ajtj, the only argument I have against the three scenarios is that the late July bottom was an extreme in negative sentiment readings, worthy of the beginning of a cyclical bull move with a double bottom 10/9 in which the sentiment indicators were far from those reached in late July. In many historical cases (and yes, history does not repeat itself, and sep 21 2001 was similar). If the current run up is aborted here and negative sentiments increase (though not as much as in either July or October) appreciably, there is a strong possibility that the cyclical bull case could reestablish itself. For that, we will need not to close above 1515 in the near future, and the retrenchment must be deep enough to once more bring fear in the market, but without much deterioration in leadership. I can't in a short post point out to all the parameters I am looking for and what levels I would prefer (I should say my regression equations would prefer). For now, however, I am staying bearish, but because I am a natural optimist, I am rooting for a sharp three weeks decline to the 1220/1263 area, if that (and few other things happen), I may turn bullish again. If some of the other things do not happen, one of your three scenario could indeed take place, and I hope that by the time we get to 1300 or so, I'll be able to see if the cyclical bull is still in play or not.
On the other hand, if we continue up here, I think the max we go is the 1619 area (the 4/01 and what was "supposed" to be the Sep 16/01 bottom (sans 9/11)). That probably in two steps, a decline from 1515 or so to the 1426 area (probably a goof 20 Naz points above) and then a final run to 1619 by early January, from there, a rough, though not exact, copy of the events in 2001 from January to late March/mid May. Right now, I don't have anything worse than 950 (though a drop from 1600 to 950 is going to be extremely painful).
Zeev
