Gold 1 bracket is broken
Haji_Saab
Member Posts: 5,837 ★★★★★
https://www.reddit.com/r/ContestOfChampions/comments/89vae9/analysis_of_aw_season_1_tier_vs_bracket/
Please have a look at the data collected by someone on the tiers. Lets look at Gold 1 and Platinum 3 brackets.
You only need to be in tier 7 (or tier 6) in order to gain Gold 1 rewards. And you need to be in tier 3 to gain platinum 3. From that, you can clearly see that the alliances in tier 4 are totally shafted. We're playing at a much higher level than tier 7 but sharing the same rewards. Most players can see that and that is why there was so much upheaval in gold top 150 alliances because all the motivated and ambitious people left them looking for platinum 3 alliances.
At the end of the season, we came across quite a few alliances who were totally demotivated because they were in the same position. What's the point of striving in the AW seasons when you can lose 15 wars and still stay in the same bracket for rewards?
I see that platinum 3 is a jump up from gold 1 in terms of difficulty and they need more rewards but equating tier 4 with tier 7 doesn't seem right to me. Hopefully you will consider it for season 2.
Please have a look at the data collected by someone on the tiers. Lets look at Gold 1 and Platinum 3 brackets.
You only need to be in tier 7 (or tier 6) in order to gain Gold 1 rewards. And you need to be in tier 3 to gain platinum 3. From that, you can clearly see that the alliances in tier 4 are totally shafted. We're playing at a much higher level than tier 7 but sharing the same rewards. Most players can see that and that is why there was so much upheaval in gold top 150 alliances because all the motivated and ambitious people left them looking for platinum 3 alliances.
At the end of the season, we came across quite a few alliances who were totally demotivated because they were in the same position. What's the point of striving in the AW seasons when you can lose 15 wars and still stay in the same bracket for rewards?
I see that platinum 3 is a jump up from gold 1 in terms of difficulty and they need more rewards but equating tier 4 with tier 7 doesn't seem right to me. Hopefully you will consider it for season 2.
12
Comments
From what I understand, DNA had a (wonderful) breakdown of the math behind what is required to get where. This post kind of looks at it retroactively as to what exactly happened. Two different approaches, both have their arguments I can understand. But the main point behind this post is saying (for the majority) alliances in tier 4 and tier 7 received the same rewards, and there should be more of a division there. I don't think the rewards were BAD per se, but I wouldn't oppose an extra reward bracket
Edit: Didn't realize it was DNA. I guess this is just posed as a "Gold 1 is too big" argument again.
No worries. I found your post to be insightful and I have to agree with your suggestion. My point was more meant for @GroundedWisdom since before I discount your opinion (since you are Platinum 3) I would like to know what tier and what season rewards bracket he ended up in.
You just got great rewards for AW, be happy you got anything.
Well, Tier 4 is the top 2-3% of Alliances and Tier 7 is the top 7-9%. It kinda seems silly to me that there's so many rewards brackets ahead of the top 2-3% if, in fact, the majority of Tier 4 Alliances wound up in Gold 1. If that is actually the case, then they probably need to broaden the Platinum Brackets to include more Tier 4 allies.
Thing is, DNA's study may be good enough to set goals by, but it's by no means a replacement for hard statistical analysis. It relies on several unrealistic assumptions, like an Alliance will win exactly 50% of their wars and never win enough in a row to temporarily bump up a tier or lose enough to temporarily drop down a tier
http://forums.playcontestofchampions.com/en/discussion/60672/analysis-of-aw-season-1-tier-vs-bracket#latest
That should be the right link
DNA's assumptions are actually fairly accepted statistical norms, barring manipulation of the system. There are always outliers, but thats why we call it an average.
I guess the weird thing about being in the gold 1 bracket is you are aiming to get into a pool of 300 people in platnum. and even then, because their rewards vary so greatly, you're just trying to get into a pool of 200 in platnum 3. The next few pools for rewards are 1,500 alliances each. This is a large jump of alliances/players to a much smaller group of rewards. The image becomes one of "1% has 75% of the wealth" with a huge jump to the next group, that outnumbers them.
We were in tier 3 from the 3rd AW until the end of season and still finished in Gold 1 (4th). We had almost twice as many points as the bottom ranked team that it showed in Gold 1 rankings (700 I believe). It's really stupid that we got the same rewards as someone almost 900 spots lower yet had to play on the top map for basically the entire season. We could have taken off the last 2 weeks and still got the same rewards we finished with. That makes no sense and should be fixed.
There's no way these are statistical norms. Look at Silver 1, for example. There's no Alliance out there that averaged 165k per war in Tier 18 the entire season while winning 50% of their wars. Maybe in DNA's tier, those are statistical norms, but they don't hold true up and down the board. A better exercise if you didn't want to take the time to examine the allies in each tier and come up with a true average would be to take the cut line for each tier and make a table for each showing how many points/war you need to are the cut based on your War Tier. Even that would still be only nominally useful since the vast majority of the player base are in tiers where 1-2 wins or losses is enough to move you up or down a tier.
