GAC - Division populations
September 1, 2020 1:31AM
edited March 2021
This study is done by Taliana who doesn't use the forums but frequent discord channels. It's striking to see the shift in time. She says the hardest data points are the lowest divisions and they may be the ones that diverge most from actual densities. This data is collected via reports of people that are still at 0 banners towards the end of the first week in each division. The lowest one gets the cake but many people report nearby spots (being tied at 0 banners) anyway.
Update:
https://www.reddit.com/r/SWGalaxyOfHeroes/comments/jpeh92/what_a_difference_a_year_makes_watching_the/
@CG_Doja_Fett
Update:
https://www.reddit.com/r/SWGalaxyOfHeroes/comments/jpeh92/what_a_difference_a_year_makes_watching_the/
@CG_Doja_Fett
Post edited by Kyno on
6
- 61.6K General Discussion
- -35 Developer Q&A Archive
- 2.6K New Players
- 30.7K Arena and Character Strategy
- 3.2K Individual Events
- 5.5K Guild Events
- 4.7K PVP and Rivalry
- 51K Guild Recruitment
- 751 Art and Media
- 4.2K Off Topic
Howdy, Stranger!
Replies
Same goes for GAC. Pretty soon we'll have 4 GLs and there will be some who have unlocked, 4 GLs. How is it going to feel for those that might have one GL (or even none!) to face someone with 4 GLs? Which is entirely possible with how matchmaking works right now.
Thanks for the info!
Is this data all pulled from everyone in a discord server or they pulling data from a 3rd party?
The data is pulled by self reports through this sheet:
https://docs.google.com/forms/d/e/1FAIpQLSeAPQ8Yb-OQkZ1jPbmESI3fTOPWwCmymAP5ev3BsihBtIo5Uw/viewform
You can contact her if you want to:
Taliana
#2524
The server I came across her was Clash's Scoundrels server. I think she've been posting it to other number crunchey servers as well. She published it on reddit 6 months ago but hasn't updated it since then. Will prolly do the new one soon with the above data. She doesn't use the forums as she thinks they are a dumpster xD
While the data also pertains to total population of active players to some degree (it will miss peeps not hitting join ever) the point of doing it is GAC oriented. The obvious thing is the reports has to be done while everyone is still in carbonite. You're free to doubt the method, I don't. Any ill-willed report would be immediately noticable since reporting players don't know rest of the data and i.e. someone reporting 30kth spot while all others are reporting close spots on 23k is immediately noticable. If only a few people report in lower divisions, that makes the data further from the reality. But there's no such problem for upper divisions.
Don't they both use GP to match ?
GAC uses the GP your top X toons, where X is number of defence slots x10.
So they are different
TW matchmaking also tries to prevent repeat opponents for a few battles, but in GAC you can face an opponent in one bracket in week one, and the same opponent again next week in a new bracket
I did not doubt the method because I did not know the method. I wanted to find out whether this data allows to draw conclusions about the total number of players.
We can draw some conclusions. We can get “players who log in during the first week of GA and who are over level 85, with some caveats”. Assume probably 1-8k possible additional players per division depending on who I get to share data each month and when they enter.
So it’s a pretty good proxy for active playerbase.
My reply was general. There's no way to verify this but I think it's safe the assume those who simply do not join gac for the free rewards are vastly a minority as that's a click away even if they don't play the event. Then there are newbies below the entry barrier. I doubt the no of active accounts are more than %10 above the total numbers on the chart.
She posted it on reddit now:
https://www.reddit.com/r/SWGalaxyOfHeroes/comments/ikox24/gac_divisions_and_state_of_the_playerbase_10
Score hidden
·
6 minutes ago
Methodology: I collect data through a google forms survey on several discord servers Operation Metaverse, Planet Coruscant/HSTR Project, 50 Shards of KAM, Clash's Scoundrels server, and friendly shard, guild, and alliance servers. The instructions for the google form are quoted below:
We are estimating the size of the active player base, through finding the lowest ranked (highest number) player in each division for GAC. Please look at your guild and ally GAC leaderboards. For each division, find the first person on the list from each division who has 0 banners. Click on them to go to their profile, and enter their current GAC ranking as a number. If you don’t have any friends in that division, enter 0.
Because of CG’s coding, we know the first person listed (at 0 banners) from each division will be the worst-ranked, so you only have to check once for each division.
If so inclined you can also work through your Arena and Fleet Top 50shard rankings list to do the same. This does require more clicks but let’s you sort through an extra 100 people!
Please feel free to fill this out again later in the week when people start earning banners. That can help us find a “true bottom”.
