I feel like this is confirmation bias, especially since it's a very large number of polls from reputable organizations being compared against one person's opinion.

Even if the true probability isn't 50% I doubt it's that far off.