"There ’s a special providence in the fall of a sparrow. If it be now, ’t is not to come; if it be not to come, it will be now; if it be not now, yet it will come: the readiness is all. Since no man has aught of what he leaves, what is ’t to leave betimes?" Hamlet V.2
The paper examines the timing of shifts in presidential vote sentiment over 5 previous election cycles, going back to 1981. In all of them the crucial turn seems to occur around mid-January to early February. And what we see now is two striking convergences: Hollande's early lead in the first round is evaporating, as his vote converges with that of Sarkozy. Marine Le Pen has begun a slow decline from her peak, which came in late November, while François Bayrou has been rising sharply since early November.
These trends are contradictory in import. Le Pen's slide, if it turns out to be real, obviously profits Nicolas Sarkozy. But Bayrou's rise is ambiguous. As in 2007, voters may be turning to the centrist because they have never liked Sarkozy but, having looked over the Socialist candidate, aren't pleased with what they see and want to keep their options open. On the other hand, the Bayrou total may be swelled by center-rightists who, when faced with the moment of choice, can't accept another five years of Sarkozy. No doubt there is a mixture of both in the MoDem vote, and we won't really know how it shakes out until the second round.
Some strategic maneuvering may lie ahead as well. Bayrou, who apparently contemplated a deal with Royal in 2007, might be more likely to go through with a bargain with Hollande, a more predictable partner than Royal would have been. It's hard to imagine a Bayrou-Sarko deal, since Bayrou has spent the last five years attacking every aspect of Sarkozy's reign, but, well ... Paris vaut bien une messe.
Interesting to speculate, but of course hindsight--which is what Le Monde's charts represent--is always more perspicacious than foresight.