Why Dem Primary Anomalies Must Be Thoroughly Investigated Before Choosing a Nominee

[Time for change]

8. It's ridiculous to say that they aren't more accurate than pre-election polls

It isn't the size of the sample that makes them far superior to pre-election polls.

It's the fact that: 1) They assess who people actually voted for rather than who they intend to vote for at a later date; 2) They don't need to rely on models that may or may not be reliable, which attempt to determine who is a "likely voter", since they only assess people who actually voted, and 3) it is far easier to get a random sample or close to a random sample of voters than in pre-election polls because they don't rely on telephone sampling, which is likely to miss certain demographics of people who don't have landlines or cannot be contacted for whatever reason.

The discrepancy in NY was huge, and far beyond the margin of error, which takes sample size into account.