I believe how the system works is it gives a weighted score to alliance war rating and alliance prestige. Then the scores are tabulated, and #1 matches #2, #3 matches #4 etc.
In essence, the matchups will almost definitely be a fair one.
One issue this system doesn’t address is tanking and shell alliances. This can help by changing the way war ratings are calculated from an alliance perspective to an individual perspective, with the alliance rating being the average of all individuals in the alliance, and war ratings being part of score calculation. (I suggested this somewhere in the forums before)
This topic mentions collusion, and the company is working to address this with revisions to the mode, but this also seems to cause more problems for players. As was mentioned above, some teams intentionally plan their matchmaking searches for legitimate reasons, like making sure their teammates are free from work, school or are not sleeping. The game is global, and some players timezones will limit their availability to fight, especially if they are waiting for a node to be removed.
It just seems like a lot of work is being done to work around cheating, but this hurts more alliances in the process. Isn't it just time to remove rewards period to cheating teams and/or players (both seasonal and individual match rewards)? There are people who track cheating teams, they're exposed on places like YouTube and Reddit, and constantly we see the same names of teams and players who seem to hurt the game more.
I fully realize that there is revenue to be made, and that not all accusations are proof, and that also not all cheaters get caught (unfortunately), but this change in timing, while hopefully helpful for matchmaking, seems like a problematic response to teams that keep getting away with breaking the rules.
If the same cheaters keep getting around your attempts to counter them, when will they be removed?
As for the timezone issue, it would be best if Alliance Wars lost their timers then, so that players could move as they needed. Or that timers were permanently shortened, so that schedules could be worked around.
2 whole weeks off? One of which there’ll be no wars in which to earn loyalty. How about you actually give us a bit of time to enjoy wars without the pressure of being in season and a chance to restock some boosts and potions? Just a thought.
My alliance can 100% our AW map 6-7 hours after the first person starts moving with 5 energy. I don’t see how schedules is a matchmaking issue. It’s more like an efficiency issue.
2 whole weeks off? One of which there’ll be no wars in which to earn loyalty. How about you actually give us a bit of time to enjoy wars without the pressure of being in season and a chance to restock some boosts and potions? Just a thought.
This topic mentions collusion, and the company is working to address this with revisions to the mode, but this also seems to cause more problems for players. As was mentioned above, some teams intentionally plan their matchmaking searches for legitimate reasons, like making sure their teammates are free from work, school or are not sleeping. The game is global, and some players timezones will limit their availability to fight, especially if they are waiting for a node to be removed.
.
Who in the world will be affected the most? In the two alliances I have run, I have had 3 Battlegroups based on nationality (Plat 3 and now I run a Gold 1 to relax). One is EU based, one is SE Asia and Australian and one is US. It is very simple to work out - if I have a US based person who works night shifts, they go in a different BG.
It will work out, people are worrying about a minor thing that has an easy fix.
Love all the complaints by people saying timing is going to be an issue now.
Why was this complaint not made about aq over the years? Aq works but war wont even though it has far less energy requirements?
Also, how many of you complaining about coordination are in top alliances?
Scared of a fair fight?
we are gold 1 verge of plat 3, so not a top alliance but we can hold our own. The difference between AQ and AW though is we don't need all hands on deck to finish AQ. we have coordinated our bg's to spread out our international players to where if they don't move in phase 3 (early am for them) we can still clear with 8 guys (5x5), however with a fixed AW start/end we will not be able to rely on them to help take down bosses if needed which will mean unnecessary use of items for others. When we got to choose the start time we at least know our international guys will be awake and can throw in to take down bosses.
@Kabam Miike
Based on what I have seen the only restriction on possible out comes is 2 alliances cant face each other back to back. There needs to be 1 war they don't face each other between wars. WIll any information be provided on a restriction such as this now that its not able to be manipulated?
(btw thanks for your input on name policy yesterday)
If one alliance in Master bracket have no luck and get multiple Times NY and lose many wars because of one Alliance....
When you lose to them, your rating drops and it’s highly unlikely you’ll get them again in the next matchup given the unified matchmaking. The only way you have a chance of matching them again is you win your next war and they lose their next.
