How to Lose Credibility

Here’s the headline: Democrats flip 43rd state legislative seat since Trump took office

Now read carefully down to the 3rd from the last paragraph: “The 43 wins for Democrats have not been a net gain, however.”

What exactly is going on? Have democrats gained or lost legislative seats? How many?

I went to Ballotpedia for a count. I combine numbers from 2017 and 2018 to date.

Under the heading “Flipped Seats,” we find that 17 seats flipped in 2017, 14 from Republican to Democrat and 3 from Democrat to Republican, for a net Democratic gain of 11 seats.

For 2018, under the same heading we find that 10 seats have flipped in special elections, 9 in favor of Democrats and 1 in favor of Republicans, for a net gain of 8 seats.

Between the two sets of numbers we have a net gain of 19 in special elections. In addition, I found a net gain of 3 for the Democrats in New Jersey, and a net gain of 15 in Virginia (all in the House of Delegates). The net total would be 37.

My point is not where the other seats might be, but that the two statements are inconsistent. Is 43 flipped seats to the Democratic party net? Apparently not, and if I didn’t count net seats, I would be close to 43. But 37 net gain is still a net gain, even if not of 43 seats. Perhaps they mean that 43 is not the net number. So why not give us the net number? I’m not paid for this, so I’m not going to try to track down the rest of the numbers. Politifact is paid, and you can read what they found earlier in the year. Their text and then their rating illustrates why I tend to read them to raid their sources, but pay little attention to their final rating.

37 seats is interesting in itself, though the meaning can be debated. But this kind of loose reporting, with a headline that would suggest something different than the text and numbers that might (or might not) reflect something different than the text shows why the media is having a hard time getting accepted as fact checkers.

I think it is unfortunate that many Americans have gone from a biased source to sources without any moorings at all. But having your expectations trampled upon repeatedly does not make for confidence. Getting basic data right would be helpful.