Here are the averages broken up by what I think are recognizable periods in New Hampshire:

Period	Clinton	Obama	Edwards	Richardson	Lead
Dec 11 - 18	35.25	25.75	15.5	6.75	9.5
Dec 1 - 11	30.8	27.8	15.3	6.8	3.0
11/27 - 12/1	33.8	25.5	16.8	9.8	8.3
11/1 - 11/27	36.3	22.3	13.5	8.7	14.0
October	41.0	21.8	12.5	6.5	19.2

It is interesting to note that the direction in Iowa is less clear. The ARG poll in Iowa found a 5 point shift to Clinton (from down 2 to up 3) and but is a little suspicious in that it shows Edwards at only 18 (no one has found him that low). In contrast, the Rassmussen poll this morning showed Clinton picking up only a point (from up 3 to up 4). While the averages look better for Clinton, this is in part attributable to the mix of pollsters. A CNN poll shows Clinton up 2 this morning, but has no prior benchmark. However, these polls do refute the idea that Clinton is somehow collapsing in Iowa.

As I have written before about New Hampshire, I am firm believer in the 30-35-40 rule for front runners. A front runner close to 30 is in serious trouble. A front runner at 40 in New Hampshire is a likely winner. With Clinton bouncing from 30 to 35, her numbers look signifcantly better to me. While Clinton's lead is too small to likely survive and Obama win in Iowa, her lead may be big enough to withstand an Edwards win. (though in the end I doubt it)

In any event the numbers show significant volatility - and a race still pretty open.