Jump to content


Alternative Matchmaker

MM Matchmaker 3-5-7 Tier 8 Frustration Battle Weight Old System MM 2.0 pre-9.18

  • Please log in to reply
12 replies to this topic

speedphlux #1 Posted 28 May 2018 - 12:11 PM

    Lieutenant

  • Player
  • 42029 battles
  • 1,681
  • [TZAR] TZAR
  • Member since:
    01-28-2011

Since I don't exactly remember how the Battle Weight system was made up and what values it had, I took it upon myself to try figure out a way to fix MM, with a similar system, instead of "the new and improved MM 2.0". First off, unlike WG, I'm going to try and balance the vehicles from the Tier 1 upwards (they did it from Tier 10 downwards in case you didn't knew). Please note, no changes (buffs or nerfs) to current vehicles would be made.

 

A completely stock vehicle, has the Battle Weight score of it's own Tier, as an example. A stock IS-3 has BW of 8. A stock JagdTiger has BW of 9. And so on. Once a module is upgraded, a certain value is added to that number. In the case of the IS-3, a top gun would add 0.22 to the BW score (fictional number that I just came up on the fly). A top turret adds another 0.17 to it. A top engine adds another 0.08 to it. I don't think that Tracks and/or Radios should add any points, but we're free to discuss those things as much as we'd like to. Obviously, we can't compare a Stock IS-3 to Elite IS-3 on the battlefield, hence the difference in their scores. With this system, a stock IS-3 is considered an IS-3 that lacks the BL-9 gun, doesn't have the top turret and lacks the top engine. Clearly - a Tier 8 vehicle with the score of 8 BW. Fully Elite IS-3 on the other hand, has all the bells and whistles and has BW score of 8.47. I hope you start to see where this is going.

 

With these scores in mind, all we're going to ask from the BW MM, is to make up teams with as close scores as possible and with same types of vehicles as possible. Keeping the limit of 3 arties per side and maybe introducing another one of max 5 TDs per side. I don't think "campy" games are good for business, but that suggestion can also be discussed. Of course there should be a margin of error on the BW scores of each team, as it would probably be impossible to make them absolutely equal and still have low queue times. As to what number should the difference be allowed to grow, I'm not prepared to give an answer to at the moment, so I'll just leave it "free". Obviously, you won't have a team consisting of fully stock Tier 8s vs a team that is made up of fully elite Tier 8s, as their scores would differ way too much and MM would always look around to shuffle the tanks a bit, so that the team numbers are as close as possible.

 

But what about platoons ? Well, if we have a platoon of stock Tier 8s, vs platoons of elite Tier 8s, they obviously can't be on the same team, as this would create a disadvantage for their opponents. So the platoon with the fully elite Tier 8s, is just going to have to be on the team with few more stock Tier 8s to even the playing field. 15 stock vs 15 elites, right ? You can't cut them in equal halves anyways, so one team is always going to have 8 elites and 7 stock and the other team is going to have 7 elites and 8 stocks. Guess where the platoon with the elite Tier 8s is going to go to ? To put it simple, we can maybe add +1 to BW score for a 2-men platoon and +2 to 3-men platoon.

 

As we know, some tanks are "more equal" then others. In a completely Soviet system of socializm and their equality of outcome, we're just going to accept this, without nerfing or buffing tanks, and assign each BW score individually. That's why I took the IS-3 as a starting point - because it's your Joe Average in terms of Tier 8 Heavy performance. But what about a Tiger II ? Well, it's also a Tier 8, obviously, with not as stelar armor, but pretty OK gun, even when stock, so I guess it can also start up with BW score of 8 when fully stock. However, I don't think that it's top gun should add the same score as the IS-3's top gun, as it's simply inferior. Sure, it's more accurate, but the lack of alpha for that class of vehicles is rather annoying, so I'd give it 0.20. Not much of a difference compared to the IS-3 and that would certainly not break the MM if those two tanks happen to be the "odd difference" between the two teams. Is the Tiger II top turret as good as the IS-3's ? Yes and no - more view range, but annoying weakpoint and huge flat side profile. But also AFAIK, it gave boost to the gun's performance, so I would say - 0.16 BW. The engines on the two are comparable upgrades, but because the Tiger II is the heavier vehicle, I'd give it higher score to underline that the Tiger II actually does need that upgrade. So - 0.1. The end score - Elite Tiger II BW is 8.46. Just 0.01 difference with the fully elite IS-3, which kinda makes sense in most cases.

 

