Top-25
Column
Top-25 Players in Elo Rating
| Name | Elo | dElo |
|---|---|---|
| Paul McBeth | 1890.0 | 7.9 |
| Richard Wysocki | 1888.5 | 3.2 |
| Calvin Heimburg | 1867.8 | 11.2 |
| Gannon Buhr | 1861.5 | 13.9 |
| Chris Dickerson | 1855.1 | 3.3 |
| Kyle Klein | 1852.8 | 8.3 |
| Matthew Orum | 1851.3 | 10.8 |
| Joel Freeman | 1840.2 | 8.3 |
| James Conrad | 1832.2 | 9.0 |
| Kevin Jones | 1831.2 | 3.2 |
| Chris Clemons | 1822.2 | 3.9 |
| Eagle McMahon | 1819.2 | 5.6 |
| Simon Lizotte | 1818.5 | -7.0 |
| Corey Ellis | 1817.8 | 3.1 |
| Bradley Williams | 1814.0 | 6.5 |
| Niklas Anttila | 1812.6 | 19.0 |
| Ezra Aderhold | 1811.4 | 4.9 |
| Isaac Robinson | 1808.4 | 14.1 |
| Alden Harris | 1807.8 | 8.7 |
| Nikko Locastro | 1806.7 | 0.3 |
| Andrew Marwede | 1806.1 | 4.8 |
| Andrew Presnell | 1804.8 | 5.8 |
| Adam Hammes | 1804.5 | 2.3 |
| Anthony Barela | 1803.7 | 7.5 |
| Väinö Mäkelä | 1801.5 | 6.0 |
Column
Bar Chart of Top-25 Players (float cursor over bars for more info)
Data Exploration
Column
Plot of Top-25 Players Over Time (float cursor over line(s) for more info)
Win probabilities
Column
Win probabilities for the next DGPT or Major event.
| Players | Win Prob (%) | Elo |
|---|---|---|
| Paul McBeth | 10.9 | 1890.0 |
| Richard Wysocki | 10.7 | 1888.5 |
| Calvin Heimburg | 7.7 | 1867.8 |
| Gannon Buhr | 6.9 | 1861.5 |
| Kyle Klein | 6.0 | 1852.8 |
| Matthew Orum | 5.8 | 1851.3 |
| Joel Freeman | 4.8 | 1840.2 |
| James Conrad | 4.1 | 1832.2 |
| Kevin Jones | 4.1 | 1831.2 |
| Chris Clemons | 3.4 | 1822.2 |
| Eagle McMahon | 3.3 | 1819.2 |
| Simon Lizotte | 3.2 | 1818.5 |
| Corey Ellis | 3.2 | 1817.8 |
| Bradley Williams | 3.0 | 1814.0 |
| Ezra Aderhold | 2.8 | 1811.4 |
| Isaac Robinson | 2.7 | 1808.4 |
| Alden Harris | 2.6 | 1807.8 |
| Nikko Locastro | 2.6 | 1806.7 |
| Andrew Marwede | 2.5 | 1806.1 |
| Andrew Presnell | 2.5 | 1804.8 |
| Adam Hammes | 2.5 | 1804.5 |
| Anthony Barela | 2.4 | 1803.7 |
| Väinö Mäkelä | 2.3 | 1801.5 |
Column
Column
Methodology:
These win probabilities are experimental. The probabilities are generated using a logistic regression and are based exclusively on player Elo rating. The better a rating the better the probability of winning an event.
Below is a plot of win probability by Elo rating. The probability follows a sigmoidal curve. FLoat cursor over points to see player name, rating, and win probability.
