Why matchmaking needs a fix sooner rather than later

[QUOTE=Millerman78;9341856]

Quote:

---

Originally Posted by cnaf

I doubt that. IMO, the overall majority of them will either complain about the matchups, or simply put the game down and move on to another one.

I don't think SuperCell wants that.I believe one of their big focuses is trying to help the casual player, the unwitting rusher as some will say. Daily bonuses, Healer tweaks, that term balance that SC likes to use- I don't believe balance to them means balancing a Town Hall offense to it's defense. I believe it's meant in relationship between a bad player and a good one. Recent war performance has been added to the MM algorithm. Why? So bad clans have a better chance at matching with bad clans. Somebody gets a bone there. These and many other tweaks have been implemented to help the casual or "bad" player. Poison spell. EQs. Convenient, eh? And as if there was even more of a need to assist... we just received an even larger radius for the poison. Who do you think really needed that?

I'm starting to think there must be a very large segment of players that have bases with progression levels that appear similar to lopsiders- in one way or another. Something that is keeping SC from tweaking the MM in a way that would squash lopsided bases in pairings... that somehow anything they've come up with thus far- would severely affect the unknowing, casual player as well.

---

This ^. The casual player dilemma.

Clash of Clans is a 'pick up and play' mobile/tablet game. This 'pick up and play' market is the biggest volume and growing market in the world.
The vast majority of players probably don't know what 'PBT' is, let alone ever reach it.

The 'Casual players' feed the hardcore pushers,farmers,war guys. They are the backbone of the game.

@skrags and cnaf.. well put. In terms of the rushers, I like to still say lesser of two evils, sort of like terrorists blending in with refugees (this came to my mind no offense). And again, rushed th8's and even th9's now are not determining the outcome of a war. They need to tread carefully with th10 and of course th11. So however they work out the computations with the MM, which is not my strength, they need to account for it. Or take into account the town hall, army camps, and particular troops that were upgraded and act accordingly. Also, people when they play this game tend to either really take to it and go all in, or not into it. So like i said about the rushers, if they really love the game and start taking to it, they will adapt IMO.

Quote:

---

Originally Posted by cnaf

I doubt that. IMO, the overall majority of them will either complain about the matchups, or simply put the game down and move on to another one.

I don't think SuperCell wants that. I believe one of their big focuses is trying to help the casual player, the unwitting rusher as some will say. Daily bonuses, Healer tweaks, that term balance that SC likes to use- I don't believe balance to them means balancing a Town Hall offense to it's defense. I believe it's meant in relationship between a bad player and a good one. Recent war performance has been added to the MM algorithm. Why? So bad clans have a better chance at matching with bad clans. Somebody gets a bone there. These and many other tweaks have been implemented to help the casual or "bad" player. Poison spell. EQs. Convenient, eh? And as if there was even more of a need to assist... we just received an even larger radius for the poison. Who do you think really needed that?

I'm starting to think there must be a very large segment of players that have bases with progression levels that appear similar to lopsiders- in one way or another. Something that is keeping SC from tweaking the MM in a way that would squash lopsided bases in pairings... that somehow anything they've come up with thus far- would severely affect the unknowing, casual player as well.

---

Quote:

---

Originally Posted by Skrags

This ^. The casual player dilemma.

Clash of Clans is a 'pick up and play' mobile/tablet game. This 'pick up and play' market is the biggest volume and growing market in the world.
The vast majority of players probably don't know what 'PBT' is, let alone ever reach it.

The 'Casual players' feed the hardcore pushers,farmers,war guys. They are the backbone of the game.

---

^these posts are one of the reason why I initially proposed my "weight tax" idea. The casual/premmie players who rush their TH without upgrading their troops, heroes, walls, and point defenses just to get the flashy new lvl1 inferno tower at TH10 don't really have highly disparate accounts (and if anything, their defense outweighs their offense) and so a weight tax measuring (offense > defense) disparity wouldn't apply to their bases and matchmaking would treat them exactly as it currently does. Which, they are currently already at quite the disadvantage, but at least my proposed "solution" doesn't hurt them any further.

The only people a weight tax would negatively effect would be highly disparate engineered accounts. Defenseless TH9+ and TH10+ minimax accounts with TH7 level defenses and no xbows or infernos, etc. And when I suggested it, I was explicitly careful to note that the tax shouldn't put them at an artificial weight ABOVE a maxed TH equivalent, I don't want to "punish" these types of accounts, I just wanted to bring them in-lane a bit more so that their advantage is lessened.

I said at the time, and I still believe that:

Quote:

---

Originally Posted by nerfedname

Thanks. My idea is not about "punishing" engineered clans or "giving maxers the advantage instead," it's about crafting ways to provide fair and equitable matchups to ALL clans, regardless of their clan composition, TH levels, offense/defense progression, maxer/rusher, etc., etc.

I want a system where EVERY clan, even one full of premmie bases, can perform a war search and have some reasonable expectation of getting a match where, if their attacking skill is good enough, they have a mathematical chance of winning.

