Starting from a premise of a 7 point lead for Clinton is building a castle in the air. Huffpost, which give some weight to trend lines, shows Clinton 53 and Sanders 38 — a 15% lead with the lead expanding since NH. Realclearpolitics which ignores trend lines shows Clinton with a ten % lead.
Starting from a false premise gets you wherever you want to go. But the details are also instructive. The post relies on an analysis by MattTX. Sanders is already well off the analysis of a presumed Sanders rally and victory in the results of SC and Super Tuesday (HRC was only supposed to widen her margin by 43 on 3/1). But the metrics of that analysis as well as the metrics of 538 will provide a basis for analyzing on March 15th whether Sanders is making any progress.
So under the MattTX analysis, if Sanders were to rally to even (which is not enough to catch up) you would expect the following primary results in northern states coming up: HRC wins Michigan by 5%, wins MS by 20%, Sanders wins Ohio by 2%, and loses MO by 8%, NC by 6%. The following week Sanders wins AZ by 2% and holds HRC to a 1% margin in Illinois.
Alternately, under the 538 analysis, in a 50-50 race, Sanders should hold HRC to 365 delegates on March 15th, and he needs to do better than that.
Kos’s position is that one evaluates the race on March 15th, and one sees if Sanders makes real, rather than fantasized, progress towards victory. Each primary he loses by a wider than expected margin increases the size of the blowouts required at the end to catch up. And as Armando states, the rules he proposes for after March 15th in the absence of Sanders progress are modest. At some point, if Sanders does not make progress significant enough to win, it is time to adjust to the concept that there’s more than a dime’s worth of difference between Trump and Clinton, just as there was more than a dime’s worth of difference between Gore and Bush.