But what about the O-Ho(ho-ho-ho) ? Well, first off - this tank doesn't even have a Turret upgrade, so we can factor that into it's base BW level straight away. That way this tank can start up with BW of 8, plus the average score for Tier 8 Heavy Top Turret upgrade. Let's say it's 0.16. That makes the stock O-Ho start off as 8.16 straight away. Of course, we can look at the turrets individually and disregard the fact that the tank doesn't have "a top turret upgrade" and just individually give it's turret score, which will give away how good it is compared to other Tier 8 Heavies upgraded turrets. In that case, the O-Ho might score higher, boosting it's BW score quite a bit. Also, given that the tank's top engine is researchable at Tier 5 (!), it's very likely that everyone would have it by now from their first battle in the vehicle. But let's not assume that and just score it accordingly (NOTE: that engine will have different score on each tank that can use it, because tank weight plays a bit factor on how an engine performs and boosts your mobility). The power-to-weight ratio of the O-Ho is still comparable to the Tiger II, but the tank's top speed and ground resistances let it down, so I'd give that engine a BW score of 0.05. With the guns however, is where things gets interesting. Now, for the sake of testing out, we're going to assume that the 10cm Kai gun is the top gun of that tank. It's an OK gun. Inaccurate, but higher DPM then the Tiger II's, but that tank also has armor, so it can use it at close distance, unlike the Tiger II, so it's basically around the same score as the Tiger II's - 0.20. But what about the 15 cm Derp gun ? Well, we can straight up consider "Derps" as alternative top guns and score them accordingly, because they are actually "always a viable option". Yes, even on the M46 Patton, as proven by Circon. So what about the Derp gun on the O-Ho ? Well, far more inaccurate, but as we established already - that tank can brawl at close range. To top it off, that gun always does damage when it hits a target, plus the potential of module and crew damage. It's also not as useless as the 10cm Kai gun when faced with Tier 10s, so I would actually dare to score that gun as 0.25. So what the score ends up being ? Elite O-Ho with the 10cm Kai gun is 8.41 if we just take the average Tier 8 Heavy Turret score. Which clearly needs to be altered, because there is no way in hell that tank's weight in battle is less then that of the Tiger II or the IS-3. So we'll have to go back and look at that Stock/Elite Turret score again. 200mm armor all around, decent view range ... I'd say maybe a score of 0.30. With those changes, the math shows a score of 8.55 for the 10cm Kai gun and 8.60 with the Derp gun. Sounds about right to me.

 

With those things in mind, we can keep on assigning scores to every single vehicle in the game, balancing them according to their own tiers first and foremost. Then we can look up at how are their handing pressure versus higher tier machines and consider those things into our simple math by adjusting the numbers accordingly. Even if we scrap the equal class distribution, that BW MM is still going to prove superior to WG's MM 2.0. Specially when dealing with tanks that are not Tier 10s. Teams may end up unbalanced, like one team having 2 extra bottom tiers and 2 less top tier vehicles, but the raw numbers of the BW would show that the teams are actually well balanced. And if the extra 2 top tiers on one side help their side wins, is not because the other team had 2 less top tiers, but because their top tiers were not adequate tankers. Imagine a 3-5-7 vs 5-5-5 teams, in a tier 8 game. the 3 is a platoon of O-Ho's vs the enemy's 5, which consists of Caernarvon, Tiger II, WZ-111, Super Pershing and JadgTiger 8.8. I'd say that is perfectly balanced top of the team. You may argue that the numbers will overwhelm the trio, but what you're missing is that the equal 5xTier 7s on each team were also part of the balance equation. In other words, the trio may have some really strong Tier 7s to back them up, while the other team's Tier 7s may include few Stock vehicles to balance the equation out.

 

Maybe as we're trying to balance out tanks in that fashion, we'd have to consider other values and factors as well. Think about set Tier and Class for example. A Tier 8 TD in general overvalues a Tier 8 Heavy, which in terms overvalues a Medium, which in terms overvalues a Light tank, which in terms overvalues an SPG. Differences between Class and Tier may varry thruout the tech tree, as some Mediums and Heavies for example, are much closer to each other in terms of performance (think Tiers 4 & 5 as an example). Some TDs, even when stock, completely overshadow everything else in their Tier (think WT Pz IV) in terms of what influence they can have on the battlefield, while others are utter garbage that doesn't even challenge same Tier Light tanks (think ARL v39 while stock).

 

Obviously, you get the idea that in some vehicles, you might be bottom tier a bit more often then with others. Unfortunately, without proper vehicle balance, some of those OP tanks, would have to "suffer" because of their OP status. As for how exactly are we going to balance out the Premium vehicles in that BW system - it's rather simple really. We keep the PMM status on whatever vehicle has it, and then adjust their numbers according to their performance. So yeah, a Defender might end up having BW score of 8.45 as an example, while an IS-6 may dip to 7.8 BW score, simply because the recent power creep made it feel like a sub-Tier 8 vehicle. Still thou, a bit too OP to be classified as a straight up Tier 7 machine.

 

As you can see, balance is a tricky act. But while the old Battle Weight System was flawed and resulted in utter garbage matchups, those games were extremely rare. The new system is just flawed, because it doesn't make difference between an VK 4502 A and O-Ho. between a Scorpion G and a JagdTiger 8.8, between Primo Victoria and T-34-2. All it cares about is that "similar Classes" face off each other. The 3-5-7 distribution is not that bad, but when you present the teams with completely different set of vehicles on each level, you get those 3 minutes 15-5 games that are extremely frustrating. Even when you're on the winning side, because you rarely have time to rack up the damage you'd know you're capable of dishing out.

 

Also, please note that the numbers I talked about are just some random figures that do not represent an actual study of the tanks mentioned. It's just something of a guideline under which we can all sit down and talk about in a calm manner. And if you have any idea about other numbers we'll need for the equation, please feel free to suggest them.



jabster #2 Posted 28 May 2018 - 12:30 PM

    Field Marshal

  • Beta Tester
  • 12516 battles
  • 22,149
  • [WSAT] WSAT
  • Member since:
    12-30-2010

View Postspeedphlux, on 28 May 2018 - 11:11 AM, said:

Since I don't exactly remember how the Battle Weight system was made up and what values it had, I took it upon myself to try figure out a way to fix MM, with a similar system, instead of "the new and improved MM 2.0". First off, unlike WG, I'm going to try and balance the vehicles from the Tier 1 upwards (they did it from Tier 10 downwards in case you didn't knew). Please note, no changes (buffs or nerfs) to current vehicles would be made.

