Jump to content
Queen of Love

Matchmaking and Extiction of middle threat levels.

Recommended Posts

1 hour ago, glaciers said:

it’s easier to upthreat than dethreat due to poorly balanced score awards and anti dethreater mechanics g1 implemented, and threat percentages are not dynamic so there’s no limit on how many players can be gold

 

7 hours ago, CookiePuss said:

Sounds like they have been playing against players less skilled than them. The only way you go up in threat is finishing in the top half of the scoreboard. Your friends have been doing that.

 

They're probably getting just enough bonus score from completing and defending objectives to prevent them for losing threat. There are some missions where they get 0 score, which means it doesn't count against their threat. I don't get the bonus for defending objectives since I'm usually away from the objective spawn camping, I mean tactically watching for spawns.

Share this post


Link to post
Share on other sites
On 10/30/2022 at 10:20 AM, Revoluzzer said:

True, because everyone manually joins districts.

Letting the game decide which district it puts you in should prevent this. Although colour-locked districts mean that system is either disabled these days or severely handicapped.

 

You need to update your knowledge, then, because those things are very much incorrect.

That's how it was a year or so after gold lock when they changed to the Win/Loss top 50% bottom 50% system where score in a match became the determination for threat level rather than just straight win/loss that it used to. There may still be tiers and the like, but there is definitely a "match number" limit that determines rank (which I never heard they removed) you are based on the last x matches, and they still should have the top bottom 50% rank increase decrease based on match score. Because they never said they removed it.

 

I know they made changes since it's been 5-6 years since that implementation, as i haven't been nearly as active as i used to keep tabs. But i also never heard of them changing from the score based and past x match system combination, even with the glicko tiers likely still in place.

Edited by Noob_Guardian

Share this post


Link to post
Share on other sites
On 11/4/2022 at 4:18 PM, Noob_Guardian said:

That's how it was a year or so after gold lock when they changed to the Win/Loss top 50% bottom 50% system where score in a match became the determination for threat level rather than just straight win/loss that it used to. There may still be tiers and the like, but there is definitely a "match number" limit that determines rank (which I never heard they removed) you are based on the last x matches, and they still should have the top bottom 50% rank increase decrease based on match score. Because they never said they removed it.

There never was a 50/50 split. Ever. Some people misread the contents of the blog post which introduced the score-system and concluded there was .

 

Unfortunately, that old blog post seems to have been removed when LO bought G1 and changed the blog.

Fortunately, the internet archive exists. So here's the archived version of the 'Settle the Score' update blog post.

Edit: I've been informed by @Amayii that the old blog-entries have been archived on blogspot. So here's the blogspot-link for that particular article.

 

And here's the relevant bit from that blog post:

"At the end of each Mission the system will rank everyone according to their Score (people that win matches get a bonus to their score that averages around the same amount of Score as about five kills). The Threat system looks at everyone’s score and then ranks them against each other as if it were a free-for-all match. While it’s likely that the team that won will have the most score, it’s not always the case. We've seen occasions on OTW where the losing side had higher individual scores but these were when matches were incredibly close calls."

 

As the blog indicates, this part was edited for clarity after the initial release. Particularly because of the 50/50 issue, iirc.

 

Also interesting:

"Getting a significant amount of score in one match compared to everyone else won't move your threat any more than a very close win but it will give you a significant amount of cash as a bonus."

 

This suggests there is a hard limit to threat mobility per mission. Or in other words: Even if you significantly exceed the threat-system's expectations, it won't move you around drastically all at once.

It probably impacts your confidence-value more significantly, though.

Edited by Revoluzzer
  • Like 1

Share this post


Link to post
Share on other sites
1 hour ago, Revoluzzer said:

There never was a 50/50 split. Ever. Some people misread the contents of the blog post which introduced the score-system and concluded there was .

 

Unfortunately, that old blog post seems to have been removed when LO bought G1 and changed the blog.

Fortunately, the internet archive exists. So here's the archived version of the 'Settle the Score' update blog post.

Edit: I've been informed by @Amayii that the old blog-entries have been archived on blogspot. So here's the blogspot-link for that particular article.

 

And here's the relevant bit from that blog post:

"At the end of each Mission the system will rank everyone according to their Score (people that win matches get a bonus to their score that averages around the same amount of Score as about five kills). The Threat system looks at everyone’s score and then ranks them against each other as if it were a free-for-all match. While it’s likely that the team that won will have the most score, it’s not always the case. We've seen occasions on OTW where the losing side had higher individual scores but these were when matches were incredibly close calls."

