The only thing I would do different is make it weekly, rather than daily because real life events like blizzards and such will impact daily sample size introducing variation that might generate incorrect conclusions later when we look back at this.
I mean missing data can never be fixed through averages. For example, if this storm was in Iowa, the state that appears to be Cruz's strongest, it might punish him more than the others. Likewise New York or New Hampshire, etc, for different candidates. On the other hand maybe it keeps people inside so they can participate more?
Seeing how this storm has completely shutdown the District of Criminals ( not to mention NYC ), I am not complaining whatsoever. I wish it would stay there permanently.
I understand your argument, but I want to voice an alternative view.
Trump is now leading the polls in Iowa. Doesn’t the bad weather impact him more? Does that help Cruz?
I want a daily consensus because I would run one or two Caucuses, and we’d be to February 2nd already.
I want almost a couple of weeks worth of days look before that.