First! Also, I like seeing confirmation that efforts are producing positive results. It's always heartening to see developers take their games seriously, and attempt to remediate issues. Doubly so when a game is basically 'dead' or there is otherwise incentive to stop working on it. Hopefully this means an infusion of new life, because with all its flaws, this is one of the best Mechwarrior games I've played.
@@harackmw I just read through the notes. Nearly all of them are good changes. It's interesting to see their note about the current meta because it's pretty much the same as when I stopped playing. The new values should bring life back to the game.
I just started playing in early june so it was interesting to see these changes. When I first started, it seemed like my PSR would steadily go up with wins no matter what. When they changed the tier system, I was still quite a noob. I started in tier 3, fell down to about halfway through tier 4 so in that regard I think the system is working. Then I finally began to understand the game after about 300+ matches and started to rise out of tier 4. Eventually, I got to a point where I kept just reaching tier 3, but I couldn't stay there. A few losses in a row in quickplay games would knock me back down to tier 4, but one game at tier 4 and I'd get 3 kills and be back to tier 3. Finally got good enough to gain consistently through tier 3 and I'm up to halfway. When I try out a new mech or a build I'm not familiar with, my rating tends to go down so it seems like it's working as intended. It's not without flaws, however! #1 problem is that it's putting cadets in tier 3 to start. They usually just get out of position and get toasted. The matchmaker has no concept of balance regarding cadets. Often, I've been on a team with 2 cadets while the other team has zero. This usually equals a loss, especially when the cadets account for one of our assaults. They're probably going to get wrecked anyway, but better to start them in tier 4 where they won't get CT crit in 2 volleys. #2 problem is that the score is still calculated poorly. My best mech right now is the Black Lanner. I have a positive win ratio and over a 2:1 K/D in that mech. The problem is, it's a low scoring mech. My build is for a highly mobile alpha striker. I typically do a 60 point alpha, often to the enemy's rear armor, then I have to retreat to cool off and not die. Most of it is pinpoint damage on CTs and rear armor. I'd say my average match result for that mech is 2 or 3 killshots and about 300/350 damage. Seems low, but it's important damage and often I take down an assault by myself. By the same token, some jerk just shooting off pointless LRMs on the right map can easily get 700+ damage and not really contribute to the team. The mech also lacks AMS which I think is another big score booster. I doubt it will happen, but they should give torso damage a higher score value, maybe even adding more for rear shots. #3 problem is that a lot of people haven't played enough games to fall out of tier 3 yet. This luckily will eventually fix itself but right now I would say there's still a LOT of players in tier 3 that do not belong there. I've looked up average match scores and K/D ratio and they're way down in both but haven't dropped out yet. #4 problem is that the way it weighs losses isn't really fair. Sure, it counts everyones damage, even in a loss. But it's impossible to get high damage in short games for some builds. If some of your teammates get rolled in the first few minutes, it doesn't matter how great a pilot you are when it's 8 vs 1. It's also almost impossible to get many assists when you get rolled. A lot of times, I'll see my stats on a 12-2 stopfest where I got 1 kill, 1 assist, about 200 damage and the team dies in 5 minutes. My rating still goes down usually on losses. Bottom line- Is it better than it was? Yes! Does it work like it's supposed to? I guess so because as a noob I fell out of tier 3. It could be improved by moving cadets to tier 4 or 5 and counting them differently.
Thanks for this very thorough response. You're absolutely right that they need to start new players out in T4 or simply move the player up or down quickly in their first few games.
@@BlackhawkSC Thanks for all the thoughtful videos man! I know you've taken a break but I'm new to MWO but a huge tabletop / Mechwarrior series fan. I realize that I'm super late to the game but I didn't have a PC that could run it for many years. I find it really interesting to see certain builds of mechs in action as compared to how they play on the tabletop. Also, to play a proper 12 vs 12 engagement in tabletop takes a good 8 hours lol! Compared to tabletop, ECM seems really overpowered and in the board game, you cannot pinpoint aim. But otherwise, it seems like a pretty faithful reproduction and it's made me rethink stuff like mech side profiles, hard points, etc. Keep up the good work!
Thanks, yes I'm well. I logged on to one of my accounts the other day. Kinda waiting to see if the weapon changes are going to do anything interesting.
Good video appreciate the coverage on a topic a lot of players are talking about. I have heard more and more negative talk about the new PSR system which is better than the previous system. I think the issue is the matchmaker and a lot of people don’t have the patience for people to settle in their proper Tiers because that is going to take more time.
It’s better because tiers now represent corresponding skill, what still needs work is how the teams are balanced. They should do that through season average match score or something similar to decrease the stomps rate.
Late to the game ;) but, yes, much better than the old system. Playing again after almost three years, I appreciate pgi's belated effort. I also appreciate any content you put up. You are consistently the best mwo youtube analyst/critic.
