Related questions
🏅 Top traders
| # | Name | Total profit |
|---|---|---|
| 1 | Ṁ1,648 | |
| 2 | Ṁ768 | |
| 3 | Ṁ754 | |
| 4 | Ṁ531 | |
| 5 | Ṁ329 |
https://fivethirtyeight.com/methodology/how-our-polling-averages-work/
538 dynamically adjust our average based on two different aggregation models. We want to avoid scenarios where there aren’t many polls for a while, then a deluge of new data creates a whiplash in the average. To help with that, when we have more data in a recent time period, we rely less on our slow-to-update exponentially weighted moving average and more on our more aggressive polynomial regression trendline. This also has the benefit of giving us averages that are more responsive to quick movement in the data when multiple polls reflect that movement, without reacting too aggressively to any one individual survey showing a big change.
538 makes loads of adjustment to either an exponential weighted moving average or a polynomial regression. Which means, no one has any specific idea how the averages move with any particular poll.
The h2h is harder to predict because they don't release the actual average like they do with the Presidential approval polling average. However, if you track the movement of the average each time a new poll goes in, you can reverse engineer the black box to the point that when I see, for instance, a poll like the NYT come in, I know that even the house effect adjustments aren't going to stop it from helping Trump.
Ergo, someone (me) has a pretty specific idea how the average moves with a particular poll. In fact, with the approval average, I can usually predict the number within 0.03.