 

As the blog indicates, this part was edited for clarity after the initial release. Particularly because of the 50/50 issue, iirc.

 

Also interesting:

"Getting a significant amount of score in one match compared to everyone else won't move your threat any more than a very close win but it will give you a significant amount of cash as a bonus."

 

This suggests there is a hard limit to threat mobility per mission. Or in other words: Even if you significantly exceed the threat-system's expectations, it won't move you around drastically all at once.

It probably impacts your confidence-value more significantly, though.

It's hard to say that it was done more towards "clarity" rather than to be more tight lipped on how to effectively dethreat.

 

Even if it pits -everyone- against the other as if it was a free for all. There still has to be a determining factor with score, by points and where each person places, that influences whether someone went up or down and by how much.

 

You are free to assume then that the top 50% which would likely be the winners, would likely gain threat, and the lower ones lose it. This percent ofc could be different, it could be top 40/30/20% of earners go up, have a no change, and then -reduce threat- percentage. With the percentage simply being the number of players that scored lower.  - With it at 50%, even with the losing team, they could still gain rank to some degree, as they would have performed better than at least 1/2 the winning team if they did well. If it's you need to be top 40 or 30% of earners to go up, that's fine as well. But it certainly does NOT seem the case by any means. If it's a different percentage that gains threat based on the "free for all", that's fine. But for simplicity sake,  (and lack of knowing the further intricacies and possibly seeing a few posts about it at some point) 50% is a good median number to indicate how well you -need to do- to -not lose threat-.

 

If you are second from top score consistently, but not the top, you should - not lose- threat. It would be dumb to assume you did, considering that wouldn't be fair especially when on the winning team consistently, imagine going silver by winning games but still being 2nd from top like that. The only times i've seen people lose threat in this system (even with a win) is when they are consistently bottom 50% low performers, even among the winning team. (won, but provided nothing, compared to the other team and their own)

 

Even if it's a "misread". It's still a fair assumption, and with how G1 was handling things back in 2013, it's likely they wanted to limit knowledge on "how to dethreat effectively" by being vague and editing it. (We had a massive dethreat problem back then.)

Remember, you used to need negative score with 1 kill to dethreat a -lot-? They fixed it so you negative shouldn't count it towards dethreating. Which means you still need to score, but score -lower- than some amount of players. If placing bottom 50% is wrong, that's fine, but it's still more accurate then pretending that there isn't some % of players in the match you have to do better than, to not lose threat. We may not know that % completely, but it's definitely there, if each players score is weighted against the other to place who gains and who loses.

Edited by Noob_Guardian

Share this post


Link to post
Share on other sites
1 hour ago, Noob_Guardian said:

If you are second from top score consistently, but not the top, you should - not lose- threat. It would be dumb to assume you did, considering that wouldn't be fair especially when on the winning team consistently, imagine going silver by winning games but still being 2nd from top like that. The only times i've seen people lose threat in this system (even with a win) is when they are consistently bottom 50% low performers, even among the winning team. (won, but provided nothing, compared to the other team and their own)

you've never seen someone win a mission but dethreat or vice versa?

 

thats because the player failed to achieve the performance that the matchmaking system predicted, regardless of where they place on the scoreboard - if the system thinks you'll obliterate 2 bronzes and you only end up barely breaking even scorewise, you'll lose threat even as the top scorer

 

edge cases aside, from a practical perspective you're right that the finer details and exact percentages don't matter for the large majority of players, especially without much more granular threat representation

  • Like 1
  • Thanks 1

Share this post


Link to post
Share on other sites
On 11/8/2022 at 5:34 PM, glaciers said:

you've never seen someone win a mission but dethreat or vice versa?

 

thats because the player failed to achieve the performance that the matchmaking system predicted, regardless of where they place on the scoreboard - if the system thinks you'll obliterate 2 bronzes and you only end up barely breaking even scorewise, you'll lose threat even as the top scorer

 

edge cases aside, from a practical perspective you're right that the finer details and exact percentages don't matter for the large majority of players, especially without much more granular threat representation

I said i have, but it's in cases where the player is consistently in the bottom 50% of everyone in the match score wise.

