**Mastery Loadouts**
Due to issues related to the release of Mastery Loadouts, the "free swap" period will be extended.
The new end date will be May 1st.
Due to issues related to the release of Mastery Loadouts, the "free swap" period will be extended.
The new end date will be May 1st.
We need to have a serious conversation AW manipulation
This discussion has been closed.
Comments
You missing a point here, they play in tier 2/3/4 to get easy match ups and these tier rewards have plenty of 5*/4* shards and they get these without even using items
Isn't the point of war to test your metal against a similar strength opponent? Is it really more rewarding to accumulate more rewards to get more champs to then just use them against weaker opponents? Idk just sounds like cowardice to me
It's not okay to bog down the forums with more of this excremental thread nonsense.
I understand the mathematics behind war rating just fine. You get more war rating for beating a better alliance, you get less for beating a less capable alliance.
But you're missing the point. The point is that there are alliances who are manipulating the system so they never have difficult matches. This comes at the expense of weaker alliances who are "going to war" in good faith.
Also I'm in an alliance that's a month old and we are 6.8 mill, we all have to start from the bottom
Takes like a month just to get to tier 9, totally not worth it
This would make the most sense for AW, taking an alliances overall prestige as the base, by-passes those people selling off 2-3*, it also ensures higher ranked alliances can't match up with a bunch of 4* 3/30 guys.
The problem with this is that it's a victimless crime. A lower ally (such as the ones that are the "victims" that are mentioned through this thread) may see one really good ally each month or longer that is gaming the system. They take the loss and move down in rating, but it does not affect their tier or war rating in any prolonged meaningful way. The likelihood is high they would lose to an equal ally within that week. At most, it changed the pattern that their win cycle would have occurred naturally.
This is nothing like ally swapping to get weekly cats with minimum effort before. The "victim" ally wins the next one and is in the same spot as before, then they can win the next match if they are good enough. So if their record likely would have been WWL, now it has changed to LWW. Same spot, same rewards, no big deal.
I would also think there are other possibilities of "seasons" where everyone is reset, this though would require a restructuring of reward payouts and could be quite complicated.
This is all predicated on the notion that there is actually a large number of alliances manipulating the system which hasn't been shown yet.
The second alliance won't be at 0 but at a lower point in the upper tiers. So, they will revive the same rewards for easier matches. Without naming names, the alliance(s) beginning this new swapping have established sister alliances in place already.
There are many "dead" alliances that disbanded at Tier 2-4 (war rating anything from 1400-1600).
This Tier 1 alliances with above 2000 war rating shift their whole roster to the dead alliance and started playing tier 2-4 wars against weaker opponents. How they get hold of an dead alliance is not rocket science.
And the alliances taking this up are notorious for previous exploits, so, expect it to work well and take off fast. Shards for up and comers will soon drop off.
If they swap to tier 2 or 3 and run 3 wars a week they will be back in tier 1 in 2 weeks maybe 3. They already gave up 1 aq rank rewards and then they are going to swap back and miss aq rank rewards 2 to 3 weeks later? It's not rocket science...that's asinine.
This was my original idea. New Season, clean slate. Each would give the possibility of new Rewards (Events, Titles, etc.). Each Season the Rating is cleared. The idea is that each Ally that participates would grow a bit more each Season. There could also be a Placement Phase that operates based on Ally,/Average Rating, so there is less chance of surprise extreme Matches. Then it would lean on War Rating later on.
A few points of clarification.
1) It is not your top alliances, if I had to guess it is your alliances outside the top 100 that still place in tier 1 or tier 2 of war. Probably covers somewhere between alliances 100-500 (just a pure guess though).
2) The alliances that do this are usually a part of a network of alliances. And there is a swap (not a change to a brand new alliance). The swap usually puts them back into tier 4 for war but allows them to stay at the expert tier for AQ, but they miss out on a week of rewards.
3) Tier 4 war ratings are somewhere within the 1400-1600 range, so it takes more than 1-3 weeks to be back in tier 1 for war.
Def a top alliance and part of a network. They were one of the first to start the original swap and also originated other game exploits.
I have thought about it for a while now, and I haven't yet figured out a way to construct a system that incorporates this kind of information that isn't gameable yet. I actually believe ranking based on anything other than win/loss record is always going to allow for crazy loopholes and exploits, because win/loss record minimizes the computational "fairness" of the matchups globally. The problem is it can, in some implementations, create bad ones locally. Meaning you may find yourself in a wildly unfair matchup today operating against you, but the system will balance out by eventually matching you with a wildly unfair matchup in your favor, so the net win/loss error is about zero.
Still, it is worth seeing if those can be dampened. I just think it isn't simple to do so. I think the best way to explicitly tackle swaps might be to target them directly, and not indirectly with different matchup logic. For example, starting a brand new alliance at zero rating is technically a way to represent zero knowledge: we don't have any wins or losses recorded yet, so we cannot compute a rating at all. If there was a way to better estimate rating for a new alliance based on membership in the absence of a win/loss record, that might ameliorate the problem somewhat. It wouldn't solve it, because I don't think you can "know" what that rating should be, but close enough might have to be good enough.
One idea I was toying with, but have not analyzed enough to know if it is mathematically sound, is to consider a points system where every member of an alliance got their own war rating. The crux of the idea is say your alliance has a war rating of 1500. Every member contributed to that rating, and we could use the battlegroup percentage as a rough proxy for that contribution. Let's say your overall percentage was 4%. We take war rating, and multiply by your percentage and get 1500 * 0.04 = 60. Now, if a new alliance is formed, then on its very first war its initial rating would be the sum of the ratings of its members at that time. If you had thirty members all with a 60 rating, the initial alliance rating would be 60 * 30 = 1800 (you were higher than average). If the same thirty people try to form a new alliance, they would end up with the same initial rating: 1500. We then use win/loss record moving forward to modify that rating. And we use moving weighted averages to come up with a way to continuously monitor individual player rating. The individual rating would only be used when new alliances were formed: they would have no impact on an ongoing alliance.
There are some weird corner case ways to exploit this kind of system, but it is harder to do. But it would take some time to determine what the best possible implementation of this kind of system would be, and whether it is ultimately sound.
This actually subtly hints at an unresolvable problem with AW. This statement encapsulates a definition of what a fair contest would be that is fundamentally and irreconcilably different from the definition many other people are implicitly using when they talk about fair and unfair matchups. You seem to be saying that the strongest alliances end up being matched against each other and so can only win half the time, when the strongest alliances should win more than half the time. That pivots around a value judgment: should the strongest war alliances win more often than the weaker ones on average, or should the strongest alliances be matched against each other in "even" contests which by definition they will only win about half the time?
I'm of the opinion that AQ incorporates a stronger meritocratic reward system, and AW more of an even odds match up system (granted: the rewards get higher at higher tiers, but to a significantly muted extent relative to AQ) and that's intentional. But that design decision is largely an arbitrary one.