Couple points: 1) Nightbird has posted some pretty good distribution of WLR graphs on the PGI forums. We're really no better off than we were before the PSR reset. Admittedly he has an agenda in showing this as his WLR based ranking system should have made for much closer matches. Additionally stomp rates seem about the same or slightly worse than they were after the queue merge (which made things worse in solo queue from a stomp standpoint). 2) This is due in big part to the fact that there are so few people in Tier 1 (or 5 even) that you cannot maintain a 1 Tier separation and get decent matchmaking times. 3) The system still cannot account for group quality in making matches. And balancing mech classes and/or tonnage seems completely impossible in many matches. The PSR rework was definitely needed. JayZ's system wasn't the best, but with regular resets it can work. However, PGI's artificial limitations on what they were willing to change, and the vast disparity between number of players in each tier makes it tough to effect a major improvement. I really don't see it getting much better unless they start setting the Tier boundaries dynamically, but even then groups still throw a monkey wrench into the works. Player numbers were trending down for July until PGI announced the very grindy event in the 2nd half of the month which boosted participation. It looks like they will probably continue down now that the event is over. I don't know how many times PGI can go to the well with events to boost player numbers before people burn out. They certainly haven't convinced me to start playing again.
In general Matchmaking has two problems, how to do the distribution and the second is population size MWO has essentially a population size problem, In that off peak means that the queues are going to be skewed. If you only have 24 players it matter not what the distribution shape looks like you are just not going to get a very good match. In low population periods I find there are massive stomps because essentially you could have a tier 5 matched up with tier 1 because of population size in higher population periods I see less stomps. I have even noticed it in my match score So whilst I think the new PSR is much better (and I suspect it is too late to save he game) the PSR hid what was one of the ain problems that PGI has done very little to add to the game. I stopped playing in 2017 and only came back for the PSR2.0 there is one new map in all that time no new game modes no new weapons not a great welcome back in that sense. Just to be clear I am not sure how many players actually play the game but looking at the leader boards at most there are 15K player that play at least 10 games a season, there are not more than 1500 player that play on average 10 games a day or more which means the majority of people play for less than a couple of hours a night which creates a very narrow window of a full populations You need to that problem or else you are trying to create a decent distribution curve with too few players. I joked there are more paedophiles in the UK than there are MWO players in the world. You need a popultions of nearly double this for any system to work well all the time or else you will basically have two different sorts of games a set of low population games where the number of stomps go up and the then the high population games where the PSR works as it is supposed to do remember the PSR is supposed to pit like skilled players against each other and that only happesn when there is enough like skilled players in the queue
I haven't seen what Nightbird posted, but before July 15, there essentially was no matchmaker. Everyone was Tier 3 for the most part and there's no 1-tier separation. I noticed that after the 1-tier separation, the players I faced in general were better. That said groups are definitely messing with the MM, but I'm not sure how to address this with such a low player population.
@@BlackhawkSC Essentially the number of people falling into each WLR bracket (centered around 1.0) is similar to what it was a couple months ago. Reducing the Tier separation to 1 resulted in Tier 1 players facing long wait times until the gates opened because a bell type Tier distribution with fixed breakpoints really doesn't work with low population. The bigger problem really is that there is no class or tonnage matching taking place in most games. If a group comes light and/or they are low skill, they have as big an impact as a max tonnage elite group, only in the opposite direction. They need to go back to solo only if they're not going to change how groups are matched. It's the only way.
"Essentially the number of people falling into each WLR bracket (centered around 1.0) is similar to what it was a couple months ago" -- Re-distributing W/L curve isn't the goal of the new PSR/MM on aggregate. The difference would be who fell in which bracket, not the number of people in each bracket. If a player was always losing before due to bad MM, he may win more, and a player who was always winning before may lose more. On net, there should be no change.
@@BlackhawkSC Actually that's not correct. An ideal matchmaker would be matching people of near equal skill as often as possible, which should result in WLR for the population being grouped much tighter around 1.0 than it has been (or is now). If you don't narrow the overall WLR distribution then you haven't improved matchmaking one bit. For example, if you and Sean Lang are both playing at the same time, you should very frequently (if not always) end up on opposing teams because you skill level is very close (according to stats). Assuming the rest of your teams are similarly matched skill wise, this would drive both of you down from a WLR of 1.5-2 closer to 1.0 because you should win nearly equally if the MM has done its job.
The PSR works as intended. No issues there. The things that need to be fixed are spawn points and equalizing tonnage across the two teams. Both these defects can be negated with good team comms to an extent, but the lighter team has an uphill battle due to the extra amour of the extra assaults on the enemy team. There's no extra credit for making a kill which is unfortunate. If you are in a lighter mech and kill an assault, but are taken out early, your PSR goes down versus someone who does a lot of damage but makes no kills. I'm a bit on the fence about this aspect of the PSR because the incentive is for cautious play and I've seen assaults hang back rather than be the tip of the spear. This causes high causalities against the lighter mechs who run up against entrenched enemies. Also, the emphasis on damage means certain weapons are favored over others such as ATMs which can put out tremendous damage at short range.