 

A completely stock vehicle, has the Battle Weight score of it's own Tier, as an example. A stock IS-3 has BW of 8. A stock JagdTiger has BW of 9. And so on. Once a module is upgraded, a certain value is added to that number. In the case of the IS-3, a top gun would add 0.22 to the BW score (fictional number that I just came up on the fly). A top turret adds another 0.17 to it. A top engine adds another 0.08 to it. I don't think that Tracks and/or Radios should add any points, but we're free to discuss those things as much as we'd like to. Obviously, we can't compare a Stock IS-3 to Elite IS-3 on the battlefield, hence the difference in their scores. With this system, a stock IS-3 is considered an IS-3 that lacks the BL-9 gun, doesn't have the top turret and lacks the top engine. Clearly - a Tier 8 vehicle with the score of 8 BW. Fully Elite IS-3 on the other hand, has all the bells and whistles and has BW score of 8.47. I hope you start to see where this is going.

 

With these scores in mind, all we're going to ask from the BW MM, is to make up teams with as close scores as possible and with same types of vehicles as possible. Keeping the limit of 3 arties per side and maybe introducing another one of max 5 TDs per side. I don't think "campy" games are good for business, but that suggestion can also be discussed. Of course there should be a margin of error on the BW scores of each team, as it would probably be impossible to make them absolutely equal and still have low queue times. As to what number should the difference be allowed to grow, I'm not prepared to give an answer to at the moment, so I'll just leave it "free". Obviously, you won't have a team consisting of fully stock Tier 8s vs a team that is made up of fully elite Tier 8s, as their scores would differ way too much and MM would always look around to shuffle the tanks a bit, so that the team numbers are as close as possible.

 

But what about platoons ? Well, if we have a platoon of stock Tier 8s, vs platoons of elite Tier 8s, they obviously can't be on the same team, as this would create a disadvantage for their opponents. So the platoon with the fully elite Tier 8s, is just going to have to be on the team with few more stock Tier 8s to even the playing field. 15 stock vs 15 elites, right ? You can't cut them in equal halves anyways, so one team is always going to have 8 elites and 7 stock and the other team is going to have 7 elites and 8 stocks. Guess where the platoon with the elite Tier 8s is going to go to ? To put it simple, we can maybe add +1 to BW score for a 2-men platoon and +2 to 3-men platoon.

 

As we know, some tanks are "more equal" then others. In a completely Soviet system of socializm and their equality of outcome, we're just going to accept this, without nerfing or buffing tanks, and assign each BW score individually. That's why I took the IS-3 as a starting point - because it's your Joe Average in terms of Tier 8 Heavy performance. But what about a Tiger II ? Well, it's also a Tier 8, obviously, with not as stelar armor, but pretty OK gun, even when stock, so I guess it can also start up with BW score of 8 when fully stock. However, I don't think that it's top gun should add the same score as the IS-3's top gun, as it's simply inferior. Sure, it's more accurate, but the lack of alpha for that class of vehicles is rather annoying, so I'd give it 0.20. Not much of a difference compared to the IS-3 and that would certainly not break the MM if those two tanks happen to be the "odd difference" between the two teams. Is the Tiger II top turret as good as the IS-3's ? Yes and no - more view range, but annoying weakpoint and huge flat side profile. But also AFAIK, it gave boost to the gun's performance, so I would say - 0.16 BW. The engines on the two are comparable upgrades, but because the Tiger II is the heavier vehicle, I'd give it higher score to underline that the Tiger II actually does need that upgrade. So - 0.1. The end score - Elite Tiger II BW is 8.46. Just 0.01 difference with the fully elite IS-3, which kinda makes sense in most cases.

 

But what about the O-Ho(ho-ho-ho) ? Well, first off - this tank doesn't even have a Turret upgrade, so we can factor that into it's base BW level straight away. That way this tank can start up with BW of 8, plus the average score for Tier 8 Heavy Top Turret upgrade. Let's say it's 0.16. That makes the stock O-Ho start off as 8.16 straight away. Of course, we can look at the turrets individually and disregard the fact that the tank doesn't have "a top turret upgrade" and just individually give it's turret score, which will give away how good it is compared to other Tier 8 Heavies upgraded turrets. In that case, the O-Ho might score higher, boosting it's BW score quite a bit. Also, given that the tank's top engine is researchable at Tier 5 (!), it's very likely that everyone would have it by now from their first battle in the vehicle. But let's not assume that and just score it accordingly (NOTE: that engine will have different score on each tank that can use it, because tank weight plays a bit factor on how an engine performs and boosts your mobility). The power-to-weight ratio of the O-Ho is still comparable to the Tiger II, but the tank's top speed and ground resistances let it down, so I'd give that engine a BW score of 0.05. With the guns however, is where things gets interesting. Now, for the sake of testing out, we're going to assume that the 10cm Kai gun is the top gun of that tank. It's an OK gun. Inaccurate, but higher DPM then the Tiger II's, but that tank also has armor, so it can use it at close distance, unlike the Tiger II, so it's basically around the same score as the Tiger II's - 0.20. But what about the 15 cm Derp gun ? Well, we can straight up consider "Derps" as alternative top guns and score them accordingly, because they are actually "always a viable option". Yes, even on the M46 Patton, as proven by Circon. So what about the Derp gun on the O-Ho ? Well, far more inaccurate, but as we established already - that tank can brawl at close range. To top it off, that gun always does damage when it hits a target, plus the potential of module and crew damage. It's also not as useless as the 10cm Kai gun when faced with Tier 10s, so I would actually dare to score that gun as 0.25. So what the score ends up being ? Elite O-Ho with the 10cm Kai gun is 8.41 if we just take the average Tier 8 Heavy Turret score. Which clearly needs to be altered, because there is no way in hell that tank's weight in battle is less then that of the Tiger II or the IS-3. So we'll have to go back and look at that Stock/Elite Turret score again. 200mm armor all around, decent view range ... I'd say maybe a score of 0.30. With those changes, the math shows a score of 8.55 for the 10cm Kai gun and 8.60 with the Derp gun. Sounds about right to me.

