Matchmaking and Threat

[M4Ni4C 18]

I want to point out another flaw in the current system. You start a mission without opposition and then you teammates quit or get disconnected. If you get opposition now you most likely end in a 1 vs 1 mission. If you are unlucky neither you nor your opponent can call backup.

Is it possible to limit missions down to 2 vs 2 players? I know there are players who probably like such missions but I think this shouldn’t happen in a team based pvp game.

[HighSociety 148]

4 hours ago, Clandestine said:

Nothing will stop dethreating. Even if it would take 100 missions playing bad some people would do it.

Maybe right but also keep in mind there are some people who just dethreat to play together with friends who are new to APB 

[Lord Cashpoint 269]

Thanks for the post MattScott. Something you didn't quite touch on although I'm sure you've thought about is how the game calculates threat. By which I mean - how does the game determine how well you did in a mission?

 

Very early on in APB: Reloaded threat was determined by winning a mission - this had a huge range of issues, such as losing threat for being called in as backup at the last second or losing large amounts of threat for being one kill down in a deathmatch when the timer ran out. A scoring system was introduced which supposedly would help better determine who had done the best during missions.

 

This has definitely helped but it also has a few problems as well, namely how score is calculated. Medals often award disproportionate amounts of points for what are sometimes trivial actions (You get more points for killing someone with a grenade after you have died than if you were still alive when the grenade went off for example). Attackers can sometimes rack up to around 5 kills worth of points before the defenders have even had a chance to arrive. Running someone over in a car is apparently worth double the amount of points that just shooting them. The list goes on.

 

Additionally - the game is probably too quick to up-threat people. I've seen missions where a silver has progressed to gold simply for being the only member of the enemy team to get a kill.

 

Finally threat will always be difficult to determine when missions in APB can be so frustratingly random and unfair. I'm not expecting Little Orbit to undertake a revamp of the mission system right away (this is a huge task), but it doesn't seem fair to down threat a player because his mission asked him to drop a heavy item in the dock area of Waterfront. Or because they had to complete 3 objectives within 5 minutes (China Blues). The new score system was meant to take the sting out of losing - but it still awards points to the victorious players and often if you have a terrible attack objective you are going to die a lot more. Fixing missions will require not only looking at the objectives and timers in missions - but also the map design itself (Looking at you, both fortresses which border the skatepark in financial).

[TheMessiah 430]

9 hours ago, MattScott said:

3) What are the problems I see in APB's current matchmaking?

 

9 hours ago, MattScott said:

- Exposing the threat levels and giving players a choice on which districts they enter only breeds a way for players to undermine the necessary balance in districts.

Why dont u hide players and districts threats right now then?Also "show recommended only" tab if stay checked and hidden people will choose only players around theyr lvls.Only rank will be visible(blue and red)Current threat system needs some tweaks dou-too easy to get gold.Also if u add free for all districts where matchmaking is off and all players can group however they want-i dont see the problem.Also for bigger choice pool u can make inner faction matchmaking.Theres no good guys in this game anyway.That can be temporally solution only till more players return

[Jazeker 1082]

This is the most interesting and promising APB post I've read in years. I am serious, not kidding. I mean, anyone serious about more activity and more players knows this has always been the way to go. "Can't be done" has always been the answer. Here we have it can be done, after an upgrade. I'll always love this game but from the very start this setup has been holding the game back. I'll even go as far as saying that the system accommodated unwanted behavior within our community at very important moments in the past. I don't know what to say. Can't believe I'm reading this. APB 2019?

[Attenzione 1]

I think more important than matchmaking is the mission design.
It is completly unbalanced.
There are spots that are too easy to defend because of the spawns or what ever.
The other problem is the reward system, there is almost no difference in winning 5 stages and the last one, or only the last one.
Also when you win 5 stages but lose the last one you get not enough exp/money for it.
Then there is also a big gap between mission if you count the exp/money per minute.
Losing missions fast, give more exp/money than winning missions that take 20min+ because of the long stages

[SK4LP 62]

not really against this phasing option, but agree with others with the fun side we may loose ..

 

with new Unreal, how much player could we have on a map ? could we extend map ? ( join Financial and Water ? )

new contact and missions for clan only ( so we create a top lvl above threats ) ?!

[nndrei 129]

For me it's a great idea, from what I could understand. But I'm just asking if there will be loading screens?

[iTzBlackout 139]

2 hours ago, SK4LP said:

could we extend map ? ( join Financial and Water ? )

As I read they will focus on financial and waterfront for now. (cuz there was also a topic about Midtown)

And I think it's really hard to combine financial and waterfront because there is a huge area between them.