 

I doubt it. If you barely break even, and still get a win, you still gain score worth 5 kills over the other team. Unless you mean total. In that case, you're still 1st, you should not lose threat in that case as if free for all, you'd still be first on top. I've NEVER seen a "top scorer" who lost and was above the 50% mark dethreat -ever-, have you? I've seen the scorer on the enemy team or my own that is bottom 1/2 of 3/4 dethreat, even on a win. But i've never seen anyone who appeared to be top 50%+ dethreat. But I also never put that much thought into -that case- specifically, because the few times i recall the top scorer ever did dethreat as an enemy, was they had less score than even the worst player my team did overall. Which would have put them ~below the 50% mark.

Edited by Noob_Guardian

Share this post


Link to post
Share on other sites
On 11/8/2022 at 10:15 PM, Noob_Guardian said:

Even if it pits -everyone- against the other as if it was a free for all. There still has to be a determining factor with score, by points and where each person places, that influences whether someone went up or down and by how much.

And there is.

Your final score of the mission is put against each other's final score individually and as a result you gain or lose threat and/or confidence for each of those pairings.

High confidence means threat moves little, low confidence means threat moves lots.

Meeting expectations increases confidence, but not threat,

Missing expectations lowers confidence and threat.

Exceeding expectations lowers confidence and increases threat.

 

On 11/8/2022 at 10:15 PM, Noob_Guardian said:

But for simplicity sake,  (and lack of knowing the further intricacies and possibly seeing a few posts about it at some point) 50% is a good median number to indicate how well you -need to do- to -not lose threat-.

If everyone in the mission is roughly at the same threat level, then yes, this is a practical guideline.

When threat levels are more spread out, though, this guideline falls apart. Staying on this guideline-system for the moment: if the system expects a low threat player to perform in the bottom 25% of the field, but they perform in the bottom 25-50% (e.g. 4v4 and their total score places them at 6 out of 8), they would gain threat.

 

I would generally advise against the mindset of "need not to lose threat", by the way. Threat is not a progressive system. It's merely a tool for matchmaking. Losing threat can be perfectly fine and should lead to better, more enjoyable matches. I'm pretty sure I lost threat on every mission I played these days, because I haven't played in years.

 

On 11/8/2022 at 10:15 PM, Noob_Guardian said:

If you are second from top score consistently, but not the top, you should - not lose- threat. It would be dumb to assume you did, considering that wouldn't be fair especially when on the winning team consistently, imagine going silver by winning games but still being 2nd from top like that.

If a group of four Golds faces a group of four Silvers, the threat system will expect the Golds to achieve a certain score. If you place 2nd on the scoreboard, but don't score as much as the system expected you to, you might lose threat (and the Silvers might gain threat).

If you consistently perform as the system expects you to, it will gain confidence in your current threat level and you'll have to lower that confidence before your threat starts moving noticeably again.

 

But the system does not expect you to hit a specific placement on the scoreboard, it expects to you score in relation to each other player. Ergo it also matters who was placed 1st, 3rd and so forth.

 

On 11/10/2022 at 2:51 AM, Noob_Guardian said:

I've NEVER seen a "top scorer" who lost and was above the 50% mark dethreat -ever-, have you?

You only ever see someone dethreat if their colour changes. Which is exceedingly rare as playtime increases (because the confidence value will be more affected than threat after a while). So you will usually see it on new accounts and dethreaters.

Nevertheless a person gains and loses threat, even if the colour of their badge doesn't change.

Share this post


Link to post
Share on other sites

should have an "antiroxx system"

gold get fragile mod

green and bronze get extra hp

(and remove the ability to "abandonmission" (favorite gold trick to stay gold ;D))

Edited by SK4LP
  • Like 1

Share this post


Link to post
Share on other sites
3 hours ago, SK4LP said:

should have an "antiroxx system"

gold get fragile mod

green and bronze get extra hp

(and remove the ability to "abandonmission" (favorite gold trick to stay gold ;D))

Apparently never happen. And if it would, it will just raise up toxicity level even more, as it wasn’t that high lol. 

I remember ppl ranting about the non-threat districts year ago, they say that game will die at last... population still the same after all this time lol (yes, on NA too, my fellow burger cowboys). Pointless decision, pointless change, just more misery.

Edited by Nagletz
  • Like 1

Share this post


Link to post
Share on other sites
On 11/22/2022 at 12:01 PM, SK4LP said:

should have an "antiroxx system"

gold get fragile mod

green and bronze get extra hp

(and remove the ability to "abandonmission" (favorite gold trick to stay gold ;D))

I would be enough if the matchmaking was actually dong its job and doesn't always stack golds vs all other threats.