 

With those things in mind, we can keep on assigning scores to every single vehicle in the game, balancing them according to their own tiers first and foremost. Then we can look up at how are their handing pressure versus higher tier machines and consider those things into our simple math by adjusting the numbers accordingly. Even if we scrap the equal class distribution, that BW MM is still going to prove superior to WG's MM 2.0. Specially when dealing with tanks that are not Tier 10s. Teams may end up unbalanced, like one team having 2 extra bottom tiers and 2 less top tier vehicles, but the raw numbers of the BW would show that the teams are actually well balanced. And if the extra 2 top tiers on one side help their side wins, is not because the other team had 2 less top tiers, but because their top tiers were not adequate tankers. Imagine a 3-5-7 vs 5-5-5 teams, in a tier 8 game. the 3 is a platoon of O-Ho's vs the enemy's 5, which consists of Caernarvon, Tiger II, WZ-111, Super Pershing and JadgTiger 8.8. I'd say that is perfectly balanced top of the team. You may argue that the numbers will overwhelm the trio, but what you're missing is that the equal 5xTier 7s on each team were also part of the balance equation. In other words, the trio may have some really strong Tier 7s to back them up, while the other team's Tier 7s may include few Stock vehicles to balance the equation out.

 

Maybe as we're trying to balance out tanks in that fashion, we'd have to consider other values and factors as well. Think about set Tier and Class for example. A Tier 8 TD in general overvalues a Tier 8 Heavy, which in terms overvalues a Medium, which in terms overvalues a Light tank, which in terms overvalues an SPG. Differences between Class and Tier may varry thruout the tech tree, as some Mediums and Heavies for example, are much closer to each other in terms of performance (think Tiers 4 & 5 as an example). Some TDs, even when stock, completely overshadow everything else in their Tier (think WT Pz IV) in terms of what influence they can have on the battlefield, while others are utter garbage that doesn't even challenge same Tier Light tanks (think ARL v39 while stock).

 

Obviously, you get the idea that in some vehicles, you might be bottom tier a bit more often then with others. Unfortunately, without proper vehicle balance, some of those OP tanks, would have to "suffer" because of their OP status. As for how exactly are we going to balance out the Premium vehicles in that BW system - it's rather simple really. We keep the PMM status on whatever vehicle has it, and then adjust their numbers according to their performance. So yeah, a Defender might end up having BW score of 8.45 as an example, while an IS-6 may dip to 7.8 BW score, simply because the recent power creep made it feel like a sub-Tier 8 vehicle. Still thou, a bit too OP to be classified as a straight up Tier 7 machine.

 

As you can see, balance is a tricky act. But while the old Battle Weight System was flawed and resulted in utter garbage matchups, those games were extremely rare. The new system is just flawed, because it doesn't make difference between an VK 4502 A and O-Ho. between a Scorpion G and a JagdTiger 8.8, between Primo Victoria and T-34-2. All it cares about is that "similar Classes" face off each other. The 3-5-7 distribution is not that bad, but when you present the teams with completely different set of vehicles on each level, you get those 3 minutes 15-5 games that are extremely frustrating. Even when you're on the winning side, because you rarely have time to rack up the damage you'd know you're capable of dishing out.

 

Also, please note that the numbers I talked about are just some random figures that do not represent an actual study of the tanks mentioned. It's just something of a guideline under which we can all sit down and talk about in a calm manner. And if you have any idea about other numbers we'll need for the equation, please feel free to suggest them.

 

As a friendly note, post it in the pinned MM thread as this, like most of the other MM threads, will be quickly be closed by the mods regardless of how constructive it is.



HQ65 #3 Posted 28 May 2018 - 12:34 PM

    Warrant Officer

  • Player
  • 27876 battles
  • 511
  • [CMERC] CMERC
  • Member since:
    01-16-2012

Wow, you definitley put some thought into this...Hats off to you.

Did you consider giving extra weight to maxed out crews?

 



speedphlux #4 Posted 28 May 2018 - 01:01 PM

    Lieutenant

  • Player
  • 42029 battles
  • 1,681
  • [TZAR] TZAR
  • Member since:
    01-28-2011

View PostHQ65, on 28 May 2018 - 12:34 PM, said:

Wow, you definitley put some thought into this...Hats off to you.

Did you consider giving extra weight to maxed out crews?

 

 

Maxxed out crews are very hard to obtain and even then, I don't think it's worth "punishing" people for loving and playing a single tank for thousands of battles to get a good crew in it. I personally consider every tank crew that I have with 3 skills to be good crew. Minor exception of that rule is from my CW days, when a Bat Chat 25t was the king and it had to have 5 skills on the crew, in order to be considered good or adequate. Maxxed out crew is a luxury.

No, I don't think that crews have to have added weight to the math. If we add that, people will start to ask for Global WR weight as well and then WR per Tank weight too. Will get way too complicated.