Thing is, if you’re fighting for the top spot, and if it’s a sole contention between you and them, you’re bound to meet them multiple times.
Wait. You’re afraid?
The issue is not Master alliances being afraid to match each other.
The issue is that there is a significant difference in rewards for top 3, but the current T1 rating/multiplier encompasses about 30 top alliances. There should be a difference between the amount of points earned for #1 vs #2 ranked alliance, and a #29 vs #30 ranked alliance as the power levels are actually quite different despite these two matches both being currently classified as T1. Either everyone in T1 should have equal opportunity to match, meaning #1 plays #30 occasionally, or award more points for a harder match for top 5 alliances competing for ranks 1,2,3.
This new system will help to improve the following:
Helps ensure that Alliances are being matched against other Alliances with similar capabilities.
Minimizes Alliances missing Wars based on matchmaking time.
Helps to minimize Collusion.
These are all important steps in making Alliance Wars and Alliance Wars Seasons a fair, and competitive mode!
It seems reasonable, and I see you found the problem (and basically the same solution) with odd numbers I mentioned regarding this type of system.
I think this does represent half of the solution to the overall problem with match making in general: the specifics of *how* matches are found still has problematic properties. But if you are actually doing a top-down match making system that prioritizes war rating, then many of those issues will likely get suppressed. I'm sure we can revisit them if they continue to occur after the new system is in place. Mostly, I'm thinking one issue gets muted and one will crop up. The old issue is the need to match on both war rating and alliance rating in order to "seek" matches when the system can't find a good one in time. The new system mostly nullifies that situation. But the new problem that might come up is that the "best match" might start to become the same alliances over and over in certain pathological situations.
That might be unavoidable sometimes: maybe sometimes number one and number two should be fighting it out most of the time if they also completely destroy everyone else. But in practice, I think the system should try to find random matches within "good enough" ranges so it doesn't get stuck in a local-minima of two alliances or a tiny group of alliances just happening to be the best matches for each other repeatedly.
If one alliance in Master bracket have no luck and get multiple Times NY and lose many wars because of one Alliance....
When you lose to them, your rating drops and it’s highly unlikely you’ll get them again in the next matchup given the unified matchmaking. The only way you have a chance of matching them again is you win your next war and they lose their next.
Thing is, if you’re fighting for the top spot, and if it’s a sole contention between you and them, you’re bound to meet them multiple times.
Wait. You’re afraid?
The issue is not Master alliances being afraid to match each other.
The issue is that there is a significant difference in rewards for top 3, but the current T1 rating/multiplier encompasses about 30 top alliances. There should be a difference between the amount of points earned for #1 vs #2 ranked alliance, and a #29 vs #30 ranked alliance as the power levels are actually quite different despite these two matches both being currently classified as T1. Either everyone in T1 should have equal opportunity to match, meaning #1 plays #30 occasionally, or award more points for a harder match for top 5 alliances competing for ranks 1,2,3.
Yup I agree. This is why I suggested a follow up system to this that incorporates war rating into season score calculation as well.
Let me find the post somewhere and post the link here.
This topic mentions collusion, and the company is working to address this with revisions to the mode, but this also seems to cause more problems for players. As was mentioned above, some teams intentionally plan their matchmaking searches for legitimate reasons, like making sure their teammates are free from work, school or are not sleeping. The game is global, and some players timezones will limit their availability to fight, especially if they are waiting for a node to be removed.
It just seems like a lot of work is being done to work around cheating, but this hurts more alliances in the process. Isn't it just time to remove rewards period to cheating teams and/or players (both seasonal and individual match rewards)? There are people who track cheating teams, they're exposed on places like YouTube and Reddit, and constantly we see the same names of teams and players who seem to hurt the game more.
I fully realize that there is revenue to be made, and that not all accusations are proof, and that also not all cheaters get caught (unfortunately), but this change in timing, while hopefully helpful for matchmaking, seems like a problematic response to teams that keep getting away with breaking the rules.
If the same cheaters keep getting around your attempts to counter them, when will they be removed?