 

 

20 minutes ago, Roxane said:

For me it's a great idea, from what I could understand. But I'm just asking if there will be loading screens?

I highly doubt you have to load each time you get into a new match. You only have to download the characters and cars of the new players you get connected with and for that is no loading screen needed.

 

3 hours ago, Attenzione said:

There are spots that are too easy to defend because of the spawns or what ever.

I agree that some spots are too easy to defend. As example spots you can only reach if you use a ladder. I already saw a nice suggestion (cannot remember oldforum/newforum) replacing a lot of ladders with stairs... he also added pictures. Ofc ladders are not the only problem. But I would like to see some changes releated to mission balancing.

[Flaws 1033]

Phasing would be one of the best things to ever happen to APB, especially if the full plan with a single World works out. It would solve many issues, certainly. And I'm certain that new players would have a much more playable experience then so it's bound to attract more of those.

[BIank 109]

17 hours ago, MattScott said:

We eliminate visible threat levels, because it's not information you can act on.

1

While eliminating visible threat levels, I would hate the removal of rank icons.

 

The current solutions you have implemented have either made the icon the same for the faction (as was in open conflict) , or have removed it entirely (ps4 and xbox one builds)

 

I think it would be a good idea to combine the idea of Open Conflict icon colors with the current ranks so they would now be red or blue instead of the current bronze, silver or gold.

(very quick mockup)

 

[Revoluzzer 274]

22 hours ago, MattScott said:

3) What are the problems I see in APB's current matchmaking?

- Due to Unreal 3's notoriously slow loading times, and the fact that "phasing" or switching to another instance of the same district, isn't supported in Unreal 3 - APB was designed to only match one player against another player inside the same district on the same server.

- Working off that limitation, they needed some way to segregate districts based on "threat" or some other skill rating. In theory, with balanced districts, the players are already close in skill to any other player, so that matching can work again.

- Exposing the threat levels and giving players a choice on which districts they enter only breeds a way for players to undermine the necessary balance in districts.

 

Is Threat broken? Yes. Is it frustrating? Yes.

I'd like to discuss some problems (which partly have been mentioned in this thread already) and disagree with the idea that "threat" (i.e. the threat-calculation-system) is broken. I've detailed my thoughts on the matter in the past (of which a particularly lengthy example is also linked in my signature), which might have been more coherent than this late night write up.

 

 

Problem: The threat-scale is fixed, not dynamic.

What does this mean? In APB all active players could theoretically reach Gold threat, by "extracting" threat out of players who eventually stop playing. On a dynamic threat-scale, only a certain amount of players (e.g. 20%) could be Gold. The remaining 80% would always be split amongst Silver (e.g. 40%), Bronze (e.g. 25%) and Green (e.g. 15%). Inactive players would drop out of consideration after a certain time of inactivity.

Now the existing threat-system does have measures to deal with in activity - inactive players gradually lose threat over time. However I had barely been active over the span of almost two years, if I'm not mistaken, and during this time my threat doesn't seem to have degraded drastically enough compared to my skill. In other words, this measure is probably not set to a harsh enough level. I think it should exist in the future.

 

 

Problem: Players can see and somewhat manipulate their threat.

This can of worms is outlined in my linked/mentioned post. I think hiding threat and preventing players from manually selecting districts is crucial to creating a good matchmaking experience. If designed well, phasing players to other districts would only become necessary in very rare circumstances.

 

 

Problem: District threat levels are fixed, not dynamic.

This is obviously similar to my first point. Currently a districts is set to one threat colour, which is a massively inconvenient restriction. After all, players at the very edge of Silver and Gold (formerly known as Silver 10 and Gold 1) are much closer to each other than players at opposite ends of the same colour (e.g. Silver 1 and Silver 10).

Originally APB would create districts with a certain threat range and sort players into appropriate districts automatically, based on their threat level. Evidently this system wasn't without flaw, otherwise it would still be in use today. One flaw from back then was manual district selection (players chose the most populated district over the most appropriate) and fixed district threat. The latter means that a district might have spun up to host players from Threat 6-10 (of the original Threat Level 1 - 15 scale), but was also manually joined by players outside that range once it reached a sufficient population. The district then never adjusted its range accordingly (although eventually it would simply have been 1-15 anyway).