I agree that the rewards brackets can stand some retooling, size wise.
I don't see why they wouldn't average 165k per war if I'm being honest. This math is ignoring starting from scratch, or losing members, or drastically changing the makeup of your alliance, mainly because its all impossible to plan on. So if an already-established alliance didnt get better, didnt get worse, didnt add players, didnt lose players, 165k is what you would see at any tier. We're not including multipliers in this.
You're technically correct that those assumptions are unrealistic, but they are deliberately unrealistic in a specific way that I mention. The analysis attempted to determine the minimum average multiplier you would need to have a realistic chance of reaching a particular bracket. It does not attempt to predict what bracket you will end up in. To determine the minimum multiplier necessary, the assumptions lean in the direction of presuming the maximum realistic points an alliance can usually expect to score.
If you're assuming the alliance is doing as well as possible, then you don't assume they will win many wars in a row and climb in multiplier, because the average multiplier in that situation will be lower than if you started in that tier to begin with. And the assumption that best matches the assumption that an alliance started in, and then stayed in, the highest possible tier within their capabilities is the assumption that they win about 50% of the time. Significantly more often and they would climb tier, significantly less often and they would drop tier.
No alliance does this all the time, of course. But any alliance that doesn't do this is scoring less points than this hypothetical model, and would thus require a higher multiplier to reach the same score. As I'm looking for the minimum realistically possible multiplier, I don't account for situations with lower scoring output.
Anyone using my analysis to say something like "I was in tier 6 so I should be in Gold 1" is using it incorrectly. The analysis says you could reach Gold 1, but you aren't guaranteed to do so. You could do much less. But the analysis says you are extremely unlikely to do more.
The thought occurred to me to do a statistical analysis to try to answer the question "if you averaged tier X, what was the most likely bracket you ended up in" but I don't think the data exists (available to us) to perform that kind of analysis. Too many variables that are not visible in the final standings we can see.
Lmao tiers have EVERYTHING to do with seasons. In a system where getting the most points gets you the highest rankings, you always want to be at the highest multiplier possible to get the most points. There are many different factors and systems that go into determining your final season rank, but they are all related and nothing is irrelevant
In the very limited data I've collected, there's a hint that tier correlates strongly with bracket (i.e. the higher your war rating the higher your final bracket) from master to gold 2. Interestingly, from Gold 3 downward, that correlation seems to get much weaker. Meaning: alliances stop showing up roughly in order of rating, and start getting much more mixed up. It is unclear precisely why, although it is possible that one strong factor is that from Gold 3 and lower the competition is much less strong - alliances are not trying as hard to maximize war performance at that level, maybe because the rewards are not high enough to encourage alliances to put out maximum effort.
This is based on an extremely tiny and not statistically strong data set, so take with a pile of salt.
As an aside, it is odd that whenever I post something like this on the forums and reddit, more forum readers tend to find it on reddit rather than here.
Clearly you didn't read the following comment. I said irrelevant was the wrong choice of wording. There are Tiers (Tier 1-20), then Season Brackets. (Plat 1-3 etc). They line up somewhat but they're two different systems that operate differently. Not sure what factors and systems you're referring to in terms of Season Rank, but it's pretty straightforward. It's a Leaderboard. It's based on Points. Your Bracket determines your Multiplier, and it can flux based on the results that other Allies put up. You win, you go up. Someone scores more than you, you go down.
It helps to think before a post or at least use the edit button. Why post something and then go in the complete opposite direction right after? Alliance war tiers ARE RELEVANT with alliance war rankings. All the different systems in alliance war (tiers, war rating, leaderboard, point scoring) are relevant with each other and work together to determine your final leaderboard ranking.
You're not grasping what I'm saying and you seem to be looking for an argument so best of luck.
Yeah, I'm not sure how someone in tier 6/7 could hit Gold 1. We finished 1177 with 13.6 million pts and every single one of our matches was in tier 5 or 6, with the majority of our matches in tier 5. We won 13 and lost 11, btw. Every time we would win in tier 6, our rank would actually drop slightly once all the numbers were in that night.
I think you just can't refute that tiers are relevant to alliance war seasons, but thanks.
It's more likely that he is in an alliance where tier and multiplier do not matter as much as it does between the Platinum 3 and Gold 1 tier.
I'm going to explain it one more time and you can argue with whatever you want.
Tiers are based on War Rating. You can win or lose and alter your Rating going up or down in Tiers, and that determines individual War Rewards. This can also be affected by other Allies winning or losing, but not as instantaneously as Brackets. The War Rating determines your Matches.
Brackets are based on how many Points you score. It's a live-action Leaderboard. You can't lose Points, but what you put up versus what others put up determines what Bracket you're in. It's an Arena for Allies. You can put up Points, but you can go down after as Allies put up more Points.
The Multiplier is determined by the Bracket you're in. Not the Tier you're in. We can see a reflection of the Tier in the scoring, but they have no effect on each other by design. It's a race of Points. There's no reason why the Tier needs to line up with the Bracket you're in.