Hey Mark, thanks for taking the time to read my piece and give me such valuable critiques.
As we’re both aware, it’s been difficult to acquire any sort of data to validate certain concepts (such as the impact of group matchmaking abuse) other than word of mouth or personal experience. Without any sort of match history data, or even a quantifiable amount of SR that is gained or lost per match, this only further complicates matters. In regards to the group queue abuse point, I always made a point to ask much higher rated players in the post-game lobby how much SR they gained/lost— obviously I can’t validate what they said, but it seemed to be a substantial enough trend to warrant bringing up. I’m glad to hear that personal gains/losses are based on the personal SR vs enemy MMR differential, maybe those numbers could use another look to see if they’re in line with what people consider to be fair.
Your assumption that performance modifiers are being calculated based on percentiles is completely reasonable, but I don’t necessarily believe that such a system works fairly unless players are consistently playing with and against players of similar skill levels. An interesting example that makes me question the validity of performance modifiers is that when a very good player plays against others who are significantly lower MMR, it will be much easier to consistently outperform them (thereby cranking up their stats and boosting the performance modifier greatly). Based on the way the numbers fall, this could mean that a player can realistically be gaining/losing the same amount of SR carrying in low MMR games when compared to being average or middle of the pack in their current SR, however, their win rate would theoretically be much higher.
(TLDR: Reduced SR/MMR from sandbagging is offset by their ability to more easily carry, resulting in similar ELO gains/losses, but at a much higher win rate). Again, all of these things are contingent on the weightings of MMR/SR differential and Performance Modifiers, but the outcomes have simply become too wonky to ignore.
Lastly, I understand what you mean when you say that I began to devolve from the central thesis of the article, I’ll be sure to address that sort of issue in future pieces. This was clearly a lengthy one, and I found it hard to be thorough while not addressing some of the other pervasive issues in the game currently.
Thanks again for the critiques, cheers!