As for the timezone issue, it would be best if Alliance Wars lost their timers then, so that players could move as they needed. Or that timers were permanently shortened, so that schedules could be worked around.
No complaints with this process for Alliance Quest
This is great news! Can't wait to see the changes "live"
But one question, will you be posting the explanation also in the other foreign Kabam orums?
Thanks
What's the cut off to have all members in your alliance to participate? Does it have to be before matchmaking or before enlistment?
It will be before matchmaking. You can find a new recruit during Enlistment.
So to clarify.... whether voluntary or involuntary if a change is to be made to roster (mid war week) someone has to leave/be kicked before war is over to have new recruit in for next war?
... not to mention scheduling complications with AQ.
It looks like you have a 4 hour matchmaking session and a 20 hr put your defense in section.... Perhaps add a 2 hour personnel change window and shorten defense to 18 hours?
I believe how the system works is it gives a weighted score to alliance war rating and alliance prestige. Then the scores are tabulated, and #1 matches #2, #3 matches #4 etc.
In essence, the matchups will almost definitely be a fair one.
One issue this system doesn’t address is tanking and shell alliances. This can help by changing the way war ratings are calculated from an alliance perspective to an individual perspective, with the alliance rating being the average of all individuals in the alliance, and war ratings being part of score calculation. (I suggested this somewhere in the forums before)
Nope. It’s not fair. If you’re at the top all you match is the top. Makes room for alliances dying 30+ times to finish in top 3
If one alliance in Master bracket have no luck and get multiple Times NY and lose many wars because of one Alliance....
When you lose to them, your rating drops and it’s highly unlikely you’ll get them again in the next matchup given the unified matchmaking. The only way you have a chance of matching them again is you win your next war and they lose their next.
Thing is, if you’re fighting for the top spot, and if it’s a sole contention between you and them, you’re bound to meet them multiple times.
Wait. You’re afraid?
The issue is not Master alliances being afraid to match each other.
The issue is that there is a significant difference in rewards for top 3, but the current T1 rating/multiplier encompasses about 30 top alliances. There should be a difference between the amount of points earned for #1 vs #2 ranked alliance, and a #29 vs #30 ranked alliance as the power levels are actually quite different despite these two matches both being currently classified as T1. Either everyone in T1 should have equal opportunity to match, meaning #1 plays #30 occasionally, or award more points for a harder match for top 5 alliances competing for ranks 1,2,3.
The problem with this idea is that it runs counter to the problem Kabam addressed by reducing the multipliers at the top: that the multipliers created the potential for too large a gap to overcome by lower alliances.
If you make the amount of points earned by the top three alliances too much higher than the points anyone else can get when they fight each other, they can create a situation where no one can catch them while they continue to fight each other. It might seem unfair for them to have "harder" fights and yet no more points, but in a competition the goal is in fact to make it harder to stay at the top than to get to the top to promote more competition.
However I'm not opposed to your idea to mix up the matching in the top 30 though: I think that's actually a good idea in principle. In fact, I suggested a modification to the match making system where the system actually matched alliances based on win/loss record within a season. In other words, in war #1 everyone gets matched just like now. But in war #2, all the winners get matched against each other, and all the losers get matched against each other. So instead of #1 matching against #2, the highest winner gets matched against the second highest winner, and then on down. In war #3, all the two and zero alliances get matched against each other, all the one and one alliances get matched against each other, and all the oh and two alliances get matched against each other. And so on.
Eventually, not all top 30 alliances will get matched against each other, but winners are forced to face other winners. You can't "dodge" even accidentally the strongest alliances. Well, except by losing. In effect, war seasons gets resolved in something like a pseudo-tournament. It isn't a real tournament because there are fewer rounds than are necessary for the amount of alliances. Twelve wars can only "match up" about four thousand alliances. But that's actually comparable to the amount of alliances in Gold 3 and higher. It is *almost* a tournament for alliances above tier 9 or so.
So there are set times where alliances can enlist, basically very short time frames......and you honestly expect that your servers are going to be able to handle all of that?.....even within the current state of the game with the constant connection issues and lag?