---

THAT is what I would be trying to achieve, and what I think Supercell should be trying to achieve. Fair and equitable war matchups FOR ALL!

Edit: And yes I'm fully aware that a "weight tax" wouldn't solve all problems, or do anything whatsoever to fix roster engineering and the like, but it would be a good 1st step imo and hence why I proposed it.

Quote:

---

Originally Posted by Johnnyfsullivan

Starting in March 2016 our encounters with engineered clans increased in frequency, so much so that we started engineering in May / June and reaped the benefits from July to September. The October tweak seems to have mitigated the impact of engineering based on my experience. Our bases had a good foundation so it wasn't too hard to "quickly" get back to balanced bases.

---

Same for us, apart from the started engineering. We had people going 9.5 around Aug/Sept, but that was in most cases because they'd maxed TH9.
The MM didn't seem to improve any in March - it was fine before for us; but went downhill steadily as people developed lopsided accounts to benefit from the free offence. By October I was quite close to quitting as we were getting so many pointless wars. Several against hyper-engineered Chinese clans. When your clan is TH10 and down and they have 1+ supermax TH11s plus multiple other fully maxed TH11 offence bases, it's just a waste of time.

Quote:

---

Originally Posted by daddylindo

A-B-S-O-L-U-T-E-L-Y. Prep day is USELESS now. We can't scout enemy base. We dont need 23 hours to fill CC. Let the war search take 6 hours and then take the 6 hours off of prep day.

---

There have been other, similar suggestions to improve the MM pool. More clans searching does mean better matches. The one I like is, instead of lengthening search, to set fixed start times. Something like 0400, 0800, 1200, 1600, 2000 and 2400 GMT. You sign up before the start time, just like starting search now. And at that start time the server organises all the clans into the best combination of matches. This would significantly increase the search pool: if current searches tend to be 30 minutes or so on average, you'd instead have 4 hours worth of clans wanting to start - could be 5 to 10 times more clans.

Quote:

---

Originally Posted by daddylindo

The rushers will adapt. And anyway, the people who rushed to th8 and th9 are not hurting the wars anymore. If you rushed to th10, well, yeah you should be a detriment to your clan unless you can 3 star enemy th10

---

I doubt they'll adapt; and as others have said I don't agree with giving maxers an advantage as such. That's where we were in 2015, when war clans had perma-9s etc because they were an advantage. It seemed like SC was very keen to nerf that - from their notes it sounded like one of the motivations behind the failed March MM changes.

Quote:

---

Originally Posted by cnaf

I'm starting to think there must be a very large segment of players that have bases with progression levels that appear similar to lopsiders- in one way or another. Something that is keeping SC from tweaking the MM in a way that would squash lopsided bases in pairings... that somehow anything they've come up with thus far- would severely affect the unknowing, casual player as well.

---

I'd be cautious working backwards from what SC did to what motivates them. They seem a particularly inscrutable company, and there's plenty of things they've done that just baffle me (eg let's ignore offence in MM!)
But I'd agree there's plenty of bad bases out there - if you look at the bases you see when nexting and think which ones you might want in war with you, the answer is usually not many.

Quote:

---

Originally Posted by OnyxDS

Same for us, apart from the started engineering. We had people going 9.5 around Aug/Sept, but that was in most cases because they'd maxed TH9.
The MM didn't seem to improve any in March - it was fine before for us; but went downhill steadily as people developed lopsided accounts to benefit from the free offence. By October I was quite close to quitting as we were getting so many pointless wars. Several against hyper-engineered Chinese clans. When your clan is TH10 and down and they have 1+ supermax TH11s plus multiple other fully maxed TH11 offence bases, it's just a waste of time.



There have been other, similar suggestions to improve the MM pool. More clans searching does mean better matches. The one I like is, instead of lengthening search, to set fixed start times. Something like 0400, 0800, 1200, 1600, 2000 and 2400 GMT. You sign up before the start time, just like starting search now. And at that start time the server organises all the clans into the best combination of matches. This would significantly increase the search pool: if current searches tend to be 30 minutes or so on average, you'd instead have 4 hours worth of clans wanting to start - could be 5 to 10 times more clans.



I doubt they'll adapt; and as others have said I don't agree with giving maxers an advantage as such. That's where we were in 2015, when war clans had perma-9s etc because they were an advantage. It seemed like SC was very keen to nerf that - from their notes it sounded like one of the motivations behind the failed March MM changes.



I'd be cautious working backwards from what SC did to what motivates them. They seem a particularly inscrutable company, and there's plenty of things they've done that just baffle me (eg let's ignore offence in MM!)
But I'd agree there's plenty of bad bases out there - if you look at the bases you see when nexting and think which ones you might want in war with you, the answer is usually not many.

---

I love the fixed start time setup idea. Now if they sprinkle start time clan war tournaments on top, now we're cooking with grease!!

A fixed start time could actually be incredibly good for many clans.

Perhaps they could have a feature where a coleader or a leader locks in a lineup and at a specific time, the match is started. They could also tweak it for timezones and such.

