| To get rid of the unfair scores ('gambling effect'), when the first and second predicted athletes finish in reverse order, is decided which alternative scoring model will be used for the 2015 contest. After an evaluation a variant of the model as proposed by one of our participants Sampsa Lahtonen is recognized as the most fair and with the least unwished side effects. He calls the chosen variant the 1-1-2 variant. You get 1 point for each athlete predicted finishing in the top1, 1 point for each athlete predicted finishing in the top2 and 2 points for each athlete predicted finishing in the top3. Fundamentals as the bonus for predicting all medalists in arbitrary order is maintained as well as receiving a 10 score for having everything predicted well. The evaluation with the other serious alternatives also solving the issue with their pros and cons will become available soon. | |||||||||
| Old model (used till 2014) | New model (used for 2015) | ||||||||
| podium | podium | ||||||||
| prediction | g | s | b | bonus | prediction | g | s | b | bonus |
| g | 4 | 1 | 1 | - | g | 4 | 3 | 2 | - |
| s | 1 | 3 | 1 | - | s | 3 | 3 | 2 | - |
| b | 1 | 1 | 2 | - | b | 2 | 2 | 2 | - |
| all | - | - | - | 1 | all | - | - | - | 1 |
| Main differences are: 1. The new model rewards having predicted an athlete in the right order above having predicted an athlete on the right spot. This model solves the gambling effect of the old model in special of a head-to-head between 2 athletes for gold. 2. The new model has a side effect that it will become more important to have a medalist predicted vs not having predicted. 3. The new model has a side effect that is does not differentiate if an athlete is finishing higher on the podium as predicted. In some circumstances this is more fair in others this is questionable. | |||||||||
| Scoring model for 2016 - today | |||
| podium | |||
| prediction | g | s | b |
| g | 5 | 3 | 2 |
| s | 3 | 3 | 2 |
| b | 2 | 2 | 2 |