Good changes IMO.. sad that people abuse matchmaking so much they have to take control out of summoners hands like this especially if your alliance has some mixed time zones.
It’s a nice change, but to make it a little better you can allow the alliance to chose the time, when they decide to enlist, within the first 8hrs (11am, 12pm, 1pm ••••• 7pm).
So there are set times where alliances can enlist, basically very short time frames......and you honestly expect that your servers are going to be able to handle all of that?.....even within the current state of the game with the constant connection issues and lag?
If I'm reading the announcement correctly, you can enlist for the next war the moment the previous one starts. In other words, all the while you are placing and then attacking in one war, you can be enlisting for the next one. So you have about two days to enlist for the upcoming war. The window from 11:00 to 16:00 (11am-4pm) is when the servers will be doing match making calculations, it isn't when you have to enlist for that upcoming war.
And what will be done against shell alliances? I think even more alliances will change their alliances to trick matchmaking and getting easier opponents.
So there are set times where alliances can enlist, basically very short time frames......and you honestly expect that your servers are going to be able to handle all of that?.....even within the current state of the game with the constant connection issues and lag?
If I'm reading the announcement correctly, you can enlist for the next war the moment the previous one starts. In other words, all the while you are placing and then attacking in one war, you can be enlisting for the next one. So you have about two days to enlist for the upcoming war. The window from 11:00 to 16:00 (11am-4pm) is when the servers will be doing match making calculations, it isn't when you have to enlist for that upcoming war.
Well thank you, that sounds more manageable than I was thinking.
Comments
In essence, the matchups will almost definitely be a fair one.
One issue this system doesn’t address is tanking and shell alliances. This can help by changing the way war ratings are calculated from an alliance perspective to an individual perspective, with the alliance rating being the average of all individuals in the alliance, and war ratings being part of score calculation. (I suggested this somewhere in the forums before)
It just seems like a lot of work is being done to work around cheating, but this hurts more alliances in the process. Isn't it just time to remove rewards period to cheating teams and/or players (both seasonal and individual match rewards)? There are people who track cheating teams, they're exposed on places like YouTube and Reddit, and constantly we see the same names of teams and players who seem to hurt the game more.
I fully realize that there is revenue to be made, and that not all accusations are proof, and that also not all cheaters get caught (unfortunately), but this change in timing, while hopefully helpful for matchmaking, seems like a problematic response to teams that keep getting away with breaking the rules.
If the same cheaters keep getting around your attempts to counter them, when will they be removed?
As for the timezone issue, it would be best if Alliance Wars lost their timers then, so that players could move as they needed. Or that timers were permanently shortened, so that schedules could be worked around.
Just a thought. “Abit of time”? One week.
Who in the world will be affected the most? In the two alliances I have run, I have had 3 Battlegroups based on nationality (Plat 3 and now I run a Gold 1 to relax). One is EU based, one is SE Asia and Australian and one is US. It is very simple to work out - if I have a US based person who works night shifts, they go in a different BG.
It will work out, people are worrying about a minor thing that has an easy fix.
we are gold 1 verge of plat 3, so not a top alliance but we can hold our own. The difference between AQ and AW though is we don't need all hands on deck to finish AQ. we have coordinated our bg's to spread out our international players to where if they don't move in phase 3 (early am for them) we can still clear with 8 guys (5x5), however with a fixed AW start/end we will not be able to rely on them to help take down bosses if needed which will mean unnecessary use of items for others. When we got to choose the start time we at least know our international guys will be awake and can throw in to take down bosses.
Based on what I have seen the only restriction on possible out comes is 2 alliances cant face each other back to back. There needs to be 1 war they don't face each other between wars. WIll any information be provided on a restriction such as this now that its not able to be manipulated?
(btw thanks for your input on name policy yesterday)
The issue is not Master alliances being afraid to match each other.
The issue is that there is a significant difference in rewards for top 3, but the current T1 rating/multiplier encompasses about 30 top alliances. There should be a difference between the amount of points earned for #1 vs #2 ranked alliance, and a #29 vs #30 ranked alliance as the power levels are actually quite different despite these two matches both being currently classified as T1. Either everyone in T1 should have equal opportunity to match, meaning #1 plays #30 occasionally, or award more points for a harder match for top 5 alliances competing for ranks 1,2,3.
