(07-08-2020, 04:52 PM)Memento Mori Wrote:I agree with some of Steg's points, but I'll add: I'm not sure that there is any methodology that would produce an accurate ranking of who the best sim tester(s) is (are).
Team A may be able to drastically increase their win percentage against Team B compared to the default, but they obviously don't play against Team B's default, they play against whatever Team B decided to run based on their own testing that week. Team A may have tested really 'well' but got unlucky in that the strategy they ended up using matched up poorly with what Team B decided to use. Equally, Team A could test 'poorly' but luck out in that Team B might use strats that play into their hands.
The numbers and the concept are fun though, and memes are always good.
Oh no doubt I don’t think you can quantify it empirically. I just think you can definitely do a better job constructing a model than this did. Like I said originally, it’s still interesting and a fun read but I don’t think it’s nearly as accurate as you could potentially get.