 

In an ideal system districts will dynamically adjust their threat-range as players leave and the player-distribution-system will repopulate it with players of an appropriate threat level, thus ensuring the matchmaking-system doesn't have too much work to do because players will most of the time match up fine.

Should, however, a player (or group) shoot outside a tolerable threat-scale for the district, you could use phasing to transfer them to a more appropriate district.

 

 

Matchmaking mechanics.

Speaking of matchmaking, there are a couple of issues people usually see when it comes to finding opposition.

> For some, it is finding opposition at all. These players will usually resort to waiting for opponents at each stage of a mission. As the matchmaking system will gradually increase the tolerance for a matchup (eventually seeking opponents outside of a sensible threat range) they will either get opposed by much stronger or much weaker players.

> For some, it is finding the same players over and over and over. This is usually caused by two teams (or players) being the only good match across the whole district population. Again, some players try to solve this issue by waiting, however now outside of a mission instead of during one. This does work if the other team gets a bad match by waiting in a mission for long enough.

> For some, it is getting matched with a smaller team of stronger players. This is particularly common for randomly thrown together players of wildly different threat-levels. If they're lucky, they'll face a similar setup. But more commonly they'll be put against a (usually smaller) pre-made group of comrades, who might technically be at the same group-threat-level. Alternatively the thrown-together-bunch waits out their mission stages for opposition and gets put against a pre-made group which is way above their averaged threat-level.

 

Generally the matchmaking-system is fine. It shouldn't need to be complex, because all systems before it take care of the difficult work.

 

 

Bottom line, I think phasing might fix matchmaking-issues at large, but isn't the ideal solution.

 

 

22 hours ago, BXNNXD said:

i think mattscott has said in the past he’d like threat change to be more gradual, which would make dethreating more difficult 

 

i honestly don’t think there’s any way to make dethreating impossible without it also affecting regular players 

The current threat-system already has counter-measures built in. Threat has a volatility-level, which is very high for new players and gets lower the more they play at a certain skill level.

De-threating increases volatility. Thus players who want to de-threat have to put in quite a bit of effort to first raise their volatility, while their threat barely moves. Once their threat starts moving down, however, it will just as easily move up again.

Naturally de-threating only really works as long as you know you are making progress. Hiding any indication of a player's threat level quickly makes it nigh impossible to gauge where you stand. And if all the systems which determine how players are matched up against each other can function as they should, de-threating would hopefully be a much rarer occurrence anyway.

[Bigalbroski 1]

I think phasing districts is a good idea but only if you can't find opposition in the district you're in. I don't think plays should be phased into lower population realms as that takes awhile the feeling on a live sever.(I also don't like that this would erase sprays) For threat I don't think you should be able to go backwords into servers like gold to silver, but the population would have to increase to support this. I think threat needs to take many things into consideration like Rank, Weapons owned, Vehicles owned, Activity, Time played, etc.

Also what is the reason for not reverting conflict districts back to 50/50 population? I think this would help the current situation on Jericho.

[Kewlin 692]

21 hours ago, HighSociety said:

Maybe right but also keep in mind there are some people who just dethreat to play together with friends who are new to APB 

That's not a legitimate reason to dethreat, just F.Y.I., because the end result is 100% the same.

[Psygo 15]

4 hours ago, Kewlin said:

That's not a legitimate reason to dethreat, just F.Y.I., because the end result is 100% the same.

you're not wrong, but if threat segregation from districts was removed it wouldnt be necessary, playing with my bronze freinds as a somewhat high(i believe?) gold, he gets fucked up every time.

of course I could also end up being the one fucking everything up in other districts, but since threat segregation wasn't a thing, they might just match us with someone else who has a gold in their team.

[wayfie 0]

Sounds like they're going to make it like how Xbox/PlayStation has it, you just join a game with no other server instances. 

[TheMessiah 430]

22 hours ago, BIank said:

While eliminating visible threat levels, I would hate the removal of rank icons.

I think it would be a good idea to combine the idea of Open Conflict icon colors with the current ranks so they would now be red or blue instead of the current bronze, silver or gold.

(very quick mockup)

 

dat!!

 

Btw i was thinking about what can be done right now urgently for matchmaking and the idea to disable the option to start missions(press K)when player become gold in bronze district(like many people suggest)is actually very good.The same must be for bronzes making silver in green dist.Same gos for players dropping threats:bronzes becoming greens in silver dist. and silvers dropping to bronzes in gold dist.

