Mr. Tippy's Tips
AFL match predictions powered by wisdom of the crowd — based on real tips from the Mr. Tippy Slack bot
About the Model
Mr. Tippy's predictions are based on a wisdom-of-the-crowd model. Rather than relying on a single algorithm, tips are aggregated from real footy fans across multiple Slack workspaces running Mr. Tippy competitions.
Currently, each tipper is weighted equally. For each match, Mr. Tippy predicts whichever team is tipped by the majority. Confidence is the percentage of tippers who picked the favoured team, so if the split is close, the confidence value will reflect that in a value close to 50%. In the case of a tie, confidence is shown as 50% and the predicted winner is taken from the highest-ranked tipster across all competitions.
As the number of tippers grows and individual scores begin to differentiate, the model can be weighted towards better-performing tipsters.
How Mr. Tippy Compares
Squiggle keeps a leaderboard of AFL prediction models, comparing correct tips, bits, mean absolute error and correct percentage. Bits, developed by Monash University Probabilistic Footy Tipping, measures how much information advantage a model holds over a coin flip, based on confidence and result (higher is better). Mean absolute error is excluded here as Mr. Tippy doesn't predict margins. See how Mr. Tippy compares to other models below.
| # | Model | Tips | Bits | Correct |
|---|---|---|---|---|