Our model is probabilistic. This means the numbers outputted claim to know the likelihood of that event occurring. For example, as of writing this, the model predicts that there is a 0.93% chance that a third party candidate wins New Mexico. It is a possibility, but not probable. What is much more probable is the claim of Trump winning the state 47% of the time, which means, if you ran the state election 100 times, Trump would win 47 of those times. In no way does the model claim with absolute certainty, which is why even in liberal strong holds such as California, Trump still has a ~30% chance of winning. The model uses The 13 Keys, currently national and state polls, as well as a proprietary weighting method to compute the probability of each candidate winning that respective state.

Developed by Professor Allan Lichtman in 1981, The 13 Keys are statements that favor victory (in the popular vote count) for the incumbent party. When five or fewer statements are false, the incumbent party is predicted to win the popular vote; when six or more are false, the challenging party is predicted to win the popular vote. Something to take particular note of is that this model is used to predict the popular vote of the entire country rather than the race for electoral college votes in each state. In our forecast we use a modified version of the keys to apply a slight weight to polling averages, that within a reasonable margin of error would give a real world representation of Professor Lichtman’s model.