +10.0%.

Not exactly a doomsday signal if you ask me, and if you combine that with the breadth statistics we've seen, it doesn't seem like the prudent thing for investors to do is panic - quite the opposite, in fact.  For shorter-term traders, it might be a different story depending on how today goes, so let's focus on the morning activity and see how the initial bounce is treated.