Geno1isme #5 Posted 28 May 2018 - 01:04 PM

    Lieutenant General

  • Player
  • 40894 battles
  • 6,751
  • [TRYIT] TRYIT
  • Member since:
    09-03-2013

Don't get too focused on the actual BW computation regarding modules, that's a very tricky thing to get right. Might be better to use the actual module parameters for matching rather than some abstract summary value. Like if there is superheavy armor on one team there should be some high-penetration (and/or derp) guns on the other team. That would automatically classify stock or PMM tanks accordingly, and if modelled properly might also take care of the class balancing issue (for heavies vs. meds at least). Of course getting that right is tricky as well.

 

After all the most common complaint about being low-tier is that the gun and armore is inadequate against +2 enemies.

 

Long before the template nonsense was even considered by WG I suggested some easy changes to the old MM to prevent the extreme brainfarts it sometimes had: http://forum.worldof...1#entry12282331

 

Would require some adjustments due to balancing changes since. Main thing is to get rid of the fixed templates and give the MM some breathing space regarding tier compositions.



arthurwellsley #6 Posted 28 May 2018 - 01:10 PM

    Major

  • Player
  • 50810 battles
  • 2,662
  • [-B-C-] -B-C-
  • Member since:
    05-11-2011

This is a very well thought out proposal.

 

However, have you thought about the server computation power required?

Would it result in the MatchMaker taking longer to make matches? (does this matter if games are fairer? DezGamers poll suggested that 31k players would wait a few seconds longer to keep pref MM).

 

Does WG use Personal Rating in conjunction with the MatchMaker? If not why did they create the metric? Does the MM try and balance the PR of both teams?

 

I agree with HQ65 that crews at 50%, 75% and 100% would have to be added to the calculation.

 

Now a bit of history. On 9th December 2013 WG posted a video of how the MM worked back then. This video is no longer available. The Source: wargaming.livejournal.com/845.html . However, you can read the following about it;

"Hello everyone,

well, for once, this is something noone else seems to have, if I see it correctly :) The link above leads to the new Wargaming LJ blog (in Russian of course). How do I know it’s legit? Because Cannoneer (WG RU developer) sent me a link (so credits go to him).

So, you’ve seen the MM video, right? If you haven’t, you should, because this post refers to it. In order to repeat what was in the video though:

- there is a possibility of the MM to create non-standard teams. The amount of such battles should not exceed 5 percent from the sum of all battles, but in concrete situatios this number can vary in accordance with the theory of probability. For example, if you played 10 battles with IS-3 in a row and in 3 of these battles you encountered a breach of the “first 5 slots” rule (tiers of top 5 vehicles were not equal), it’s not a MM error and it most likely won’t repeat itself in next 10 battles.

 

- the MM tries to align the team composition with the current state of queue, so that in cases, where the team composition is non-standard, the template for selection of tanks into teams can be significantly different from the default template (used when the server is started) (SS: if you remember the video, it is stated there that “weird” battles happen for example shortly after server restart, but in fact before the templates, based on the queue are formed, the server uses a “default” template for team building). Statistics of played battles does influence the default template from the first minutes and as soon as in 30 minutes, the template corresponds to the current composition of the queue and “keeps track” of the changes in it. For example, on the test server the queue might be filled with vehicles of one type the player joing the queue with and it’s possible to notice that there are more and more of these “popular” tanks and their waiting time gets lower and lower.

If the vehicle stands in queue for more than one minute, the matchmaker tries harder to push it into battle, reducing the team composition requirements. But there are 30 vehicles entering the battle (2 teams with 15 players), so even if you wait less than one minute, you can still get into the battle, in which the team is composed in order to push the long waiting player in.

There is one more reason for the appearance of non-standard teams: platoons. If the platoon consists of vehicles of one type (for example three tier 7 heavies), the MM easily picks a team for such a platoon. And if the platoon is a “trollplatoon” consisting of one tier 10 and one tier 2 tank? The MM can correctly handle even this situation, but the composition of both teams will likely be non-standard. And there is one more type of platoons, for which the MM can make non-standard teams: platoons that have artillery or scouts in them. These classes of vehicles have their own MM rules and platoons with them should not break the MM rules.

While collecting the team for the battle, the MM fills empty slots in it while moving gradually from slot 1 to slot 15, skipping those that have already been filled. At the same time it keeps an eye on the queue and from the queue the vehicle, whose MM weight is closest to the one assigned to the empty slot is picked. But even if the MM weight of the vehicle differs from the one of the slot, the slot is filled with it and the MM weights of unoccupied slots are adjusted so that the sum of the team MM weight doesn’t differ by more than 10 percent (20 percent in case there is a player in the queue, waiting for more than three minutes). This explains the cases where the teams have a significantly weaker tank on the last slot of the team: the MM corrected the MM weights of the slots several times and the last unoccupied slot proved to be the hardest to fill.

Slots of both commands are filled in sync, from slot 1 to slot 15. “Rearranging” of 30 already picked vehicles in order to ensure the equality of compositions does not happen, so even if one team has a group of vehicles of the same type and the other doesn’t, when the other MM conditions are met, such teams are sent to battle. When filling the empty slots in teams starting from 6th slot, the MM does not crosscheck the type and tier of the vehicle against the slot with the same number on other team, so it makes no sense to compare “pair to pair” both teams in battle from 6th place onwards. But even in first 5 slots the rule of “equal pairing” can be cancelled, if there are long-waiting vehicles in the queue.

