# How are the odds in FiveThirtyEight's election forecasts calculated? [closed]

How are the odds in FiveThirtyEight's political election forecasts calculated?

For example, Bernie is predicted to win the majority of votes with odds of "1 in 2 (46%)" as of 2/23/2020. This is neither 1/(2+1) (about 33%) or 1/2 (50%).

Two days later, his odds are "2 in 5 (43%)". Again, this is not actually 2/5 (40%).

Where do these odds come from and why don't they match the percentage?

• "1 in 2" is an approximation of 46%. Feb 23 '20 at 19:27
• What does "1 in 2" mean?
– Tim
Feb 23 '20 at 19:28
• "1 in 2" means (approximately) 50%. Feb 23 '20 at 19:29
• This appears to be referring to journalistic decision of 538.com and others. Expressing as percentages gives an unrealistic level of precision (46%, and not 45 or 47!) and puts too much emphasis on "the prediction" and not the probability. To counteract this 538 and others have been expressing chances as "1 in 2" to mean "about 50%" 46% is roughly 1 in 2. It could also be roughly 3 in 7 or exactly 23 in 50, but since 46% is only an estimate, the fractions are kept as simple as possible. However this is just one journalist's algorithm. There is no particular rule. Feb 23 '20 at 20:26
• @dandavis - The percentages are very precise in the sense that they are the (probably rounded) results from running their model n number of times and in 46 out of every 100 runs, Bernie wins. But they're also imprecise because the model has a margin of error in its assumptions and output. So they're the exact results of a series of guesses, and thus shouldn't be interpreted as exact predictions. But they're not made up. Feb 26 '20 at 0:03