Hello everyone !

Many players may have abandoned the idea of rating the difficulty of a map, or rating accurately PBs. As a matter of fact, there is no absolute way of doing so, but you would be surprised of what you can do with some math and statistical analysis !

The idea here is to predict how records will be distributed on the timeline, and how this distribution will change over time, with as little data as possible. I am still in the early stage of this analysis, and there is one thing that I miss: a real-scale experiment.

Dedimania top30 is a good start but it doesn't lead very far, finding the total number of runs on a map is hard, and the average time is nowhere to be found.

That is why I am calling on willing server admins to host this experiment, which remains very simple. Ideally, I would like:

1. A new map, very classic (meaning normal style or slightly tech), 30s to 1min, and a normal server (not team-only, not too high level players, to have a representative and large enough sample).
2. Having access to all the records. I am not an expert, please tell me if there are any technical problems with that. All I know is that some servers have a webpage, that is perfectly fine if I can copy+paste.
3. A refreshing rate allowing me to check day after day the evolution.
4. The possibility to follow individual progression. Players can remain anonymous of course, it's just statistics, no need for the real name. The goal is to identify what proportion of players improve their record by how much.

I do not know yet how long the experiment will last. That is simply because I have no idea how long it will take to obtain enough times to see a tendency, it depends on your server. For now I would like to focus on one experiment, but of course in the future it would be even greater to repeat it on different environments, styles, length, etc.

I like to think that in the future, you will be able to say "I am ahead of 80% of records on this map, that has a difficulty level of 6,7" with a plugin like dedimania (maybe by dedimania), that registers only a few variables such as the average time and the standard deviation. If you share this thought with me and you would like to host the experiment, please contact me by e-mail ! I will gladly share the results with the community as soon as I can !

my e-mail adress: vipix.lt@gmail.com