28 replies to this topic

[Vlad Ward]

It seems like a good time to think about adding a Placement Match system. Perhaps even regular PSR resets + Seeding/Weighting based on previous rank (Quarterly?).

It takes hundreds of matches to grind back from Cadet T5 to T1 and every inactive player is now in the T5 bucket after the December patch. That's hundreds of matches where returning comp players and veterans are stepping on the heads of Cadets and actual T4/T5 players on their way up - through no 'fault' of their own.

I actually like the overall design of the PSR system and matchmaker. Building it to rely on Match Score lets you tweak that if you want to incentivize/disincentivize different things. The problem is that It doesn't have a mechanism for dealing with pilots who aren't actually new players after a reset.

Thoughts?

[martian]

Vlad Ward, on 28 March 2021 - 08:55 AM, said:

It seems like a good time to think about adding a Placement Match system. Perhaps even regular PSR resets + Seeding/Weighting based on previous rank (Quarterly?).

I would like to say that I do not quite agree.

One reason why I am glad that I am in T1 is that I do not have to deal with players who struggle with WASD (on both sides). I do not like the idea of having to experience it every few weeks.

Vlad Ward, on 28 March 2021 - 08:55 AM, said:

It takes hundreds of matches to grind back from Cadet T5 to T1 and every inactive player is now in the T5 bucket after the December patch. That's hundreds of matches where returning comp players and veterans are stepping on the heads of Cadets and actual T4/T5 players on their way up - through no 'fault' of their own.
...
The problem is that It doesn't have a mechanism for dealing with pilots who aren't actually new players after a reset.

If somebody really is a T1 material, he will move out of T5 to T3 in about a week or two and rise to T1 in another two weeks.

Vlad Ward, on 28 March 2021 - 08:55 AM, said:

I actually like the overall design of the PSR system and matchmaker. Building it to rely on Match Score lets you tweak that if you want to incentivize/disincentivize different things.

The new PSR system is not ideal, but it better than those system that they were before it.

[Vlad Ward]

martian, on 28 March 2021 - 09:24 AM, said:

I would like to say that I do not quite agree.

One reason why I am glad that I am in T1 is that I do not have to deal with players who struggle with WASD (on both sides). I do not like the idea of having to experience it every few weeks.

That's why you have seeding and placement matches - so you don't have to do that.

martian, on 28 March 2021 - 09:24 AM, said:

If somebody really is a T1 material, he will move out of T5 to T3 in about a week or two and rise to T1 in another two weeks.

PSR movement isn't measured in time, it's measured in number of games played. ~300 games isn't massive in the grand scheme of things but it's 300 games ruined for low tier players and 300 games of boring target practice for returning players.

That's an easy way to kill the experience for everyone.

Edited by Vlad Ward, 28 March 2021 - 09:30 AM.

[martian]

Vlad Ward, on 28 March 2021 - 09:27 AM, said:

That's why you have seeding and placement matches - so you don't have to do that.

This is PGI we are talking about ...

Haven't you noticed the mess with the latest Patch changes?

Vlad Ward, on 28 March 2021 - 09:27 AM, said:

PSR movement isn't measured in time, it's measured in number of games played. ~300 games isn't massive in the grand scheme of things but it's 300 games ruined for low tier players and 300 games of boring target practice for returning players.

PSR movement is not measurred in games - it is measured in PSR movement. Somebody can play two dozens of games and up with 0 movement up or down, while someone else can play two dozens of games in a day and move ... dunno ... a quarter or half a Tier.

[D A T A]

hundreds of games? what? it takes max 50-60 games.

https://youtu.be/IukKlhxL-70?t=8

[Nightbird]

Isn't it better to thank them for coming back by letting them farm pugs for a hundred games?

[Vlad Ward]

martian, on 28 March 2021 - 09:33 AM, said:

This is PGI we are talking about ...

Haven't you noticed the mess with the latest Patch changes?

The patch they're reverting?

martian, on 28 March 2021 - 09:33 AM, said:

PSR movement is not measurred in games - it is measured in PSR movement. Somebody can play two dozens of games and up with 0 movement up or down, while someone else can play two dozens of games in a day and move ... dunno ... a quarter or half a Tier.

Yeah, sorry man but if you're talking about two dozen games a day you're in a totally different headspace.

There's a practical upper bound on PSR movement per game, and most people aren't going to hit it every single game no matter how good they are. 200-300 games is a reasonable pace for T1-level players hit by the reset, particularly after they increased the PSR range to 5000.