It seems reasonable, and I see you found the problem (and basically the same solution) with odd numbers I mentioned regarding this type of system.
I think this does represent half of the solution to the overall problem with match making in general: the specifics of *how* matches are found still has problematic properties. But if you are actually doing a top-down match making system that prioritizes war rating, then many of those issues will likely get suppressed. I'm sure we can revisit them if they continue to occur after the new system is in place. Mostly, I'm thinking one issue gets muted and one will crop up. The old issue is the need to match on both war rating and alliance rating in order to "seek" matches when the system can't find a good one in time. The new system mostly nullifies that situation. But the new problem that might come up is that the "best match" might start to become the same alliances over and over in certain pathological situations.
That might be unavoidable sometimes: maybe sometimes number one and number two should be fighting it out most of the time if they also completely destroy everyone else. But in practice, I think the system should try to find random matches within "good enough" ranges so it doesn't get stuck in a local-minima of two alliances or a tiny group of alliances just happening to be the best matches for each other repeatedly.
Poor guy. You might have to actually compete with someone on your level instead of steamrolling the lower alliances to stay on top. Tough life.
I remember now brought this up in March 2018 at the end of season 1 glad to see action taken
Yup I agree. This is why I suggested a follow up system to this that incorporates war rating into season score calculation as well.
Let me find the post somewhere and post the link here.
No complaints with this process for Alliance Quest
But one question, will you be posting the explanation also in the other foreign Kabam orums?
Thanks
So to clarify.... whether voluntary or involuntary if a change is to be made to roster (mid war week) someone has to leave/be kicked before war is over to have new recruit in for next war?
... not to mention scheduling complications with AQ.
It looks like you have a 4 hour matchmaking session and a 20 hr put your defense in section.... Perhaps add a 2 hour personnel change window and shorten defense to 18 hours?
Nope. It’s not fair. If you’re at the top all you match is the top. Makes room for alliances dying 30+ times to finish in top 3
The problem with this idea is that it runs counter to the problem Kabam addressed by reducing the multipliers at the top: that the multipliers created the potential for too large a gap to overcome by lower alliances.
If you make the amount of points earned by the top three alliances too much higher than the points anyone else can get when they fight each other, they can create a situation where no one can catch them while they continue to fight each other. It might seem unfair for them to have "harder" fights and yet no more points, but in a competition the goal is in fact to make it harder to stay at the top than to get to the top to promote more competition.
However I'm not opposed to your idea to mix up the matching in the top 30 though: I think that's actually a good idea in principle. In fact, I suggested a modification to the match making system where the system actually matched alliances based on win/loss record within a season. In other words, in war #1 everyone gets matched just like now. But in war #2, all the winners get matched against each other, and all the losers get matched against each other. So instead of #1 matching against #2, the highest winner gets matched against the second highest winner, and then on down. In war #3, all the two and zero alliances get matched against each other, all the one and one alliances get matched against each other, and all the oh and two alliances get matched against each other. And so on.
Eventually, not all top 30 alliances will get matched against each other, but winners are forced to face other winners. You can't "dodge" even accidentally the strongest alliances. Well, except by losing. In effect, war seasons gets resolved in something like a pseudo-tournament. It isn't a real tournament because there are fewer rounds than are necessary for the amount of alliances. Twelve wars can only "match up" about four thousand alliances. But that's actually comparable to the amount of alliances in Gold 3 and higher. It is *almost* a tournament for alliances above tier 9 or so.
If I'm reading the announcement correctly, you can enlist for the next war the moment the previous one starts. In other words, all the while you are placing and then attacking in one war, you can be enlisting for the next one. So you have about two days to enlist for the upcoming war. The window from 11:00 to 16:00 (11am-4pm) is when the servers will be doing match making calculations, it isn't when you have to enlist for that upcoming war.
Well thank you, that sounds more manageable than I was thinking.