Watch Small-Caps for Signs of Trouble - TheStreet

Editor's Note: Helene Meisler's column runs exclusively on RealMoney.com; this is a special free look at her column. For a free trial subscription to RealMoney.com, click here. This article was published Jan. 10 on RealMoney.

The averages reversed late Wednesday, and no one can

pinpoint exactly why. It wasn't as though some company warned about earnings; it wasn't as though the

Federal Reserve

hiked rates. It was as if buyers just stepped away.

We've seen that before, but is it different this time? Well, it's different every time, so there's no reason to compare one reversal day to another. If we look at Wednesday's reversal on its own, it did have some noteworthy negative connotations.

As I mentioned in

Wednesday's column, the Volatility Index, or VIX, is hovering at its lows. Not every reading in the low 20s will result in a market plunge, as we saw from May to August.

Someone pointed out to me that the

S&P 500

had fallen by 200 points in that time frame, so it's negative. I agree, but those 200 points happened over the course of several months, during which there were many ups and downs.

The point is that the market doesn't have to fall off a cliff. The VIX is a measure of complacency, and the market doesn't like folks to get complacent.

Investor's Intelligence

reported its weekly sentiment readings Wednesday. I expected to see the type of numbers we've seen all along in this rally: mediocre bullishness and mediocre bearishness.

But Wednesday's figures took me by surprise: Bulls now weigh in at 52.6%, and bears are at 22.7%. I won't spend a lot of time arguing the merits of this indicator because I always get a zillion questions about its validity as soon as I mention it. Suffice it to say that this indicator has been around for decades -- and for decades, I think it's worked.

At 52.6%, the bullish percentage is the highest we've seen since late July, just as the market was peaking. Yes, a reading above 50% is high, but by no means extreme. More importantly, look at the low percentage of bears. At 22.7%, that's extreme.

Note that I took the bearish percentage chart all the way back to early 1998 because that was the last time the reading was below 23%. We've seen low readings around that level since 1998 two other times: late July 2001 and July 1998. These two readings are important to note because they both led to extreme selloffs in the market. You're already familiar with the 2001 selloff, but the 1998 selloff is interesting for a different reason.

The advance/decline line had been on a tear in 1997 right through the first quarter of 1998. On April 6, 1998, the A/D line peaked, as you can see on the chart. Five days earlier, the percentage of bears bottomed out at 22.6%. At this point, the A/D line has been on an upward slope for just under a year.

Granted, the decline in the cumulative A/D line looks more than dramatic on this chart, but if we look at the S&P 500 in the same time frame, we find that the correction from that April peak was rather mild.

The S&P went from 1130 to 1070, about 60 points. Heck, we went from 1175 to 1125, about 50 points, in one week last month and no one made a fuss, so the decline doesn't have to be dramatic. In 1998, it wasn't until the market had a blowoff top in the summer (and the bears once again plunged to low levels) that the market had enough.

I don't know which area of the market will "give" this time around, but the statistics show that when we get a bearish percentage reading this low, some area of the market gives up. Sometimes it's the whole market, or as in early 1998, it was the breadth, but something usually gives up.

I'm watching one particular area closely right now: small-caps. That's where the love affair seems to be right now, so that may be the area most vulnerable to a bout of profit-taking.

It hasn't done anything wrong yet

, but I'm watching the Russell 2000 right here. Its 35% rise since the lows has been excellent, and it's now hovering at a very steep uptrend line. If that's broken, it could spell trouble for this index. A close below 490 would certainly be a warning of some trouble.

Overbought/Oversold Oscillators

For more explanation of these indicators, check out The Chartist's

primer.

Helene Meisler, based in Shanghai, writes a technical analysis column on the U.S. equity markets and updates her charts daily on TheStreet.com. Meisler trained at several Wall Street firms, including Goldman Sachs and SG Cowen, and has worked with the equity trading department at Cargill. At time of publication, she held no positions in any securities mentioned in this column, although holdings can change at any time. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. She appreciates your feedback and invites you to send it to

Helene Meisler.