| |
BGonline.org Forums
Win Rate as a function of PR difference - XG Feature
Posted By: AdamStocks In Response To: Win Rate as a function of PR difference - XG Feature (Maik Stiebler)
Date: Friday, 28 August 2015, at 4:05 a.m.
Sorry for my late response - I was away from computer for a few days.
If you compared the two types of server (unramped and ramped, perma or otherwise), and imagine both had a fixed initial player pool (let's just assume it's the same pool) with no further memberships allowed after launch day, and the players all played a very large number of matches, the max and min (in my and Bob's sense) and the average ratings of all players of both servers will approach the same values because the initial effect of the ramp on all players will have diminished to near-zero, such that the average ratings of any two given players will have approached the level that means that the difference in their ratings will give the appropriate MWC% that the (unramped) formula implies. So, in the limit, there would be no difference between the ramped and unramped server ratings, and therefore no inflation. But in a real server, players are continually joining after launch day, so the players can never get to the long run before new players join and yet again temporarily expand the max-min range. In a sense, the ramped server never gets much past launch day, because new players are being regularly launched. That's why the real server data such as the data I looked at gives a larger delta than for the purer, more theoretical system. Hope that clarifies it a bit.
| |
BGonline.org Forums is maintained by Stick with WebBBS 5.12.