You're correctly pointing out several issues with MM and match score. Right now the series of inputs are as such: match score -> PSR formula -> matchmaker, and then the output of the matchmaker feeds back in the match score. They only fixed 1 element in that cycle, which is the PSR formula. Match score and matchmaker are not fixed directly as a part of this patch.
Question: Won't this edge out players taking on support roles (eg scout w/ Narc) who typically contribute more than raw damage _but_ who's contributions do not translate into match score?
When you're playing support, you should also be doing damage. There are matchscore bonuses for support activities, so combined with the damage that you do, you should roughly equal out. However I'd argue that the best support mechs are ones that do damage while others tank, so I think the MS formula reflects that reality.
so everyone should start playing assaults to farm more damage? i know some people are proficient on everything but a lbx fafnir can make anyone a farmer. how does this play out for lighter mech classes and PSR level
Like everyone I was reset to T3, but have been getting towards the top of T3. Not wanting to venture into T2 (fearing longer matchmaker times) I have started playing non meta meduim/heavy mechs and lights. This seems to have stopped my upward creep.
The next step would be to lower the value of raw damage in match score and increase the value of kills and component destruction. Also lights get buffs to match score through activities that are performed by light my experience has been actually only a slight difference of like 20 less average match score in lights
I wish the tables shown to us showed how big tier 3 is as that can be quite important when trying to figure out if it's "working". To try to illistrate, imagine if results were always like they were in day 1. This'd mean 100% of people would be in tier 3, an obvious downgrade from the older system. If a large majority of people are staying in tier 3 and we are only seeing a small fraction of players reaching the other tiers then it similarly wouldn't be much of an improvement. We'll have to wait and see if we can get the whole picture once more time has passed and the tiers have a greater chance of spreading out. As with all these bar graphs, we don't know how players are distributed among those brakets, which does matter since the matchmaker does use PSR to try to fine tune balance. A crucial factor to still consider is matchmaking time. If only a very slim number of players reach tier 1 or tier 5 then they might be subject to rather lengthy search times, something you don't necisarily want to subject your very best (and worst) players to in a dwinding game...
For the distribution to start happening, people need to start playing :P But what you're really pointing out is that even if all active players start playing, it takes a long time to get out of Tier 3 (on the way down or up). I'd imagine a player who should be in Tier 2 takes ~50 matches to get to Tier 2 and some of them don't have the time to play that many games.
@@BlackhawkSC Seeing the size of Tier 3 would help give us an idea of what proportion of players have actually moved out of their tiers. Seeing the distribution of players within the tiers will help us see the distribution more fully, on a local and global level. Moving out to the less populated tiers also comes with the drawback of increased search times, it's important to know how big a drawback this is.
Yes, I like the changes to the lasers overall. Some of the clan laser vomit builds were improved quite a bit. I want the cACs to be better, but they're still inferior to cUACs. cUAC20 got a bit better. The HML + cUAC20 combo got better as a result. You can use cACs on some niche builds though and they wouldn't be completely stupid now.
They are not taking in to account on games like conquest where a fast mechs win the games buy capturing bases but does not do a lot of fighting. If this is going to stay the ways it is WHY should be play conquest. They may as well take conquest out and put incursion back in.
as a tier 2 player, i have had a pretty horrible experience with it. the matchmaker always puts me in tier 1 games, and i take down arrows to psr because its telling me i underperformed in those games. well duh. if you put me in games with doobix and scout derek and backshot and rkshz and a million people i recognize from comp, toc, etc, i'm going to be worse than them. i dont need psr to tell me that. complaining aside, this is the other side of the issue you discussed in the video, namely, each match being its own PSR evaluation. If the matchmaker does not correctly select games for an individual, then the result of the self-contained PSR evaluation will also be incorrect. if i could opt out of tier 1 games and only play against other tier 2 players, or even, if i could just get a balance of tier 1 and tier 3, i think the system would be working as intended. but that's not what's happening. people keep saying it'll get better, but its not. two days ago i played 6 games, all 6 were tier 1. until this particular problem is fixed, its a horrible system for me, no matter how well it works for people who are less sensitive to piss poor matchmaking.
There's a 1 tier separation. So Tier 2 should play against Tier 3 and Tier 1. Now if you are truly Tier 2, you should score higher than Tier 3 and lower than Tier 1, and you would pretty much stay around where you are. If you reached Tier 2 prior to July 15th, the results are still early. The longer the current PSR system stays in place, the more accurate it is.
@@BlackhawkSC agree. however i always play tier 1. that's the problem. to be more clear, due to the 1 tier separation rule, the games i play in are either explicitly tier 1+2 or tier 2+3. tier 1 cannot play tier 3, so that eliminates the possibility of both tier 1 players and tier 3 players being in a game with me. due to that, the PSR methodology only makes sense if i play an equal number of tier 1+2 and tier 2+3 games. at the times when i play, this never happens. i only play against tier 1. i'm sure you can understand why i'm not a fan.