Quote:

---

Originally Posted by ExclusiveInvicta

A fixed start time could actually be incredibly good for many clans.

Perhaps they could have a feature where a coleader or a leader locks in a lineup and at a specific time, the match is started. They could also tweak it for timezones and such.

---

or

modders return to their TWC tournaments elites or whatever and stop ruinind the fun for regular clans
cause you fail a lot without TPAs

I never knew matching was broken. It appears just fine for me and the clan I'm referring to doesn't engineer.

Quote:

---

Originally Posted by nerfedname

Edit: And yes I'm fully aware that a "weight tax" wouldn't solve all problems, or do anything whatsoever to fix roster engineering and the like, but it would be a good 1st step imo and hence why I proposed it.

---

My query about the weight tax now is whether there exists any problem except roster engineering. When you first suggested it, my line was it was the best of the suggestions on here, but I was still opposed to it. At that time the problem was that offence was ignored. Ignoring offence and then adding a tax if a clan had too much offence seemed a corkscrewy way of dealing with it - better to count offence properly before adding other complications. SC clearly improved the weighting of offence in October.

What we see now is a lot of complaints along the lines of "they have all these TH11s with 1 cannon defence and the war is unwinnable". My question is though - is the problem caused by the TH11 offence, or the 1 cannon defence? And from what I've seen it's mostly the zero defence. Every genuine bad match I've been in or visited since October, the enemy has had a substantial number of zero defence bases. But they could be either defenceless TH11s, or TH3s. It seems to make little difference.
Conversely my clan did have one match against a very lopsided clan without defenceless: their offence was a TH higher everywhere, often 2. Several TH9.11s at the top (we had 9.5s, 9s and 8s). The war ended in a predictable double-perfect draw - I really can't see the point of their line-up. But my point is it didn't feel that they had an advantage from those 9.11s and 8.10s. In fact, quite the opposite: we tripled IIRC all their top 5 bases twice. We had a perfect score with none of our best 2 or 3 guys attacking. Yet the enemy only finished clearing us on their last viable attack - it actually looked for a long time as if we might win. So it felt that their lopsidedness if anything made them weaker than us, not stronger. But it's an odd case - I've not seen many very lopsided opponents without (semi) defenceless bases.

It's the lopsidedness of defenceless bases that's visible, and that's what people complain about. But I'm not convinced it's the cause of the problem. So my feeling about the weight tax remains much the same: it's better than the other ideas; it wouldn't cause any harm, and would do some good. But better would be to fix the actual underlying problem first. Then assess whether there remain edge cases that need extra checks and taxes.

Quote:

---

Originally Posted by OnyxDS

My query about the weight tax now is whether there exists any problem except roster engineering. When you first suggested it, my line was it was the best of the suggestions on here, but I was still opposed to it. At that time the problem was that offence was ignored. Ignoring offence and then adding a tax if a clan had too much offence seemed a corkscrewy way of dealing with it - better to count offence properly before adding other complications. SC clearly improved the weighting of offence in October.

What we see now is a lot of complaints along the lines of "they have all these TH11s with 1 cannon defence and the war is unwinnable". My question is though - is the problem caused by the TH11 offence, or the 1 cannon defence? And from what I've seen it's mostly the zero defence. Every genuine bad match I've been in or visited since October, the enemy has had a substantial number of zero defence bases. But they could be either defenceless TH11s, or TH3s. It seems to make little difference.
Conversely my clan did have one match against a very lopsided clan without defenceless: their offence was a TH higher everywhere, often 2. Several TH9.11s at the top (we had 9.5s, 9s and 8s). The war ended in a predictable double-perfect draw - I really can't see the point of their line-up. But my point is it didn't feel that they had an advantage from those 9.11s and 8.10s. In fact, quite the opposite: we tripled IIRC all their top 5 bases twice. We had a perfect score with none of our best 2 or 3 guys attacking. Yet the enemy only finished clearing us on their last viable attack - it actually looked for a long time as if we might win. So it felt that their lopsidedness if anything made them weaker than us, not stronger. But it's an odd case - I've not seen many very lopsided opponents without (semi) defenceless bases.

It's the lopsidedness of defenceless bases that's visible, and that's what people complain about. But I'm not convinced it's the cause of the problem. So my feeling about the weight tax remains much the same: it's better than the other ideas; it wouldn't cause any harm, and would do some good. But better would be to fix the actual underlying problem first. Then assess whether there remain edge cases that need extra checks and taxes.

---

I've always been a supporter of the weight tax proposal, but you're right. The issues it was proposed to fix have been, largely, negated. At least in my experience.

Roster engineering on the other hand is still an issue, and I like you am uncertain if it's offense or the defense that is the issue.

It is certainly a difficult situation. How does a computer program determine the difference between a low weight base at the bottom that is put there so the player could be in war, and the one that was put there to intentionally attempt to skew MM. At some point the program has to make a choice and that choice will either give an advantage to both clans or punish both clans.