I don't get why starting a mission without opposition is even still a thing. How hard it is to make it start only when there are enough people in both teams?

And speaking of teams, 3v3  should be the minimum team size to even have a proper match. 

 

Fix that and the population may get above 200 people. 

Edited by PotatoeGirl
  • Thanks 1

Share this post


Link to post
Share on other sites
6 minutes ago, PotatoeGirl said:

I would be enough if the matchmaking was actually dong its job and doesn't always stack golds vs all other threats.

I don't get why starting a mission without opposition is even still a thing. How hard it is to make it start only when there are enough people in both teams?

And speaking of teams, 3v3  should be the minimum team size to even have a proper match. 

 

Fix that and the population may get above 200 people. 

Because it has literally no other choice. Otherwise you would be waiting forever to get a game, which would drive away even more people than bad games.

Share this post


Link to post
Share on other sites
5 minutes ago, PotatoeGirl said:

I don't get why starting a mission without opposition is even still a thing. How hard it is to make it start only when there are enough people in both teams?

And speaking of teams, 3v3  should be the minimum team size to even have a proper match. 

because the more the matchmaking system is forced to follow the rules, the less matches it creates

 

the alternative is no unopposed matches and no unfair matches, but 30/60/90 minutes waiting for a mission instead

Share this post


Link to post
Share on other sites
14 minutes ago, NotZombieBiscuit said:

Because it has literally no other choice. Otherwise you would be waiting forever to get a game, which would drive away even more people than bad games.

Well, you are not playing much if all you get is no opposition matches.

The other option is to at least give minimal reward for completing non opposed missions to encourage staying to the end (which to be given only at certain server population to prevent farming on empty servers)

 

Edited by PotatoeGirl

Share this post


Link to post
Share on other sites
3 hours ago, ninjarrrr said:

game has no players because everyone is in social doing cybersex

Maybe you are.

 

 

3 hours ago, PotatoeGirl said:

Well, you are not playing much if all you get is no opposition matches.

The other option is to at least give minimal reward for completing non opposed missions to encourage staying to the end (which to be given only at certain server population to prevent farming on empty servers)

 

I've had sessions where I would struggle to get a single match an hour. For multiple hours. Of course this was when grouped up though, but even solo I've struggled sometimes.

Share this post


Link to post
Share on other sites
1 hour ago, NotZombieBiscuit said:

Maybe you are.

can't say i have but i do think you are the type of person to do it for sure

  • Like 2

Share this post


Link to post
Share on other sites
5 hours ago, ninjarrrr said:

can't say i have but i do think you are the type of person to do it for sure

Why are you assuming such a thing?

Share this post


Link to post
Share on other sites
9 hours ago, NotZombieBiscuit said:

Why are you assuming such a thing?

Probably because APB community is full of deviants, just my personal observation. Noticing how much time ppl waste in the game, y’all should be still virgins.

Edited by Nagletz

Share this post


Link to post
Share on other sites
On 11/24/2022 at 11:12 PM, NotZombieBiscuit said:

Why are you assuming such a thing?

australia 

Share this post


Link to post
Share on other sites

Another option is to compare the scores of rivals. Very simple. The ratio of kills to deaths affects the remoteness of the spawn point. Killed three, died 1 time - respawn 300 meters from the point. Your score is 10 to 10 - respawn 100 meters from the point. You are a noob with a score of 0 to 5 - respawn 50 meters from the point. When the multiplier is increased by one, the spawn point moves away by 100 meters. With a negative multiplier, the spawn point gets closer by 10 meters.

 

Merged.

 

Today I played for three hours. All missions were difficult for me, it is exhausting and demotivating. Playing the role of a piece of meat is not fun. Removed all the mechanics that restrain strong players from the game. The fifth threat level is now meaningless. There is no division of regions. Everything is aggravated by the loss of frames during a shootout, macros, high ping, long processing of data on the server. We are all waiting for 10 years of wonderful moment, customer upgrade! We look forward to and believe in this as in the third coming! Especially developers. They are diligently repairing the engine, but do not pay attention to the fact that the gas is almost over. And all you need to do is to hammer a small stick into a leaky gas tank!