I see your point. That might be what's happening. The issue is I think most people are still in Tier 3 and Tier 1 and 2 are only a few players, so it's possible the MM keeps sticking 1s and 2s together whenever they're online and searching.
@@BlackhawkSC yup. tier 3 can self solve, but the MM needs tier 2 to solve for tier 1 wait times. its been a month since reset, and the MM is still doing it. that said, if i were a better player, this wouldn't matter at all, so i keep telling myself to practice and improve so i dont have to care. i figure the better you are, the less it matters to your enjoyment of the game what matches you get put in. i hope so anyway. thanks as always, for everything you do.
It’s too easy to reach tier 1. I don’t know. I’m tier 1 with a full bar and it wasn’t difficult to obtain it. Way less tier 4 and 5s and everyone still stuck in 3.
Please please please tell me they're dropping new players into Tier 5, not Tier 3. I'd rather not have our community die overnight because we scare away literally every new player on their first day.
@@BlackhawkSC So new people on their first day are dropping with Tier 1's. Got it. Their first time playing a video game they're getting stomped by the best of the best. On their first day. Aww, that's too bad. I missed the days when MWO had new players joining MWO. Remember those days? :( Seriously, I only found the game in November - can we start and end every single conversation about MWO with this fact, until someone upstairs realizes they did an oopsie? I don't know what could kill off new players joining faster than that.
Well since July 15 they're trying to do a 1-tier separation, meaning they try to drop Tier 3s with 4s and 2s. However because they relax the MM after 4 minutes, right now sometimes Tier 1 can drop with 3s or 3s can drop with 5s. It's best for new players to drop during North American prime time to have the best chance to get a decent match.
Chiming in on this, I believe new players are not directly scared away, as often had the feeling that the Trial mechs were my most limiting factor starting out some years ago. I only started climbing tiers and objectively becoming better, after starting to look for resources like Blackhawks channel which teach you the fundamentals as well as some more in depth concepts. Even though some of it is outdated, having veterans explain their way of thinking as well as guides on building mechs as well as not being restricted by a trial mech was the breakout point for me. So maybe T3 IS a good place to start, so you can see what a neutrally okay game would look like.
Overall it's a little better than before, I guess. Still, I was back in T1 in no time and still don't deserve to be there with my current skill set. So, the games are much more difficult than before because I face much better players all the time. Everyone under 90% in the Jarls just should no be in T1. tha
Before July 15th there was no matchmaker essentially. Most people were Tier 3 and there was a 2-tier separation, meaning that even when you got to T1 you were still playing T3, who might be newbies or people who belong in T5. After July 15th, you should be mostly be matched with T1 and T2. So if you did not belong in T1 you would either be dropping out or not seeing your progress bar move. Suppose you're staying in T1 for an extended period of time, that means you should be matched with other T1s and T2s and you're doing better or at least matching the average of your teammates/the other team. That by itself affirms that you should be in T1. Jarl's % is good when everyone played against each other, which is no longer the case after the new PSR + 1-tier separation. Let's take this example, if you made it to the Olympics, but you always were 5th place out of 10, does it mean you're just an average athlete? No it means you're the middle of the elite athletes, you're still elite (T1).
Wow Blackhawk still playing MWO? Can't believe anyone still left in this mess of a game. I thought every skilled player had already left for games that are actually ..... Well good. I tried MWO three or four times after I quit 3 years ago. And it is a pain. They have done everything to ruin the game and they did. In my point of view MWO is unplayable in the current status. I wonder why you still punish yourself
First! Also, I like seeing confirmation that efforts are producing positive results. It's always heartening to see developers take their games seriously, and attempt to remediate issues. Doubly so when a game is basically 'dead' or there is otherwise incentive to stop working on it. Hopefully this means an infusion of new life, because with all its flaws, this is one of the best Mechwarrior games I've played.
Damn bro, miss you. Hope you're well.
TY! I'm OK. Family obligations keep me away from gaming in general these days.
@BlackhawkSC thank you for replying. You're definitely one of my favorite youtubers.❤️
Miss you BlackhawkSC, hope to see you return to the game. =)
Thanks man. I hope I can come back at some point too. Miss shooting mechs :)
@@BlackhawkSC hey man hope you been following the news, huge changes coming!
@@harackmw I just read through the notes. Nearly all of them are good changes. It's interesting to see their note about the current meta because it's pretty much the same as when I stopped playing. The new values should bring life back to the game.
