Beware Biased Endorsements

Test Your Knowledge: New York Times Editorial Board

The New York Times endorsed 12 candidates in last week's mid-term elections. What would you guess is the probability that all 12 of the candidates they endorsed were Democrats?

a) 4095 to 1 (about .02% chance)
b) 100% chance
c) No way to tell - they evaluate each candidate on his or her merit, without regard to party affiliation.

Score:

If you selected (a), give yourself 1 point. This is the mathematically correct answer, assuming they planned to evaluate each candidate on his or her merit, without regard to party affiliation. But you only get 1 point, because this was not the correct answer.

If you selected (b), give yourself 2 points. This turned out to be the correct answer as I myself witnessed on the New York Times website endorsement page prior to the election.

If you selected (c), give yourself 0 points. You clearly do not have any familiarity with the New York Times.


What is the point of the above? The point is that you should not go to the New York Times for thoughtful analysis of elections. It is equivalent to asking a cheerleader for a High School football team, or the parents of the quarterback, which team they think should win. They will give you an honest answer, but it is completely impossible for them to remove their bias. And thus it is a mostly worthless exercise to consider their predictions, if you are truly trying to determine which side is better than the other.

By the way, if you think this endorsement thing is just an unusual phenomenon related to maybe President Bush, or the war in Iraq, did you know that the New York Times hasn't endorsed a Republican candidate for President since Dwight Eisenhower in 1952 (54 years ago)?