Ranked matchmaking, the LP system, and report shielding

aspacecodyssey·8/15/2019, 6:20:03 PM·2 votes·1,045 views

I'm in the middle of the all-too-standard experience of feeling like in my last ~20-50 games, the number of games in which my team had an AFK/inter/troll was much higher than the number of games with that player on the enemy team. One of my most recent games had an AFK ADC, we voted no on remake for some reason, then the ADC showed up and was mad and inted, followed by our top lane AFKing. I openly complained/tiled about these things in chat. I received a chat restriction.

I'm having trouble understanding 3 things:

  1. It obviously could be confirmation bias re: uneven # of games with troll vs against. But it seems to me this should be something the matchmaking system takes into account, i.e. "player luck". Something like, for every game with a player like that on your team, you are given leniency in matchmaking in terms of either MMR, or being matched with honorable players/vs toxic players. Spitballing, but the point is: why isn't that taken into account?

  2. It also seems to me that in games where there is an AFK, the other players on that team should have some amount of immunity/shielding from reports/restrictions, i.e. an understanding that it is an inherently tilting experience and unless they are egregious in spitting venom, the bar for restriction of any kind is higher.

  3. This relates to a broader point. Something akin to what I'm describing in #1 there should be implemented based on individual score, i.e. losing a close game where you played well shouldn't result in as much LP loss as a stomp where you fed (likewise you should gain more LP for carrying/playing well than for getting carried). This is a team game but the ranked system is individual, so such a heavy weight being put solely on game outcome doesn't account for the actual quality of play of individual players.

It just strikes me that there are going to be players on the extreme ends of the luck bell curve because the difference in LP gain/loss is too flat. A more progressive system, i.e. you can lose as little as 5 and gain as much as 30 (again spitballing) based on various factors (game outcome, individual play/score, whether your team had someone else feeding, enemy team had a smurf, etc.) would account better for this and would result in an overall healthier matchmaking system.

11 Comments

rujitra8/15/2019, 7:03:33 PM2 votes
  1. There are 5 enemy players and 4 teammates. Assuming you are not a troll, it is 20% more likely that a troll ends up on the enemy team than on your team.

  2. Two wrongs do not make a right, period.

  3. No. Your lane performance does not matter if you ignore your teammates. Your CS does not matter if you get it by AFKfarming and letting your base fall. There is no objective way to determine someone's "personal performance" in the game.

Makarakarn8/15/2019, 6:24:37 PM1 votes

As far as AFK goes, that's just what it is. However a lot of players (specifically, most of them), seem to confuse genuinely bad games/bad matchups and inting/trolling.

Dukues8/15/2019, 7:40:17 PM1 votes

Yea idk man my experience earlier in the season from like gold 4 - silver whatever is basically ... just a headache idk shits fucked. Idk how it could get this bad it's legit like..... every game someone has their worst game of their life lmfao.

Once I got to gold 3 mmr my games were actually "fun" and a majority of them were competitive and evenly balanced. I have heard horror stories of Plat as I just reached it yesterday. Unsure how it will be haven't played yet but.... up to this point from like gold 3 into plat 4 it was an actual ranked feeling experience to me.