If the amoung of players playing on the server is low or if there is an overflow of one type of vehicles in the queue and the battle waiting time increases, in such a cases following exceptions in MM rules are made:

- there can be battles with more than 5 arty per team (if an arty player waited for more than three minutes for battle)
- teams that are not full can be released into battle (7-14 vehicles), both teams have to have the same amount of vehicles though (this happens when any player’s waiting time is longer than three minutes)
- light tanks with increased MM weight (scouts) can assume the slots for medium and heavy tanks on the top of the team
- a notable disbalance of tech class and tiers might appear within the sum of MM weights of the team (this concerns arty)
- starting from battletier 7, a medium tank, TD or arty can appear on the top position of the team
- if the waiting time is longer than 5 minutes, the tank drops back to the hangar from the queue

All cases written above are however an exception."

 

The above comments come from FTR [note it is banned on WG forums to link to FTR], but if you google FTR and 10th December 2013 and matchmaking you will get it.

Conclusion is that the MM used to do what speedflux is suggesting.



speedphlux #7 Posted 28 May 2018 - 01:49 PM

    Lieutenant

  • Player
  • 42029 battles
  • 1,681
  • [TZAR] TZAR
  • Member since:
    01-28-2011

View PostGeno1isme, on 28 May 2018 - 01:04 PM, said:

Don't get too focused on the actual BW computation regarding modules, that's a very tricky thing to get right. Might be better to use the actual module parameters for matching rather than some abstract summary value. Like if there is superheavy armor on one team there should be some high-penetration (and/or derp) guns on the other team. That would automatically classify stock or PMM tanks accordingly, and if modelled properly might also take care of the class balancing issue (for heavies vs. meds at least). Of course getting that right is tricky as well.

 

After all the most common complaint about being low-tier is that the gun and armore is inadequate against +2 enemies.

 

Long before the template nonsense was even considered by WG I suggested some easy changes to the old MM to prevent the extreme brainfarts it sometimes had: http://forum.worldof...1#entry12282331

 

Would require some adjustments due to balancing changes since. Main thing is to get rid of the fixed templates and give the MM some breathing space regarding tier compositions.

 

It might be tricky to get right, I agree, but since we're seeing time after time failed attempts on WG' side to balance out tanks, it might even be easier then we think. I'm open minded to the idea of assigning BW values according to pure tank's parameters, rather then Elite or Stock status. It was just that the Elite vs Stock was the first thing to pop into my head.

Of course, we should specify Tier, Class and even Roles into the equation. Thanks for that reminder. A Borsig is by no means equal to a Ferdinand, even thou both are Tier 8 TDs in general. That being said, perhaps Roles should be the least of MM computations worry ? Unless too specific, as with the case of Autoloaders for example.

 

The +2 MM complains about guns and armor being inadequate, just leads back to the error that WG made with balancing the game from the Top down, instead of from the Bottom upwards. We can't get around that without proper tank balance, but that is by no means fault of the Matchmaker.

Adding additional rules like your suggestions is not a bad idea. I also had thought about hard-capping the TDs per side, as I often see them as the reason for unenjoyable games. It might complicate the computations for the BW MM to figure out teams, but I don't think many of us would mind to wait a bit in queue just so that we get into a match we can actually enjoy.

 

View Postarthurwellsley, on 28 May 2018 - 01:10 PM, said:

This is a very well thought out proposal.

 

However, have you thought about the server computation power required?

Would it result in the MatchMaker taking longer to make matches? (does this matter if games are fairer? DezGamers poll suggested that 31k players would wait a few seconds longer to keep pref MM).

 

Does WG use Personal Rating in conjunction with the MatchMaker? If not why did they create the metric? Does the MM try and balance the PR of both teams?

 

I agree with HQ65 that crews at 50%, 75% and 100% would have to be added to the calculation.

 

Now a bit of history. On 9th December 2013 WG posted a video of how the MM worked back then. This video is no longer available. The Source: wargaming.livejournal.com/845.html . However, you can read the following about it;

"Hello everyone,

well, for once, this is something noone else seems to have, if I see it correctly :) The link above leads to the new Wargaming LJ blog (in Russian of course). How do I know it’s legit? Because Cannoneer (WG RU developer) sent me a link (so credits go to him).

So, you’ve seen the MM video, right? If you haven’t, you should, because this post refers to it. In order to repeat what was in the video though:

- there is a possibility of the MM to create non-standard teams. The amount of such battles should not exceed 5 percent from the sum of all battles, but in concrete situatios this number can vary in accordance with the theory of probability. For example, if you played 10 battles with IS-3 in a row and in 3 of these battles you encountered a breach of the “first 5 slots” rule (tiers of top 5 vehicles were not equal), it’s not a MM error and it most likely won’t repeat itself in next 10 battles.

 

- the MM tries to align the team composition with the current state of queue, so that in cases, where the team composition is non-standard, the template for selection of tanks into teams can be significantly different from the default template (used when the server is started) (SS: if you remember the video, it is stated there that “weird” battles happen for example shortly after server restart, but in fact before the templates, based on the queue are formed, the server uses a “default” template for team building). Statistics of played battles does influence the default template from the first minutes and as soon as in 30 minutes, the template corresponds to the current composition of the queue and “keeps track” of the changes in it. For example, on the test server the queue might be filled with vehicles of one type the player joing the queue with and it’s possible to notice that there are more and more of these “popular” tanks and their waiting time gets lower and lower.

If the vehicle stands in queue for more than one minute, the matchmaker tries harder to push it into battle, reducing the team composition requirements. But there are 30 vehicles entering the battle (2 teams with 15 players), so even if you wait less than one minute, you can still get into the battle, in which the team is composed in order to push the long waiting player in.