I just started playing in early june so it was interesting to see these changes. When I first started, it seemed like my PSR would steadily go up with wins no matter what. When they changed the tier system, I was still quite a noob. I started in tier 3, fell down to about halfway through tier 4 so in that regard I think the system is working. Then I finally began to understand the game after about 300+ matches and started to rise out of tier 4. Eventually, I got to a point where I kept just reaching tier 3, but I couldn't stay there. A few losses in a row in quickplay games would knock me back down to tier 4, but one game at tier 4 and I'd get 3 kills and be back to tier 3. Finally got good enough to gain consistently through tier 3 and I'm up to halfway. When I try out a new mech or a build I'm not familiar with, my rating tends to go down so it seems like it's working as intended. It's not without flaws, however!
#1 problem is that it's putting cadets in tier 3 to start. They usually just get out of position and get toasted. The matchmaker has no concept of balance regarding cadets. Often, I've been on a team with 2 cadets while the other team has zero. This usually equals a loss, especially when the cadets account for one of our assaults. They're probably going to get wrecked anyway, but better to start them in tier 4 where they won't get CT crit in 2 volleys.
#2 problem is that the score is still calculated poorly. My best mech right now is the Black Lanner. I have a positive win ratio and over a 2:1 K/D in that mech. The problem is, it's a low scoring mech. My build is for a highly mobile alpha striker. I typically do a 60 point alpha, often to the enemy's rear armor, then I have to retreat to cool off and not die. Most of it is pinpoint damage on CTs and rear armor. I'd say my average match result for that mech is 2 or 3 killshots and about 300/350 damage. Seems low, but it's important damage and often I take down an assault by myself. By the same token, some jerk just shooting off pointless LRMs on the right map can easily get 700+ damage and not really contribute to the team. The mech also lacks AMS which I think is another big score booster. I doubt it will happen, but they should give torso damage a higher score value, maybe even adding more for rear shots.
#3 problem is that a lot of people haven't played enough games to fall out of tier 3 yet. This luckily will eventually fix itself but right now I would say there's still a LOT of players in tier 3 that do not belong there. I've looked up average match scores and K/D ratio and they're way down in both but haven't dropped out yet.
#4 problem is that the way it weighs losses isn't really fair. Sure, it counts everyones damage, even in a loss. But it's impossible to get high damage in short games for some builds. If some of your teammates get rolled in the first few minutes, it doesn't matter how great a pilot you are when it's 8 vs 1. It's also almost impossible to get many assists when you get rolled. A lot of times, I'll see my stats on a 12-2 stopfest where I got 1 kill, 1 assist, about 200 damage and the team dies in 5 minutes. My rating still goes down usually on losses.
Bottom line- Is it better than it was? Yes! Does it work like it's supposed to? I guess so because as a noob I fell out of tier 3. It could be improved by moving cadets to tier 4 or 5 and counting them differently.
Thanks for this very thorough response. You're absolutely right that they need to start new players out in T4 or simply move the player up or down quickly in their first few games.
@@BlackhawkSC Thanks for all the thoughtful videos man! I know you've taken a break but I'm new to MWO but a huge tabletop / Mechwarrior series fan. I realize that I'm super late to the game but I didn't have a PC that could run it for many years. I find it really interesting to see certain builds of mechs in action as compared to how they play on the tabletop. Also, to play a proper 12 vs 12 engagement in tabletop takes a good 8 hours lol! Compared to tabletop, ECM seems really overpowered and in the board game, you cannot pinpoint aim. But otherwise, it seems like a pretty faithful reproduction and it's made me rethink stuff like mech side profiles, hard points, etc. Keep up the good work!
I'm glad they are making an effort to provide a better player experience, even if there has been no new content.
I miss your gameplay videos, hope you're well!
Thanks, yes I'm well. I logged on to one of my accounts the other day. Kinda waiting to see if the weapon changes are going to do anything interesting.
@@BlackhawkSC Good to hear!
Let's hope for some positive changes then :-)
Nice. Thanks for breaking it down for us. I certainly appreciate the new PSR more than before. Feels like I actually have to work harder.
Good video appreciate the coverage on a topic a lot of players are talking about. I have heard more and more negative talk about the new PSR system which is better than the previous system. I think the issue is the matchmaker and a lot of people don’t have the patience for people to settle in their proper Tiers because that is going to take more time.
Thanks for explaining this in this Video.
It’s better because tiers now represent corresponding skill, what still needs work is how the teams are balanced. They should do that through season average match score or something similar to decrease the stomps rate.
Hope you're doing alright.
I'm good thank you
@@BlackhawkSC you coming back to mwo?
I saw you back in MWO (was watching someone's stream). Looking forward to seeing you on the battlefield or linking up with ya. Cheers o7 - BrioS
Thanks Brios :D I left you a msg in the game chat just now explaining things
@@BlackhawkSC Yup, read it. Real life > Mechs so take care but it's always a pleasure to have you in game. o7 - BrioS
5:17 Oh, hi! Mind sharing your BAS-B build?