D A T A, on 28 March 2021 - 09:38 AM, said:

hundreds of games? what? it takes max 50-60 games.

https://youtu.be/IukKlhxL-70?t=8

Great video. Feels like you're waiting for me to say "Not everyone's as good as you, DATA!" While that's an accurate statement, being, what, the #1 player worldwide right now? It anchors expectations around an unrealistic ideal and turns the conversation away from useful discourse and into a pissing contest over who "really deserves" to be T1.

I think you know as well as I do that they vast majority of T1 is still pretty bad. I think you also know that no matter how bad T1 is, they still absolutely crap all over T2, T3, T4, and T5 - which is the entirety of the problem. It's still boring for a more typical capped-bar T1 to play against T2/3/4/5, and it's still unfun for T2/3/4/5 to get smashed by a more typical capped-bar T1.

Nightbird, on 28 March 2021 - 09:43 AM, said:

Isn't it better to thank them for coming back by letting them farm pugs for a hundred games?

Lol

Edited by Vlad Ward, 28 March 2021 - 09:54 AM.

[East Indy]

Group and weight imbalances are elephants in the room and until those are addressed, PSR is a passenger.

In my experience the PSR change worked. I used to see sub-20th percentile players all the time and now it's only if they're part of a group. When groups are evenly matched or less influential than the sum of solos, all-Tier-1 matches feel very well-placed.

Now, regarding placement, I think PGI missed an opportunity to use matchscore rankings. Tossing everyone into Tier 3 allowed me to jet to Tier 1 before valves were tightened; now, whenever I play non-group-thrown Tier 1 matches I'm absolutely pressed to gain or maintain PSR. But again, when a significant number of matches are decided by groups before the start, it's not of as much use.

Edited by East Indy, 28 March 2021 - 10:11 AM.

[Bud Crue]

Vlad Ward, on 28 March 2021 - 08:55 AM, said:

That's hundreds of matches where returning comp players and veterans are stepping on the heads of Cadets and actual T4/T5 players on their way up - through no 'fault' of their own.

I don't see how a periodic PSR reset has a positive impact on either the virgin T5 player or the returning veteran in such a schema.

When a new player comes to the game they either figure out how to play it or they don't. If a new player comes in and figures out the game mechanics, they will readily climb the tiers, and getting "stepped on" by returning veterans is not any more of an issue for them at T5 as it is for us at T1 when actual good players are dropping. If they don't figure it out, they leave, usually after a few games. Do returning veterans really have some sort of dramatic impact on that theoretical new player population, whereby via their farming ways, they are driving the incompetent new players out faster than normal or preventing the competent new players from climbing the tiers? I don't see how that would work as the returning veterans are just as likely to be grouped with a new player as they are to be grouped against that same new player.

Mediocre new players will rise in the tiers with or without veteran players impacting their matches. For obvious reasons, long term T5 players aren't going to notice their games being impacted by returning "veteran" players. As such I don't see how a periodic PSR reset would affect this situation.

[Vlad Ward]

Bud Crue, on 28 March 2021 - 10:32 AM, said:

I don't see how a periodic PSR reset has a positive impact on either the virgin T5 player or the returning veteran in such a schema.

When a new player comes to the game they either figure out how to play it or they don't. If a new player comes in and figures out the game mechanics, they will readily climb the tiers, and getting "stepped on" by returning veterans is not any more of an issue for them at T5 as it is for us at T1 when actual good players are dropping. If they don't figure it out, they leave, usually after a few games. Do returning veterans really have some sort of dramatic impact on that theoretical new player population, whereby via their farming ways, they are driving the incompetent new players out faster than normal or preventing the competent new players from climbing the tiers? I don't see how that would work as the returning veterans are just as likely to be grouped with a new player as they are to be grouped against that same new player.

Mediocre new players will rise in the tiers with or without veteran players impacting their matches. For obvious reasons, long term T5 players aren't going to notice their games being impacted by returning "veteran" players. As such I don't see how a periodic PSR reset would affect this situation.

Plenty of online PvP games regularly reset their version of PSR/MMR for a multitude of reasons. For instance, it helps prevent people from getting too "stuck" in tiers they aren't suited for due to inactivity.

Zero-sum PSR will always "work itself out" given a large enough sample, but you have to assume that people are going to be willing to grind out a bunch of poorly fitted matches before getting into a place that works. Periodic reset and acceleration makes this process less tedious and helps stave off burnout.

Also, you assume being on the winning side of a stomp where one player is 30-40% of the total team damage is going to have a positive impact on new player retention. That's just silly. All that does is reinforce the idea that MWO matchmaking is stupid when the problem has nothing to do with the matchmaker.