 

In bronze dist.(talk about citadel)this golds are not actually true golds-thats high silvers(most,like me)True golds play in silver dist.Very little start dethreating.High/more experienced silvers start losing missions in silver dist. and takes them like at least hour to drop theyr gold.When they go in bronze and become gold again after 1-2 missions and cant play anymore(K disable)-this will at least limit theyr time in bronze dist.Most gonna start playn in theyr own silver dist. and find matches.Just not gonna be worthy the time to dethreat anymore.So in resume:still gonna be 2 threat dist. playable.In bronze where-new players,bronzes and middle/low skill silvers and in silver dist. where-middle/high skill silvers and true golds.Gonna be somethin like that till game become more populated.Thats how new/low skill players gonna be at least a bit more protected from good/high skilled players (in my opinion)

Edited October 21, 2018 by TheMessiah

[Poperon 141]

Is it just me, or i think we should consider a more simplistic approach?

In my opinion, we should just have 2 kinds of districts:

1) A Beginner district (tutorial/prep district), where only players ranged from R1 to R50 or R100 would be able to join, and this would serve as a preparatory district/tutorial district for players. Players higher than R100 (or any feasible number to be considered as a "ready" status for a player, meant that they already had enough time to "learn" about the game) would not be able to join. Here players would have the opportunity to learn about game mechanics, drive around, do their thing "in peace", in a slower pace. Even if rerolls (by rerolls here i mean players that already have other characters, creating a newer character) would not "stomp" players all that much, since their new characters are still "beginners". With this mechanic, the problem with dethreaters would cease, meaning that to join a "beginner district", a player would need to start all over again with a newer "fresh" character (meaning limited equipments), and just for a limited time period.

2) A General district (after tutorial/prep district), where all kinds of players could join. Here, players that achieved R50/R100/whatever chosen Rank/ would be ready to fight against everybody else. And here's the special thing about it: Even newly created characters may join this district. If they are rerolls or, if they know a little more about the game mechanics(returning players) and think they are already "suited" for this district and its challenges, he may join by his own will. This would not force "beginners" or "rerolls" to stay in the beginner district, and that would prevent the case of players suffering with a low pop/empty beginner district. In this scenario, i consider a growing gaming environment/population, where we always get newer players.

 

The threat icons would still remain just as a cosmetic/visual thing, without being considered as a "skill" factor for anything, and wouldn't be removed...
Rank would just be used as a value to be considered ONLY while joining one of both districts, and nothing else...

 

If you guys want to develop some other complex algorithm to add inbetween this concept, you feel free to do it.

(maybe use K/D for anything whatsoever...)

But i'm basically saying what i see in many online games, where they just calculate simple parameters and join people altogether. (Battlefield, COD, PUBG, etc)
The more complex a system is, the more the AI will toast its brains, and the mess we already have now will just endure, but in a "newly different way".

Edited October 21, 2018 by Poperon

[Kewlin 692]

8 hours ago, Psygo said:

you're not wrong, but if threat segregation from districts was removed it wouldnt be necessary, playing with my bronze freinds as a somewhat high(i believe?) gold, he gets fucked up every time.

of course I could also end up being the one fucking everything up in other districts, but since threat segregation wasn't a thing, they might just match us with someone else who has a gold in their team.

 

It's not necessary, they can just play with you in Gold, lol. There is no excuse for dethreating, purposefully dethreating means you're a selfish jerk, end of story.

 

27 minutes ago, Poperon said:

Is it just me, or i think we should consider a more simplistic approach?

In my opinion, we should just have 2 kinds of districts:

1) A Beginner district (tutorial/prep district), where only players ranged from R1 to R50 or R100 would be able to join, and this would serve as a preparatory district/tutorial district for players. Players higher than R100 (or any feasible number to be considered as a "ready" status for a player, meant that they already had enough time to "learn" about the game) would not be able to join. Here players would have the opportunity to learn about game mechanics, drive around, do their thing "in peace", in a slower pace. Even if rerolls (by rerolls here i mean players that already have other characters, creating a newer character) would not "stomp" players all that much, since their new characters are still "beginners". With this mechanic, the problem with dethreaters would cease, meaning that to join a "beginner district", a player would need to start all over again with a newer "fresh" character (meaning limited equipments), and just for a limited time period.

2) A General district (after tutorial/prep district), where all kinds of players could join. Here, players that achieved R50/R100/whatever chosen Rank/ would be ready to fight against everybody else. And here's the special thing about it: Even newly created characters may join this district. If they are rerolls or, if they know a little more about the game mechanics(returning players) and think they are already "suited" for this district and its challenges, he may join by his own will. This would not force "beginners" or "rerolls" to stay in the beginner district, and that would prevent the case of players suffering with a low pop/empty beginner district. In this scenario, i consider a growing gaming environment/population, where we always get newer players.