Hehe here you go - th-cam.com/video/Kzx6CU5FRzA/w-d-xo.html
Late to the game ;) but, yes, much better than the old system. Playing again after almost three years, I appreciate pgi's belated effort. I also appreciate any content you put up. You are consistently the best mwo youtube analyst/critic.
Couple points:
1) Nightbird has posted some pretty good distribution of WLR graphs on the PGI forums. We're really no better off than we were before the PSR reset. Admittedly he has an agenda in showing this as his WLR based ranking system should have made for much closer matches. Additionally stomp rates seem about the same or slightly worse than they were after the queue merge (which made things worse in solo queue from a stomp standpoint).
2) This is due in big part to the fact that there are so few people in Tier 1 (or 5 even) that you cannot maintain a 1 Tier separation and get decent matchmaking times.
3) The system still cannot account for group quality in making matches. And balancing mech classes and/or tonnage seems completely impossible in many matches.
The PSR rework was definitely needed. JayZ's system wasn't the best, but with regular resets it can work. However, PGI's artificial limitations on what they were willing to change, and the vast disparity between number of players in each tier makes it tough to effect a major improvement.
I really don't see it getting much better unless they start setting the Tier boundaries dynamically, but even then groups still throw a monkey wrench into the works. Player numbers were trending down for July until PGI announced the very grindy event in the 2nd half of the month which boosted participation. It looks like they will probably continue down now that the event is over. I don't know how many times PGI can go to the well with events to boost player numbers before people burn out. They certainly haven't convinced me to start playing again.
In general Matchmaking has two problems, how to do the distribution and the second is population size
MWO has essentially a population size problem, In that off peak means that the queues are going to be skewed. If you only have 24 players it matter not what the distribution shape looks like you are just not going to get a very good match. In low population periods I find there are massive stomps because essentially you could have a tier 5 matched up with tier 1 because of population size in higher population periods I see less stomps. I have even noticed it in my match score
So whilst I think the new PSR is much better (and I suspect it is too late to save he game) the PSR hid what was one of the ain problems that PGI has done very little to add to the game. I stopped playing in 2017 and only came back for the PSR2.0 there is one new map in all that time no new game modes no new weapons not a great welcome back in that sense.
Just to be clear I am not sure how many players actually play the game but looking at the leader boards at most there are 15K player that play at least 10 games a season, there are not more than 1500 player that play on average 10 games a day or more which means the majority of people play for less than a couple of hours a night which creates a very narrow window of a full populations
You need to that problem or else you are trying to create a decent distribution curve with too few players.
I joked there are more paedophiles in the UK than there are MWO players in the world. You need a popultions of nearly double this for any system to work well all the time or else you will basically have two different sorts of games a set of low population games where the number of stomps go up and the then the high population games where the PSR works as it is supposed to do
remember the PSR is supposed to pit like skilled players against each other and that only happesn when there is enough like skilled players in the queue
I haven't seen what Nightbird posted, but before July 15, there essentially was no matchmaker. Everyone was Tier 3 for the most part and there's no 1-tier separation. I noticed that after the 1-tier separation, the players I faced in general were better.
That said groups are definitely messing with the MM, but I'm not sure how to address this with such a low player population.
@@BlackhawkSC Essentially the number of people falling into each WLR bracket (centered around 1.0) is similar to what it was a couple months ago. Reducing the Tier separation to 1 resulted in Tier 1 players facing long wait times until the gates opened because a bell type Tier distribution with fixed breakpoints really doesn't work with low population.
The bigger problem really is that there is no class or tonnage matching taking place in most games. If a group comes light and/or they are low skill, they have as big an impact as a max tonnage elite group, only in the opposite direction.
They need to go back to solo only if they're not going to change how groups are matched. It's the only way.
"Essentially the number of people falling into each WLR bracket (centered around 1.0) is similar to what it was a couple months ago" -- Re-distributing W/L curve isn't the goal of the new PSR/MM on aggregate. The difference would be who fell in which bracket, not the number of people in each bracket. If a player was always losing before due to bad MM, he may win more, and a player who was always winning before may lose more. On net, there should be no change.
@@BlackhawkSC Actually that's not correct. An ideal matchmaker would be matching people of near equal skill as often as possible, which should result in WLR for the population being grouped much tighter around 1.0 than it has been (or is now). If you don't narrow the overall WLR distribution then you haven't improved matchmaking one bit. For example, if you and Sean Lang are both playing at the same time, you should very frequently (if not always) end up on opposing teams because you skill level is very close (according to stats). Assuming the rest of your teams are similarly matched skill wise, this would drive both of you down from a WLR of 1.5-2 closer to 1.0 because you should win nearly equally if the MM has done its job.