There is one more reason for the appearance of non-standard teams: platoons. If the platoon consists of vehicles of one type (for example three tier 7 heavies), the MM easily picks a team for such a platoon. And if the platoon is a “trollplatoon” consisting of one tier 10 and one tier 2 tank? The MM can correctly handle even this situation, but the composition of both teams will likely be non-standard. And there is one more type of platoons, for which the MM can make non-standard teams: platoons that have artillery or scouts in them. These classes of vehicles have their own MM rules and platoons with them should not break the MM rules.

While collecting the team for the battle, the MM fills empty slots in it while moving gradually from slot 1 to slot 15, skipping those that have already been filled. At the same time it keeps an eye on the queue and from the queue the vehicle, whose MM weight is closest to the one assigned to the empty slot is picked. But even if the MM weight of the vehicle differs from the one of the slot, the slot is filled with it and the MM weights of unoccupied slots are adjusted so that the sum of the team MM weight doesn’t differ by more than 10 percent (20 percent in case there is a player in the queue, waiting for more than three minutes). This explains the cases where the teams have a significantly weaker tank on the last slot of the team: the MM corrected the MM weights of the slots several times and the last unoccupied slot proved to be the hardest to fill.

Slots of both commands are filled in sync, from slot 1 to slot 15. “Rearranging” of 30 already picked vehicles in order to ensure the equality of compositions does not happen, so even if one team has a group of vehicles of the same type and the other doesn’t, when the other MM conditions are met, such teams are sent to battle. When filling the empty slots in teams starting from 6th slot, the MM does not crosscheck the type and tier of the vehicle against the slot with the same number on other team, so it makes no sense to compare “pair to pair” both teams in battle from 6th place onwards. But even in first 5 slots the rule of “equal pairing” can be cancelled, if there are long-waiting vehicles in the queue.

If the amoung of players playing on the server is low or if there is an overflow of one type of vehicles in the queue and the battle waiting time increases, in such a cases following exceptions in MM rules are made:

- there can be battles with more than 5 arty per team (if an arty player waited for more than three minutes for battle)
- teams that are not full can be released into battle (7-14 vehicles), both teams have to have the same amount of vehicles though (this happens when any player’s waiting time is longer than three minutes)
- light tanks with increased MM weight (scouts) can assume the slots for medium and heavy tanks on the top of the team
- a notable disbalance of tech class and tiers might appear within the sum of MM weights of the team (this concerns arty)
- starting from battletier 7, a medium tank, TD or arty can appear on the top position of the team
- if the waiting time is longer than 5 minutes, the tank drops back to the hangar from the queue

All cases written above are however an exception."

 

The above comments come from FTR [note it is banned on WG forums to link to FTR], but if you google FTR and 10th December 2013 and matchmaking you will get it.

Conclusion is that the MM used to do what speedflux is suggesting.

 

Sure, we might have to wait a bit for matches. By how much depends on how many rules we apply to the BW MM I assume. At the moment, I honestly haven't had to wait for more then 40 seconds in queue when the server is least populated and while trying to play an odd Tier vehicle. I don't think that this is going to result in much higher then 10-15% longer queue times.

Since WG stated more then once that Skill MM is out of the question, I highly doubt that their Rating system has anything to do with the MM. Just install XVM and play 10 games and you will see that it has absolutely no influence on the teams you're getting into or the matchups that you're getting into. Why did they introduced it then ? Because WNx is flawed. Always has been, always will be, and I guess WG wanted us to have an alternative values for our e-peens, because let's face it - having a larger e-peen is very very important to some people.

As I've said already, if we start to factor in external stuff like "crew proficiency" and others, it would not only make the MM queues longer, but also start to factor in other issues, like Global WR and per-tank WR and so on. Bad business IMHO. As I've said, I'm not looking to "balance" things out - no changes to vehicles, not taking into account crew skills, ammo loadouts, consumables, permanent camos and what not. This is just a base "scoring system" to match up our tanks only. Just the tanks, not the crews, not the players, not the ammo loadouts. That way I believe that the players would be the ones having the power to influence the battles, despite their tanks' Battle Weight values.

http://wiki.wargamin...Matchmaker_(WoT)

You can see in here, they even left out some info on what MM was doing pre-9.18. Do you had any issues with the queue times back then, when supposedly MM system was so heavy and complex ? Because I don't remember having to wait much. Post-9.18, rules are less, there is almost no complexity whatsoever and we're all unhappy about the matches that we have to play.



Aikl #8 Posted 28 May 2018 - 02:00 PM

    Brigadier

  • Player
  • 25150 battles
  • 4,200
  • [-MM] -MM
  • Member since:
    04-13-2011

Might comment properly when I have time to read, but in the interest of staying on-topic while I reply to Anxter Jabster:

This sounds pretty good. Sort of like a dynamic version of the battle rating system War Thunder uses - a game where matchmaking is arguably far less harsh than WoT.

Worth noting they're still suffering from tier compression - it's not detailed enough to work 100%. For instance, Tiger E and Tiger H1 are equals - despite the former having access to (largely useless, but sometimes important) APCR ammo, a better cupola/mantlet (?) design and added armor on the front. IS-2 and Tiger are equals too - simply because the next step up is too much for the IS-2, and one step down is way too much for the Tiger.

 

Not sure if it would be perfect - and an apparent design goal of WoT is that it's supposed to look fair, not necessarily be fair (I'll find that quote and add it to the proper reply later). Also, +2 is really important to Wargaming. Because clubbing and getting clubbed is fun, apparently.

