@Habreno I have been following the most spedrun games by various measures for over a year. On May 2, 2019, SR.com said that SMO had 1622 players and OoT had 1450. On September 1, 2019, SR.com said that SMO had 1753 players and OoT had 1486. These numbers matched the numbers on the statistics page under the game. Thus in 4 months SMO had gained 131 players and OoT had gained 36. Today SR.com says SMO has 2807 players and OoT has 953. That means that in 3 months SMO had to have gained 1054 players and OoT had to have LOST 533 players. As someone with a strong following in SMO I can tell you there's no way it could've hit numbers like those this fall. And there's been no massive ban or any reason OoT would have lost anywhere close to that many players. However if we go to the stats page under the games we see SMO has 1808 players and OoT has 1513 players. This would mean in the past 3 months SMO gained 55 players and OoT gained 27 players. These are much more reasonable numbers.

If the numbers aren't "glitched" and are in counting players differently, how do they count it? Because I can't think of a single way to measure it so SMO has only 23 less players than SM64 and OoT has less players than Getting Over It With Bennett Foddy.

So there is some bug with the sorting algorithm. If you view all games by total players like here https://www.speedrun.com/games#platform=&orderby=mostplayers&unofficial=off you will see Super Mario Odyssey has 2800 players and Ocarina of Time has 950 total players. However under the game's pages, Super Mario Odyssey has 1800 total players here https://www.speedrun.com/smo/gamestats and Ocarina of Time has 1500 total players here https://www.speedrun.com/oot/gamestats