phroxz0n

We're as upset about it as y'all are and we've been talking about these types of cases internally.

While we've fixed erroneous seeding of people who played normal games moving forward, we can't retroactively fix the cases of people who played their first games a month ago before we pushed the fix. (we looked into it, but it can't be done at the scale or speed that we need it to 😭😭)

The second set of cases look like bots 😭. If a player plays just aram or bot games they don't go through our seeding pipeline. This is also not something we're happy with and are looking at improving, but for these cases in particular, hoping that vanguard removes them from the ecosystem entirely.

Again, apologize for the experience, we're working on things and split 2 reset should hopefully put the erroneously placed players back in their right places.

yeah :( You seem to have played your first game before we pushed the fix to the incorrect seeding issue. Really sorry, but the only way to go about this is to play your way down

yes, but it's one of those roles that's really hard to estimate

It's funny cause I was literally talking with someone the other day about how it's hard to estimate performance with 0-11 blitzcrank who could be having the game of their life. And then this thread came up 🙃

OK so I looked at your opgg, I think you're heavily underestimating your skill level!! You're not an iron player, you're doing pretty well in Gold/Plat games!!

That said, Gold/Plat in normals is probably not Gold/Plat in ranked, so we'd already been thinking about adjustments to move people even further down if we see that.

And it's not particularly effective. Bad actors use their grandma's id, neighbors, etc.

There's some accounts we intentionally want to put there though, like smurfs, etc. But agreed with the general sentiment. We are on a continuous battle to try and do a better job with seeding

Ty

Thanks for the report on this. Will do some triage... Can you dm me the acc name of your friend? You should have been placed similar to them if things are working properly...

It's something we're looking into. We've pushed some adjustments recently and continuing to do more.

There are clear problems with smurfing that we want to resolve (disruptive behavior, avoiding punishments)

Solutions to these problems are difficult and take time. We agree they're problems though!

We also need to find ways to naturally fix some of the reasons why people avoid playing on their main in the first place, eg. Playing ranked on main is not rewarding enough, playing with friends of varying skill is hard, playing off role, flex queue is perceived as less prestigious than solo/duo etc.

Not really. People get inflated because of negative lp gains

Negative lp gains = inflation, ie. Your rank is diamond 4, but you're actually emerald 1 skill = you're inflated and getting +18/-22

Yes nothing to share tho yet

Negative gains aren't expected in masters by default. There was more inflation last year because so many people had negative lp gains.

Lolalytics has positive winrates because it's comparing people on tier boundaries eg. Masters 0lp as masters winrate while they are playing in a diamond 1/2 game, same for d4 in an emerald game.

Those players are likely to have a higher winrate and there are more of them compared to people at the top of the bracket (more d4 players than d1 players)

There's still more work to be done. In emerald and above its mostly eliminated.

https://twitter.com/RiotPhroxzon/status/1760573106698244306?t=BmWSeV6M6caxEyqg9Oz3GA&s=19

More context here on why some of the cases happen.

Most of the cases now are localized to some of the fallout from: * seeding problems (which we've resolved future facing, but couldn't fix people who placed gold but had actual skill which is lower)

• people tanking on boundaries

• people on a precipitous drop

• some bugs in iron where iron 3 and 4 players were placing in iron 2

There are Champs that spend a long time being strong, but aren't strictly overpowered. Eg. A miss fortune who sits at 52% winrate for a long time. There's a bunch of Champs that could sit at 52% without being OP so they should probably take turns so the meta doesn't get stale. Champ variety is alot of what makes league fresh

Yeah I was coming in to reply this. It's not happening to everyone, but it is happening to some people. Really sorry about this 😭

We're looking at bringing item power down for Procs specifically starting at 14.3. 14.2 was a massive swing in terms of durability for classes and we're going to continue looking for ways we can make playing squishes feel better to play

It might help to get some coaching. If you don't know what you're doing wrong and you're plat you're probably doing hundreds of things wrong per game.

can you send me some examples of this, I'm curious to dig in

This is just the start. We're still going to continue working on tooltips to strike the balance between new and returning players and veteran players.