Poli sci election prediction models
Posted by Joseph Knippenberg
A couple from this year. I can’t recall a time when they’ve ever been right. Perhaps others can correct that impression, if they’ve paid closer attention.
11:43 PM / October 17, 2006
: include(/srv/users/prod-php-nltashbrook/apps/prod-php-nltashbrook/public/sd/nlt-blog/_includes/promo-main.php): failed to open stream: No such file or directory in
: include(): Failed opening '/srv/users/prod-php-nltashbrook/apps/prod-php-nltashbrook/public/sd/nlt-blog/_includes/promo-main.php' for inclusion (include_path='.:/opt/sp/php7.2/lib/php') in
There was the series in PS about six months prior to the 2004 Presidential election and if I remember correctly, a couple of the predictions were pretty darned close.
I remember all the models in 1994--I was actually on a panel that featured some--saying that the Rs would 20-some seats in the House. They turned out, of course, to gain over 50. These models, which are moderately unreliable, are literally just for fun. The predictions serve no useful purpose and are at this point displaced by what we can see with our eyes about the particular races. No model could have predicted, for example, Allen’s strange behavior or Ford’s brilliant campaign.