Khobai, on 19 January 2020 - 12:26 PM, said:
I dont doubt that youre right. My point was simply that there is no way for you to prove it. You are simply asking me to take your word for it which does not constitute actual proof. The inability to distinguish between solo and group games is one of the inherent limitations of jarl's list.
And yes good players will have better stats in the long run. I am not refuting that. My point was simply that jarl's list has limitations and as such should not be used as the sole means of determining if a player is good or not. And your stats certainly should not determine the validity of your opinion or justify behavior like statshaming. I find it hilarious that players like Ash measure people's value by their MWO stats, because their only way of validating themselves in life is how good they are at MWO. Whereas most normal people simply dont care about stats one way or another.
Right, Jarl's has limitations, as does anything. GroupQ is dead so that is not a limitation worth giving much thought unless you are looking at data from before something like July 2019. Any data since about July '19 is largely uncontaminated by GroupQ because the queue is dead.
Also, we conveniently have several other places to test our mettle, see Solaris and numerous player run competitions (ISC, MoR, MWOWCS to name a few).
I wouldn't give much stock to the opinions of someone who hasn't performed well in at least one of those areas, and it's not like there are a whole lot of high performing comp or Solaris players with horrible, or even unexpectedly low, stats on Jarl's. A player with low stats might really just have bad mechanicals skills and understand the game at a high level, but my experience as an erstwhile potato who has played against and learned from some of the top players in the game is that the higher the performance of the player the more likely they are to have a deep understanding of game mechanics and strats. That's partly that high level players have, at some point at least, devoted significant time to building that knowledge base, and because they notice things and connect dots that others might not. It's also due to the fact that high level players active in comp have a chance to test ideas against other high level players, which gives a better sense of the absolute merit of a build or strat.
I am sure I missed reasons high level players are far more likely to understand the implications of changes to the game, or understand how to build a mech or approach a particular scenario, but that's at least a start.
Any time we deal with data there are limitations. You seem to be of the mind that having any limitations in a dataset renders it invalid. That's just wrong. We've identified the issues with the data, and their likely impacts, AND that the GroupQ issue is dead along with the queue.
BTW stats or in-game achievements are just that, stats and in-game achievements. They don't mean anyone is worth more than others (though they can give some sense of who I would prefer to drop with in a competitive drop!). I've played both with people who are much better at the game than me, and significantly worse. I respect and value all of those folks.
Edited by Brauer, 19 January 2020 - 06:12 PM.