Here is an update to the spreadsheet looking at the delegates and demographics of the states Bernie Sanders has won. Again, the delegate data is taken from the New York Times and the demographic from the US census.
The top level results from looking at the spreadsheet is that Sanders did better than expected in NY. Sanders has never won a state where Black African Americas composed more than 10% of the population and there was no reason why New York would be different. As Al Sharpton said on NPR on the morning of April 20, Sanders does not do well with any group of AA, even though he does better with millennials, and that costs him primaries. While Winning New York would have supporting Sanders’ thesis that only the unsophisticated and conservative southern AA are voting against him, this did not play out.
While he did what he needed to do in New York, his performance in Wyoming and Wisconsin do not auger well for the future. The problem for Sanders is that Clinton is wining big diverse states and steadily increasing her lead, while Sanders has not won big since Utah and Alaska. Next week Maryland and Pennsylvania award 20% of the remaining delegates, and even if Sanders ties these two states and wins the remaining by 20 points, he will likely only net about 30 delegates. If Clinton wins Maryland ans Pennsylvania by 15 points, she could walk with 50 extra delegates.
I have also updated the spreadsheet to answer the question of fairness. Specifically if there is a bias toward one candidate or the other in terms of awarding of delegates. The quantifiable question I ask is if the fact that small states have proportionately more delegates has resulted in one candidates being awarded a disproportionate number of delegates. The small state advantage is real, as show by the population rate on the spreadsheet.
In order to quantify this small state advantage, I calculated the number of delegates each state would have if we simply awarded a delegate per 100,000 persons in the states, based on the US census count. This is about the distribution used in large states. I then took the percentage of delegates that Sanders won in each state and awarded him that percentage of the proportional delegates. I then scaled this total back to the existing number of delegates so the delegates for each candidate could be compared. At the end of the analysis it is shown that Sanders is awarded about 40 extra delegates due to a proportionally larger delegate count for smaller states.