 

The threat icons would still remain just as a cosmetic/visual thing, without being considered as a "skill" factor for anything, and wouldn't be removed...
Rank would just be used as a value to be considered ONLY while joining one of both districts, and nothing else...

 

If you guys want to develop some other complex algorithm to add inbetween this concept, you feel free to do it.

(maybe use K/D for anything whatsoever...)

But i'm basically saying what i see in many online games, where they just calculate simple parameters and join people altogether. (Battlefield, COD, PUBG, etc)
The more complex a system is, the more the AI will toast its brains, and the mess we already have now will just endure, but in a "newly different way".

 

It's just you, this solution wouldn't work.

 

1) I don't know where you're getting your idea from that veterans can't stomp newbies just because they aren't geared up, but it's not correct. Every time I've started up a new account I destroyed everyone the game matched me with until it decided I was Gold, and let's be honest, even once you get to Gold the only real downside you have is not having CA3, which isn't a huge deal breaker IMO.

 

2) This doesn't even remotely make any sense, because being bad at APB has nothing to do with rank. People can be legitimately Bronzes all the way up to 255.

 

3) What do you mean threat icons stay but aren't considered as a skill factor? What does that even mean?

 

People need to realize that under no circumstances should rank and skill be associated with each other, as there simply is no correlation.

[Psygo 15]

19 minutes ago, Kewlin said:

 

It's not necessary, they can just play with you in Gold, lol. There is no excuse for dethreating, purposefully dethreating means you're a selfish jerk, end of story.

 

well, see, that's my point. if the threat segregation wasn't a thing, and people simply were in districts, this wouldn't be an issue.

[Fortune Runner 796]

14 hours ago, Kewlin said:

That's not a legitimate reason to dethreat, just F.Y.I., because the end result is 100% the same.

agreed. a true person who shouldn't be gold wouldn't have to dethreat anyways.

 

10 hours ago, Psygo said:

you're not wrong, but if threat segregation from districts was removed it wouldnt be necessary, playing with my bronze freinds as a somewhat high(i believe?) gold, he gets fucked up every time.

of course I could also end up being the one fucking everything up in other districts, but since threat segregation wasn't a thing, they might just match us with someone else who has a gold in their team.

people dethreat to play opponents who aren't true golds.

Removing threat segregation would instead chase people off the game from being forced to play people who are clearly beyond their scope of playing , which is why threat segregation was made to begin with.

To not separate players from threat level (skill) means to stick them all in one pot , and that just doesn't work.

I can not think of any successful game that does not have certain areas locked in some form so that only higher players can go in who are ready for it so that they don't get crushed then frustrated and rage quit the games.

 

edit : read more of your posts.....disappointed

to throw bronzes in the same pot as golds when SO MANY CRY ON THE FORUMS FROM GOLDS IN BRONZE...... how can you be deaf to their pain so much?

that's pretty cruel

and yes that is what you are saying ....I'm disappointed in you.

Edited October 21, 2018 by Fortune Runner

[gasolina 11]

my only concern is that how would matchmaking be done otherwise? say 4 high ranks play 6 mid ranks ? if it was something like that, just know that the spawn system needs to be fixed more than anything. being able to spawn 100m away or less with car spawner and just continuously throw bodies (and nades) at the objective just makes it impossible for the attacking team. 4v4s in certain spots are already hard enough

 

otherwise, everything sounds good to me. that would be my only concern though; how would matchmaking be done then.

Edited October 21, 2018 by trustdealer

[HighSociety 148]

I think several people (me included) pointed it out so many times. >"gold" is not "gold"<

Really don't know what all of u mean with "true golds" cuz the skill gap between gold players is way too far!

I am "gold" but i get stomped from high gold players and drop down to silver some(many)times.

And yes if there is no space on silver district and i am "silver" than i join bronze district and after 2-3 missions i am gold again...

Edited October 21, 2018 by HighSociety

[SK4LP 62]

"dethreating" and in an other side "team balance"

(contact and missions only for clan)

Edited October 21, 2018 by SK4LP

[mynd 120]

If there ends up being just one game world, what is going to happen with duplicate names from different regions?  I've my mains on all the regions so I personally wouldn't be affected.

 

Honestly, I wish we could switch to a modern naming system where anyone could have whatever name they wanted, but be differentiated by #1234 etc.