 

(APCR in WT has little post-pen damage; hence why it's considered 'useless' - but strongly preferable to not actually penning in the first place.)

 

View Postjabster, on 28 May 2018 - 11:30 AM, said:

 

As a friendly note, post it in the pinned MM thread as this, like most of the other MM threads, will be quickly be closed by the mods regardless of how constructive it is.

 

​One could argue that we're not discussing the matchmaker, but rather a proposed matchmaker. 

Edited by Aikl, 28 May 2018 - 03:08 PM.


jabster #9 Posted 28 May 2018 - 02:45 PM

    Field Marshal

  • Beta Tester
  • 12516 battles
  • 22,149
  • [WSAT] WSAT
  • Member since:
    12-30-2010

View PostAikl, on 28 May 2018 - 01:00 PM, said:

Might comment properly when I have time to read, but in the interest of staying on-topic while I reply to Anxter:

This sounds pretty good. Sort of like a dynamic version of the battle rating system War Thunder uses - a game where matchmaking is arguably far less harsh than WoT.

Worth noting they're still suffering from tier compression - it's not detailed enough to work 100%. For instance, Tiger E and Tiger H1 are equals - despite the former having access to (largely useless, but sometimes important) APCR ammo, a better cupola/mantlet (?) design and added armor on the front. IS-2 and Tiger are equals too - simply because the next step up is too much for the IS-2, and one step down is way too much for the Tiger.

 

Not sure if it would be perfect - and an apparent design goal of WoT is that it's supposed to look fair, not necessarily be fair (I'll find that quote and add it to the proper reply later). Also, +2 is really important to Wargaming. Because clubbing and getting clubbed is fun, apparently.

 

(APCR in WT has little post-pen damage; hence why it's considered 'useless' - but strongly preferable to not actually penning in the first place.)

 

 

One could argue that we're not discussing the matchmaker, but rather a proposed matchmaker

 

It would seem a shame to have a useful thread locked.



speedphlux #10 Posted 28 May 2018 - 02:55 PM

    Lieutenant

  • Player
  • 42029 battles
  • 1,681
  • [TZAR] TZAR
  • Member since:
    01-28-2011

View Postjabster, on 28 May 2018 - 02:45 PM, said:

 

It would seem a shame to have a useful thread locked.

 

Sure, but if this tread gets locked, we'd know why ;)
This forum section used to be called "General Discussions" and I still have it bookmarked with that name, hence why I posted this here, for the community to discuss.

jabster #11 Posted 28 May 2018 - 03:24 PM

    Field Marshal

  • Beta Tester
  • 12516 battles
  • 22,149
  • [WSAT] WSAT
  • Member since:
    12-30-2010

View Postspeedphlux, on 28 May 2018 - 01:55 PM, said:

 

Sure, but if this tread gets locked, we'd know why ;)
This forum section used to be called "General Discussions" and I still have it bookmarked with that name, hence why I posted this here, for the community to discuss.

 

The reason being that there’s a pinned MM thread. It sounds a bit harsh but that’s the way it works.

Balc0ra #12 Posted 28 May 2018 - 04:16 PM

    Field Marshal

  • Player
  • 63409 battles
  • 14,817
  • [WALL] WALL
  • Member since:
    07-10-2012

Stock grinds should always have a "note" in the MM IMO. As in even in a +2 game. There should be at least one enemy stock tier 8 HT on their team too. Issue is that it's not something that can be a 100% sure thing. But a factor to make it more even. But then again there are stock grinds with no turret upgrade etc. So a Stock Black Prince, still has the same armor value as an elited BP. But same can't be said for a Stock T29 vs Elite etc. 

 

As for classes? They do take clip guns into consideration now vs classes. But not armor or even derp etc. HT's should be divided into two classes. HT's and Super heavies. As seeing the enemy with a 3 man Type 4 platoon on top, and you have a two man 50 120  platoon on top with a T-10.. it's not ideal. I wish MM did take that into consideration at times. As it don't differ between support HT like the 50 120 or VK A, or a brawling HT like the 257 or VK B. Or a Super Heavy like the Type 4 etc. As that with 3-5-7 anyway can make a major difference on the HT line. As the 50 120's usually don't go the HT like when there is a Type 4 or two on the enemy team. And your Tiger II and IS-3 has to deal with it. 



speedphlux #13 Posted 28 May 2018 - 05:47 PM

    Lieutenant

  • Player
  • 42029 battles
  • 1,681
  • [TZAR] TZAR
  • Member since:
    01-28-2011
I understand Jabster, but the point of having separate tread is to discuss it freely, without others bumping into old ideas and whatnot. If WG closes it and ignores it, it's on them ... I believe we're only trying to help here.

Balcora, I did had an example of the O-Ho not having "Elite Turret" available, so it's similar case to your Black Prince. I just basically had "evaluated" the O-Ho's turret as Elite and added it to the Battle Weight Tier value. All we need to do, is to simply compare the BP's turret to the turrets of other Tier 7 Heavies and we can get a proper number from there. You know - the usual armor profile, view range, gun mantlet, gun depression, does a turret improve the stats of the top gun or not and so on. And yeah, we did talked a little bit about not only Class separation, but also Roles separation. You can't compare an Object 416 to T-44 in the Roles department, but as far as MM system is concerned, those are actually differences that can be overlooked in order to speed up the queue times. I do agree that "Super Heavies" and "Autoloaders" are Roles which the MM should take a bit more serious and actually try to matchup accordingly, even if that takes longer.




1 user(s) are reading this topic

0 members, 0 guests, 1 anonymous users