A thought experiment on the impact of political TV ads

Gerry Nichols noted that a recent survey didn't do a very good job passing along the information about the newest slate of political ads: namely whether it changed anybody's minds or which demographics it influenced the most.

I made a joke that I could show stock footage of a river and then find that after viewing it 27% of the viewers supported the Liberal Party. Just in case the link rots away, a recent EKOS poll puts the Liberals at...27% In other words, the ad didn't change anybody's mind, and I could talk about how the Conservative/Liberal/NDP breakdown in the viewership was 30/27/29 but as a political ad it was garbage: it didn't change any minds.

Anyways, that was the joke. But I got to thinking...would it work?

If you got three rooms with 1000 people each, polled them about which political party they would vote for (if any), and then played the same stock footage followed only by an ad for one of the three political parties (a different party logo for each room), would it change their minds?

The conventional wisdom should be "no", and you're probably right. In an ideal world, it would be no. This lazy video didn't tell me about Party X's campaign, nor did it say what Party Y or Z would do or why I should approve of any of them.

And yet, somehow, deep in my being I have this sinking suspicion that it would make a difference. I think that afterwards you would see that some, maybe not a lot, but some of these 1000 people would change their vote to support the party whose logo was on the ad. Even more insane, I'm going to guess that each party would find different numbers of people changing allegiance based on this "ad".

I'll go on a limb here and say that the further left the party, the more this nonsensical ad will change their vote.