In today's column, I talked about a colleague's take on offense vs. defense. When the defensive stuff like health care and staples outperform chances are the market is feeling a little ill. Conversely, when the offensive stuff, like tech and cyclicals, chances are that the market is feeling better.

The offense/defense ratio I put up shows the short-term rising trend hitting the long-term declining trend either now or very soon, depending on how strict you are with the trendline construction. That suggests that the bounce rally's days are numbered.

But it also leaves the door open for a breakout through the long-term trendline. I can't see that happening unless still more technical beliefs must be set aside until the market is finally stable again.

Check out this chart of the old Morgan Stanley consumer and cyclical indices (they make me say the unpalatable "indexes" in all my DJ work).

What I see is the same sort of short-term term saint meets long-term devil. But unlike the index in the Barron's Online column, this ratio set a lower low last month. I have no real conclusions but there is one more thing to note - a big old textbook bullish RSI divergence.

It's almost as if this chart is saying that the cyclical stuff will break out relative to the consumer stuff. The problem is when we look back in time.

Now we can see that the consumer stuff beat the cyclical stuff during the 1994-2000 rally. So what gives? Is this not a sign that the bear market rally has legs?

Well, how about this? EVERYONE (caps intended) says the economy was consumer driven all this time. There was a respite in 2002 as the last bear market healed but then the cyclicals drove the market during the last bull run.

Not the consumer? Hmmm.....

And now it looks like it could, repeat could be happening again. Don't forget this is not the offense/defense stuff as the other index was. This is consumer stocks vs. industrial cyclicals. It just might pay to look at cyclicals leading the next bull and not the consumer stuff.