[Bud Crue]

Vlad Ward, on 28 March 2021 - 11:15 AM, said:

Plenty of online PvP games regularly reset their version of PSR/MMR for a multitude of reasons. For instance, it helps prevent people from getting too "stuck" in tiers they aren't suited for due to inactivity.

Zero-sum PSR will always "work itself out" given a large enough sample, but you have to assume that people are going to be willing to grind out a bunch of poorly fitted matches before getting into a place that works. Periodic reset and acceleration makes this process less tedious and helps stave off burnout.

Also, you assume being on the winning side of a stomp where one player is 30-40% of the total team damage is going to have a positive impact on new player retention. That's just silly. All that does is reinforce the idea that MWO matchmaking is stupid when the problem has nothing to do with the matchmaker.

I'm not assuming anything and I mentioned nothing about stomps. I am merely asserting that a PSR reset won't help either the veteran player who "through no fault of their own" is stomping the T5 players, nor the T5 players themselves be they a new player or of the long term T5 crowd. There may be lots of good reasons to do a periodic PSR reset, I just don't think a reset will do anything to alleviate the suffering you describe for either the returning veteran or the folks normally in T5.

[w0qj]

+1 Agree, PSR reset is not needed, as returning players in Tier 5 would rise back to Tier 1 in a relatively short period of time.
In fact, allowing Tier 5 returning veteran players to naturally rise back to Tier 1 may not be a bad thing--it may well teach other players how to play MWO properly!

Let's turn the tables around and look at things from a Tier 4/5 player's perspective (ie: me), diligently leveling each of my unskilled mechs and constantly getting killed. But I digress...

I got killed in a game midway through, and just happened to be spectating in the cockpit of a returning veteran playing a standard variant Stormcrow laser build. He did over 1,600 damage, and almost 1,000 match score, along with 6 kills. He carried our team to victory.

Needless to say, I've learnt a lot from spectating this player's game from the returning veteran player's cockpit view.

Therefore allowing returning veteran players playing Tier 5 and letting them naturally rise back to Tier 1 may not be a bad thing--it may well teach other players how to play MWO properly!

Bud Crue, on 28 March 2021 - 10:32 AM, said:

I don't see how a periodic PSR reset has a positive impact on either the virgin T5 player or the returning veteran in such a schema.

When a new player comes to the game they either figure out how to play it or they don't. If a new player comes in and figures out the game mechanics, they will readily climb the tiers, and getting "stepped on" by returning veterans is not any more of an issue for them at T5 as it is for us at T1 when actual good players are dropping. If they don't figure it out, they leave, usually after a few games. Do returning veterans really have some sort of dramatic impact on that theoretical new player population, whereby via their farming ways, they are driving the incompetent new players out faster than normal or preventing the competent new players from climbing the tiers? I don't see how that would work as the returning veterans are just as likely to be grouped with a new player as they are to be grouped against that same new player.

Mediocre new players will rise in the tiers with or without veteran players impacting their matches. For obvious reasons, long term T5 players aren't going to notice their games being impacted by returning "veteran" players. As such I don't see how a periodic PSR reset would affect this situation.

Edited by w0qj, 28 March 2021 - 12:05 PM.

[RickySpanish]

Vlad Ward, on 28 March 2021 - 09:53 AM, said:

...the #1 player worldwide right now?

Nah Data isn't quite that good, there's a bunch of EMP players ranked higher than him on Jarl's.

[PocketYoda]

D A T A, on 28 March 2021 - 09:38 AM, said:

hundreds of games? what? it takes max 50-60 games.

https://youtu.be/IukKlhxL-70?t=8

Depends on the players.

[martian]

Vlad Ward, on 28 March 2021 - 09:53 AM, said:

The patch they're reverting?

The patch that was so bad that it should have never been implemented in the first place.

Vlad Ward, on 28 March 2021 - 09:53 AM, said:

Yeah, sorry man but if you're talking about two dozen games a day you're in a totally different headspace.

There's a practical upper bound on PSR movement per game, and most people aren't going to hit it every single game no matter how good they are. 200-300 games is a reasonable pace for T1-level players hit by the reset, particularly after they increased the PSR range to 5000.

In July 2020 PGI did exactly what you are suggesting: All MWO players were moved to mid-Tier 3. Very good players needed about one week to leave Tier 3 and to move to Tier 1. Some other players needed two weeks to move to T1.

What you are suggesting is not much different - the only difference is that everybody would start in T5, but other than that it would be pretty much the same thing. Top MWO players would climb to top Tiers in no time, while other good players would climb slightly slower.

[Dozer6]

RickySpanish, on 28 March 2021 - 12:03 PM, said:

Nah Data isn't quite that good, there's a bunch of EMP players ranked higher than him on Jarl's.