Top-25 (FPO)
Column
Top-25 Players in Elo Rating
| Name | Elo | dElo |
|---|---|---|
| Kristin Tattar | 1873.6 | 9.7 |
| Paige Pierce | 1853.5 | 2.1 |
| Catrina Allen | 1844.1 | 12.8 |
| Missy Gannon | 1840.7 | 8.3 |
| Ohn Scoggins | 1827.4 | 3.2 |
| Henna Blomroos | 1826.1 | 13.4 |
| Eveliina Salonen | 1825.4 | 7.8 |
| Sarah Hokom | 1807.8 | 8.7 |
| Valerie Mandujano | 1801.7 | -10.7 |
| Ella Hansen | 1794.5 | 7.4 |
| Kat Mertsch | 1789.1 | 4.3 |
| Hailey King | 1780.2 | 16.8 |
| Macie Velediaz | 1773.7 | 8.4 |
| Natalie Ryan | 1773.4 | 15.4 |
| Holyn Handley | 1767.8 | 15.9 |
| Jessica Weese | 1763.7 | 3.8 |
| Jennifer Allen | 1757.2 | 16.2 |
| Lisa Fajkus | 1752.5 | -2.2 |
| Madison Walker | 1752.4 | 5.4 |
| Alexis Mandujano | 1752.3 | 5.9 |
| Deann Carey | 1750.7 | 2.5 |
| Heidi Laine | 1744.6 | 13.9 |
| Rebecca Cox | 1743.5 | 5.8 |
| Maria Oliva | 1742.7 | 1.5 |
| Juliana Korver | 1739.3 | 9.3 |
Column
Bar Chart of Top-25 Players (float cursor over bars for more info)
Data Exploration (FPO)
Column
Plot of Top-25 Players Over Time (float cursor over line(s) for more info)
Win probabilities (FPO)
Column
Win probabilities for the next DGPT or Major event.
| Players | Win Prob (%) | Elo |
|---|---|---|
| Catrina Allen | 9.2 | 1844.1 |
| Missy Gannon | 8.8 | 1840.7 |
| Ohn Scoggins | 7.6 | 1827.4 |
| Henna Blomroos | 7.5 | 1826.1 |
| Eveliina Salonen | 7.5 | 1825.4 |
| Sarah Hokom | 6.1 | 1807.8 |
| Valerie Mandujano | 5.6 | 1801.7 |
| Ella Hansen | 5.1 | 1794.5 |
| Kat Mertsch | 4.8 | 1789.1 |
| Hailey King | 4.3 | 1780.2 |
| Macie Velediaz | 3.9 | 1773.7 |
| Holyn Handley | 3.6 | 1767.8 |
| Jessica Weese | 3.4 | 1763.7 |
| Jennifer Allen | 3.1 | 1757.2 |
| Lisa Fajkus | 2.9 | 1752.5 |
| Madison Walker | 2.9 | 1752.4 |
| Alexis Mandujano | 2.9 | 1752.3 |
| Deann Carey | 2.9 | 1750.7 |
| Rebecca Cox | 2.6 | 1743.5 |
| Maria Oliva | 2.6 | 1742.7 |
| Juliana Korver | 2.4 | 1739.3 |
Column
Column
Methodology:
These win probabilities are experimental. The probabilities are generated using a logistic regression and are based exclusively on player Elo rating. The better a rating the better the probability of winning an event.
Below is a plot of win probability by Elo rating. The probability follows a sigmoidal curve. FLoat cursor over points to see player name, rating, and win probability.
Methodology
Elo Ratings were invented by Physicist Arpad Elo in the mid-twentieth century as a way to measure the relative skill of chess players. They are now used for a wide variety of sports (including esports).
Elo ratings were originally designed for measuring players of one-on-one sports. However, I have been working since ~2017 to modify the metric for disc golf (see this article for more details).
The two equations below, which I use for this metric, are pretty much the standard Elo equations. ES stands for expected score, S is actual score, and K is the k-factor, which is a multiplier for the amount a rating will change per round. I am currently using a k-factor of 20.
\[ Elo_i=Elo_{i-1}-K*MS*(S-ES)\] \[ES=\frac{1}{(10^{-(Elo_n-\overline{Elo})/400}+1)}\]
S is a standardardized actual score that varies between 1 (best score of the round) and 0 (worst score of the round).
MS (margin stabilizer) is an additional (and novel) multiplier that influences how much ratings change per round. However, unlike k-factor, which is the same for all players, the MS value is different for each player and based on how a player’s rating compares to the average rating of the field. If a player’s rating is low compared to the field, a very good round will cause their rating to go up a lot, but a bad round will not cause their rating to decrease. The opposite is true for players with high ratings. So it has the largest effect at the margins, hence the name.
\[MS=[e^{S-ES}]^{-(Elo_n-\overline{Elo})/100}\]