I wanted to use the 2017 UK general election to test an intuition, that instead of asking people for predictions about their own future behavior, we might be better off asking them to predict the future behavior of a friend. In short, I had a hunch that we might know the minds of others better than we know our own! Why? The behavioral sciences tells us, variously, that we can be unreliable witnesses to our motivations,1 that we can be over-confident in our ability to stick to intentions,2 that we are naturally expert at observing other people,3 that we often tell little white lies to protect our self-image4 and that, consequently, we are notoriously poor forecasters. Tetlock and Gardner (2015) show that most people are very poor at forecasting future outcomes, even within their own area of expertise.

I used voting intention to test my intuition because it involves asking a simple question that can be easily repeated and, crucially, can be validated against a definitive outcome—the general election result itself. In so doing, I also hoped to make a useful contribution to the debate about improving polling techniques.