Until they play and beat him in comp, and get the badge, he is still top dog...

[Vlad Ward]

martian, on 29 March 2021 - 02:07 AM, said:

In July 2020 PGI did exactly what you are suggesting: All MWO players were moved to mid-Tier 3. Very good players needed about one week to leave Tier 3 and to move to Tier 1. Some other players needed two weeks to move to T1.

What you are suggesting is not much different - the only difference is that everybody would start in T5, but other than that it would be pretty much the same thing. Top MWO players would climb to top Tiers in no time, while other good players would climb slightly slower.

I mean, no? I feel like we're not understanding each other. The crux of what I'm proposing are placement matches and seeding. Periodic resets are simply there to allow people to take advantage of placement matches after their cadet period ends.

The climbing will still happen in the current system, but it won't be fun for anyone. DATA posted a good example of a match that would drive new players nuts earlier. Good players will either climb via a smaller number of hugely disproportionate matches or a larger number of less egregious stomps. Either way, it's a pretty big impact on the low-mid tier population that could be avoided by common sense solutions implemented in most other ranked PvP games.

[Bud Crue]

Vlad Ward, on 29 March 2021 - 07:17 AM, said:

The crux of what I'm proposing are placement matches and seeding. Periodic resets are simply there to allow people to take advantage of placement matches after their cadet period ends.

So it sounds like you are proposing that after the cadet matches are played, some predetermined set of matches (5? 10?) have a disproportionate impact on the player's PSR bar (for good or ill) that will act as the "seeding" or initial placement matches so as to accelerate the player's presumed rise in the tiers. The basic idea being that good players will score well during this seeding phase relative to the average player, and thus get catapulted out of the lower tiers quicker than they might otherwise rise (e.g. 5-10 games to throw them up to T2-T1, vs say 50--60 games under the current system).

If I have this correct, couldn't such a system be readily abused by the folks who want to just club T5 seals (such as by intentionally playing awful builds during the seeding phase or even throwing those matches). Once the seeding phase is past, they are free to enjoy their 50-60 matches worth of normal lower level play (clubbing), or is there some mechanism that would prevent that? Also, it seems to me that a periodic PSR reset with this system would allow such players to cause even more havoc, since they would regularly be put back down to T5 where they can once again restart the clubbing.

[GARION26]

I don't see any real advantage of reseeding people regularly. Yes someone who has been out of the game may have skill atrophy but that's going to be corrected for with drops in PSR gradually based on how often they play.

The data is calculable on how long it takes to move up- max movement is 24 points per game. Someone who is a true 1% player who is somehow in tier 5 is probably going to consistently get 24 point upticks. 5 games for 96 point game in PSR, 50 games for 960 point gain. The higher they go of course the slower they would gain as they are less likely to get the 24 point increase games and in some cases as they face better opposition they will have games they go down. At some point they'll hit their cap either based on their PSR up and down games balancing out or because they hit the true PSR cap of 5000.

By definition 99% of us aren't as good as a true 1% player - we're going to go up slower on average even starting in tier 5 and we'll hit our equilibrium point at a much lower PSR value. But you can pick whatever number you want to use for average PSR gain per match from a tier 5 start to calculate how long until they hit that players equilibrium point.

[Vlad Ward]

Bud Crue, on 29 March 2021 - 08:34 AM, said:

So it sounds like you are proposing that after the cadet matches are played, some predetermined set of matches (5? 10?) have a disproportionate impact on the player's PSR bar (for good or ill) that will act as the "seeding" or initial placement matches so as to accelerate the player's presumed rise in the tiers. The basic idea being that good players will score well during this seeding phase relative to the average player, and thus get catapulted out of the lower tiers quicker than they might otherwise rise (e.g. 5-10 games to throw them up to T2-T1, vs say 50--60 games under the current system).

If I have this correct, couldn't such a system be readily abused by the folks who want to just club T5 seals (such as by intentionally playing awful builds during the seeding phase or even throwing those matches). Once the seeding phase is past, they are free to enjoy their 50-60 matches worth of normal lower level play (clubbing), or is there some mechanism that would prevent that? Also, it seems to me that a periodic PSR reset with this system would allow such players to cause even more havoc, since they would regularly be put back down to T5 where they can once again restart the clubbing.

There is no solving for bad faith. People in the current system can already just throw a few dozen matches to get back to whatever tier they want to play in.

Seeding would help prevent someone who played in Tier 1 last season from ever dropping all the way down to Tier 5, but once again if people want to game the system there's no way for an automated system to differentiate on a large scale between a player who is actually not good at the game and a good player who is playing poorly on purpose.