The PSR works as intended. No issues there. The things that need to be fixed are spawn points and equalizing tonnage across the two teams. Both these defects can be negated with good team comms to an extent, but the lighter team has an uphill battle due to the extra amour of the extra assaults on the enemy team. There's no extra credit for making a kill which is unfortunate. If you are in a lighter mech and kill an assault, but are taken out early, your PSR goes down versus someone who does a lot of damage but makes no kills. I'm a bit on the fence about this aspect of the PSR because the incentive is for cautious play and I've seen assaults hang back rather than be the tip of the spear. This causes high causalities against the lighter mechs who run up against entrenched enemies. Also, the emphasis on damage means certain weapons are favored over others such as ATMs which can put out tremendous damage at short range.
You're correctly pointing out several issues with MM and match score. Right now the series of inputs are as such: match score -> PSR formula -> matchmaker, and then the output of the matchmaker feeds back in the match score. They only fixed 1 element in that cycle, which is the PSR formula. Match score and matchmaker are not fixed directly as a part of this patch.
Question:
Won't this edge out players taking on support roles (eg scout w/ Narc) who typically contribute more than raw damage _but_ who's contributions do not translate into match score?
When you're playing support, you should also be doing damage. There are matchscore bonuses for support activities, so combined with the damage that you do, you should roughly equal out. However I'd argue that the best support mechs are ones that do damage while others tank, so I think the MS formula reflects that reality.
so everyone should start playing assaults to farm more damage? i know some people are proficient on everything but a lbx fafnir can make anyone a farmer. how does this play out for lighter mech classes and PSR level
You're correctly pointing out the issue with using match score
Like everyone I was reset to T3, but have been getting towards the top of T3. Not wanting to venture into T2 (fearing longer matchmaker times) I have started playing non meta meduim/heavy mechs and lights. This seems to have stopped my upward creep.
The next step would be to lower the value of raw damage in match score and increase the value of kills and component destruction. Also lights get buffs to match score through activities that are performed by light my experience has been actually only a slight difference of like 20 less average match score in lights
Wil you ever coming back ?
I still play but tough to make new videos with so many things going on IRL, thanks for the views.
I wish the tables shown to us showed how big tier 3 is as that can be quite important when trying to figure out if it's "working". To try to illistrate, imagine if results were always like they were in day 1. This'd mean 100% of people would be in tier 3, an obvious downgrade from the older system. If a large majority of people are staying in tier 3 and we are only seeing a small fraction of players reaching the other tiers then it similarly wouldn't be much of an improvement. We'll have to wait and see if we can get the whole picture once more time has passed and the tiers have a greater chance of spreading out.
As with all these bar graphs, we don't know how players are distributed among those brakets, which does matter since the matchmaker does use PSR to try to fine tune balance.
A crucial factor to still consider is matchmaking time. If only a very slim number of players reach tier 1 or tier 5 then they might be subject to rather lengthy search times, something you don't necisarily want to subject your very best (and worst) players to in a dwinding game...
For the distribution to start happening, people need to start playing :P But what you're really pointing out is that even if all active players start playing, it takes a long time to get out of Tier 3 (on the way down or up). I'd imagine a player who should be in Tier 2 takes ~50 matches to get to Tier 2 and some of them don't have the time to play that many games.
@@BlackhawkSC Seeing the size of Tier 3 would help give us an idea of what proportion of players have actually moved out of their tiers.
Seeing the distribution of players within the tiers will help us see the distribution more fully, on a local and global level.
Moving out to the less populated tiers also comes with the drawback of increased search times, it's important to know how big a drawback this is.
What the, how did you get that high of damage out of those matches?
PSR 2.0 was in its early stages at that time, so it was a bit of farming
I think it is kinda working, maybe a little bit better.
Do you remember TheMarvelousMike?
You giving the new patch a try?
Yes, I like the changes to the lasers overall. Some of the clan laser vomit builds were improved quite a bit. I want the cACs to be better, but they're still inferior to cUACs. cUAC20 got a bit better. The HML + cUAC20 combo got better as a result. You can use cACs on some niche builds though and they wouldn't be completely stupid now.
@@BlackhawkSC I am glad to hear that, yeah I have been loving my games in you HML + cUAC20 build. Still one of my favorites.
They are not taking in to account on games like conquest where a fast mechs win the games buy capturing bases but does not do a lot of fighting. If this is going to stay the ways it is WHY should be play conquest. They may as well take conquest out and put incursion back in.
as a tier 2 player, i have had a pretty horrible experience with it. the matchmaker always puts me in tier 1 games, and i take down arrows to psr because its telling me i underperformed in those games. well duh. if you put me in games with doobix and scout derek and backshot and rkshz and a million people i recognize from comp, toc, etc, i'm going to be worse than them. i dont need psr to tell me that.
complaining aside, this is the other side of the issue you discussed in the video, namely, each match being its own PSR evaluation. If the matchmaker does not correctly select games for an individual, then the result of the self-contained PSR evaluation will also be incorrect. if i could opt out of tier 1 games and only play against other tier 2 players, or even, if i could just get a balance of tier 1 and tier 3, i think the system would be working as intended. but that's not what's happening.
people keep saying it'll get better, but its not. two days ago i played 6 games, all 6 were tier 1. until this particular problem is fixed, its a horrible system for me, no matter how well it works for people who are less sensitive to piss poor matchmaking.