  • Like 1

Share this post


Link to post
Share on other sites
1 hour ago, Yapopal said:

Another option is to compare the scores of rivals. Very simple. The ratio of kills to deaths affects the remoteness of the spawn point. Killed three, died 1 time - respawn 300 meters from the point. Your score is 10 to 10 - respawn 100 meters from the point. You are a noob with a score of 0 to 5 - respawn 50 meters from the point. When the multiplier is increased by one, the spawn point moves away by 100 meters. With a negative multiplier, the spawn point gets closer by 10 meters.

 

 

 

 

This might be the worst suggestion I've heard yet. Kill count is not a good indicator of skill/win rate. Plus a whole bunch of other reasons.

Share this post


Link to post
Share on other sites
21 hours ago, NotZombieBiscuit said:

This might be the worst suggestion I've heard yet. Kill count is not a good indicator of skill/win rate. Plus a whole bunch of other reasons.

This is not the first time I hear this from you. You are talking about the number of murders, but I wrote about the ratio of murders and deaths. These are different things. The kill to death ratio is a good indicator of a player's abilities.

 

Merged.

 

I offered several options for finding the balance of the team. I don't know if they will work or not. To do this, you need to introduce them into the game and test them. To say for sure that this is the worst idea that I have heard is not reasonable.

This topic discusses the inequality in the level of play of opponents. Everyone knows that it is difficult to pick up an enemy because of the small population. This problem cannot be solved by separating areas according to threat levels. We need a balance of power between the teams. Not by individuals, but by teams. This balance must be found at the beginning of each mission and changed as you progress through the stages. This will make the gameplay really interesting. The weak will receive support, the strong will be restrained within certain limits. If the teams are initially equal in strength, then they will not notice any changes in the gameplay at all.

  • Like 2

Share this post


Link to post
Share on other sites
3 hours ago, Yapopal said:

I offered several options for finding the balance of the team. I don't know if they will work or not. To do this, you need to introduce them into the game and test them. To say for sure that this is the worst idea that I have heard is not reasonable.

This topic discusses the inequality in the level of play of opponents. Everyone knows that it is difficult to pick up an enemy because of the small population. This problem cannot be solved by separating areas according to threat levels. We need a balance of power between the teams. Not by individuals, but by teams. This balance must be found at the beginning of each mission and changed as you progress through the stages. This will make the gameplay really interesting. The weak will receive support, the strong will be restrained within certain limits. If the teams are initially equal in strength, then they will not notice any changes in the gameplay at all.

the problem with this is you are punishing players for playing well and rewarding them for playing poorly, which is the opposite of what should happen

 

it also seems very exploitable at first glance, what stops players from intentionally playing poorly for the entire mission so that they get extra buffs, and then playing well in the final stage while abusing those buffs?

Share this post


Link to post
Share on other sites
9 hours ago, Yapopal said:

This is not the first time I hear this from you. You are talking about the number of murders, but I wrote about the ratio of murders and deaths. These are different things. The kill to death ratio is a good indicator of a player's abilities.

 

That is pretty much under the same umbrella of what I was reffering to. Kill count, KDA, Ratio of kills between teams, ratio of one player with rest of team, blah blah, etc. They are all not good indicators of skill within the realm of APB

 

Merged.

 

5 hours ago, glaciers said:

the problem with this is you are punishing players for playing well and rewarding them for playing poorly, which is the opposite of what should happen

 

it also seems very exploitable at first glance, what stops players from intentionally playing poorly for the entire mission so that they get extra buffs, and then playing well in the final stage while abusing those buffs?

Be team of four. Have three players be the 'slayers', one be the objective abuser and get 50m spawns to just zerg the objectives. Easy wins. 

 

Hey, kevlar might even be useful if I have to only run 50m to the fight and sit my fat patootie on a objective.

Share this post


Link to post
Share on other sites
On 12/1/2022 at 8:58 PM, Yapopal said:

Another option is to compare the scores of rivals. Very simple. The ratio of kills to deaths affects the remoteness of the spawn point. Killed three, died 1 time - respawn 300 meters from the point. Your score is 10 to 10 - respawn 100 meters from the point. You are a noob with a score of 0 to 5 - respawn 50 meters from the point. When the multiplier is increased by one, the spawn point moves away by 100 meters. With a negative multiplier, the spawn point gets closer by 10 meters.

This would certainly make it more comfortable for good players to farm bad players. In fact, spawning 50 m from the objective is very much within the movement radius of decent players. Spawn. Die. Repeat. Welcome to the suck.

  • Thanks 1

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×
×
  • Create New...