There's a 1 tier separation. So Tier 2 should play against Tier 3 and Tier 1. Now if you are truly Tier 2, you should score higher than Tier 3 and lower than Tier 1, and you would pretty much stay around where you are.
If you reached Tier 2 prior to July 15th, the results are still early. The longer the current PSR system stays in place, the more accurate it is.
@@BlackhawkSC agree. however i always play tier 1. that's the problem.
to be more clear, due to the 1 tier separation rule, the games i play in are either explicitly tier 1+2 or tier 2+3. tier 1 cannot play tier 3, so that eliminates the possibility of both tier 1 players and tier 3 players being in a game with me.
due to that, the PSR methodology only makes sense if i play an equal number of tier 1+2 and tier 2+3 games.
at the times when i play, this never happens. i only play against tier 1.
i'm sure you can understand why i'm not a fan.
I see your point. That might be what's happening. The issue is I think most people are still in Tier 3 and Tier 1 and 2 are only a few players, so it's possible the MM keeps sticking 1s and 2s together whenever they're online and searching.
@@BlackhawkSC yup. tier 3 can self solve, but the MM needs tier 2 to solve for tier 1 wait times. its been a month since reset, and the MM is still doing it. that said, if i were a better player, this wouldn't matter at all, so i keep telling myself to practice and improve so i dont have to care.
i figure the better you are, the less it matters to your enjoyment of the game what matches you get put in. i hope so anyway.
thanks as always, for everything you do.
It’s too easy to reach tier 1. I don’t know. I’m tier 1 with a full bar and it wasn’t difficult to obtain it. Way less tier 4 and 5s and everyone still stuck in 3.
IT STARTED ME OUT MIDDLE OF TIER 3...LOL...!
New player right?
Please please please tell me they're dropping new players into Tier 5, not Tier 3.
I'd rather not have our community die overnight because we scare away literally every new player on their first day.
They're going into Tier 3. They should be in Tier 4 as in the earlier system.
@@BlackhawkSC So new people on their first day are dropping with Tier 1's. Got it.
Their first time playing a video game they're getting stomped by the best of the best. On their first day.
Aww, that's too bad.
I missed the days when MWO had new players joining MWO. Remember those days? :(
Seriously, I only found the game in November - can we start and end every single conversation about MWO with this fact, until someone upstairs realizes they did an oopsie?
I don't know what could kill off new players joining faster than that.
Well since July 15 they're trying to do a 1-tier separation, meaning they try to drop Tier 3s with 4s and 2s. However because they relax the MM after 4 minutes, right now sometimes Tier 1 can drop with 3s or 3s can drop with 5s. It's best for new players to drop during North American prime time to have the best chance to get a decent match.
Chiming in on this, I believe new players are not directly scared away, as often had the feeling that the Trial mechs were my most limiting factor starting out some years ago.
I only started climbing tiers and objectively becoming better, after starting to look for resources like Blackhawks channel which teach you the fundamentals as well as some more in depth concepts.
Even though some of it is outdated, having veterans explain their way of thinking as well as guides on building mechs as well as not being restricted by a trial mech was the breakout point for me. So maybe T3 IS a good place to start, so you can see what a neutrally okay game would look like.
@@mobi777 glad it helped you before
Overall it's a little better than before, I guess. Still, I was back in T1 in no time and still don't deserve to be there with my current skill set. So, the games are much more difficult than before because I face much better players all the time. Everyone under 90% in the Jarls just should no be in T1.
tha
Before July 15th there was no matchmaker essentially. Most people were Tier 3 and there was a 2-tier separation, meaning that even when you got to T1 you were still playing T3, who might be newbies or people who belong in T5. After July 15th, you should be mostly be matched with T1 and T2. So if you did not belong in T1 you would either be dropping out or not seeing your progress bar move.
Suppose you're staying in T1 for an extended period of time, that means you should be matched with other T1s and T2s and you're doing better or at least matching the average of your teammates/the other team. That by itself affirms that you should be in T1.
Jarl's % is good when everyone played against each other, which is no longer the case after the new PSR + 1-tier separation. Let's take this example, if you made it to the Olympics, but you always were 5th place out of 10, does it mean you're just an average athlete? No it means you're the middle of the elite athletes, you're still elite (T1).
Are you dead?
Wow Blackhawk still playing MWO?
Can't believe anyone still left in this mess of a game. I thought every skilled player had already left for games that are actually ..... Well good.
I tried MWO three or four times after I quit 3 years ago. And it is a pain.
They have done everything to ruin the game and they did. In my point of view MWO is unplayable in the current status.
I wonder why you still punish yourself
There's no other game that plays like MWO though :) I'm no longer playing MWO hardcore, but it's interesting to drop once in a while.