We’ve been doing this for 8 months now; results are posted on the discords for Operation Metaverse, Planet Coruscant/HSTR project, the KAM project, and at other discords on request. So far, we are seeing massive distortion of divisions 1-3, but overall the size of the playerbase has held fairly steady. Let’s see what happened in response to changes this month!
level 2
taliana_reesei
Score hidden
·
just now
Comments on data: The quality of the datapoint depends on the number of users who collect data for me. The sampling has gotten much, much better as time has gone on - I would hypothesize that some of the "growth" in early months is actually just accessing closer to true bottom by realizing we could access more data for the same numbers of players using shard and fleet arena rankings, not just guild and ally lists.
In recent months, the biggest problem has been accessing the smaller GP divisions. The older players you tend to find on discord servers are all crammed together into division 1 - it's a lot harder to find anyone who can see divisions 4-11! Because of this, assume there could be an extra ~5-8k players in those lower divisions. I've highlighted in blue the "reliable" comparison points month to month, where I feel I have enough players over enough time to trust that my numbers are within ~1-2k of true bottom.
So they both use GP to match .
When you typed out ,use , GP and match for the answers and still can't see the point ....woooooosh
matching via GP maybe the problem ( whether it's guild or your top X )as GP is not a clear reflection of how OP a toon may or may not be .
Rather have a GL Rey or kylo at 20kGP than CUP at the same GP
GP is like calories, not all are created equal when you break them down into macros .
Ultra: TW matchmaking & GAC matchmaking are different.
You: don’t they both use GP?
Me; pointed out the differences in the matchmaking to you.
You: but don’t they both use GP?
You’re being obstinate, not clever. The matchmaking is not the same. Saying they’re the same is like saying bread is the same as pasta because they both use flour as an ingredient.
Also, good luck finding a 20k GP Rey or Kylo. They are ~50K GP and the 12 toons required to unlock them add another 250K+ between them, so they already significantly inflate the matchmaking GP of any GL owner.
Regarding the off topic on GAC vs TW, major difference is not in total vs top GP matching but in the fact that TW does not account for number of active players, which causes the sandbagging issues (intentional or otherwise).
Thanks!
Edit: Nice work!!
I would question some of your assumptions however:
* we have seen screenshots of triple-digit rankings here on the forums, yet I only see one such number in your data which makes me wonder if your numbers are low
* by the end of that first week, many players have advanced one, even two leagues; the most accurate representation of a division's population size can only be acquired in the first couple of days of a GAC, before anyone can get promoted.
"Sorry. That costs money."
- CG, probably
Again you missed my point, I never once said that the match making was the same , I'm making the point that they use the same base measurement of power which is GP , and that power level which they give toons (GP) needs to be adjusted.
I'm not being stubborn I understood ultras' point from the get go , I was pointing out that they both use GP which is a hint at what might be the problem , I tried to explain that in the quote you must posted :
" matching via GP maybe the problem ( whether it's guild or your top X )as GP is not a clear reflection of how OP a toon may or may not be "
The part were i mention guild or your Top X implies I have made a distinction between them.
If you thought that I was implying that the match making was identical then sorry ,I thought that I cleared up my hint before.
If you don't like the example of GL Vs CUP how about leader Zeta's most give the same GP but if you look at the Jedi with zeta leadership's Lumi or Qui don't really hold up against JKR or bast.
Can we at least agree that not all GP is created the same.
I think all numbers come before anyone advances a league. The misunderstanding might be due to my initial representation. The window have certainly expired by the time results of 2nd match is out. And the process of reporting starts as soon as gac launches. The important thing is that the reporters start scratching the bottom once 0 banner positions start coming in. From that point on more reports will be pushing that bottom a bit lower each time, after a bit of saturation it becomes significantly harder to come across anything lower.
I think reaction is an interesting thing to ask for, as CG has more accurate data of this exact thing.
The best example for me is Geonosians. Taken all to g12 with both zetas on GBA there is pretty much nothing that matches them for efficacy for that little GP investment. Must be several lower division matches decided by having / not having that team.
Not surprised to see the almost exponential tail graph, as the game rewards you more for having better toons with every event they come out with, there isn't anything that gives new players a boost to catch up (unless they want to spend a fortune on packs, which I wouldn't think likely for a brand new player).
Does make me wonder if they might split div1 up, as very large percentage in there.
And as for the TW match up comments, I hope they do change it. Most of ours in recent months have been decided purly based on the number of GLs each side has (we win with more, lose with less). Takes the fun out of winning knowing you have a huge advantage, and sucks knowing you won't break through also.
So would you say that GP is a inconsistent and unfair reflection of a toons/ squads power , and matching by it whether it is guild total or total of your Top X could cause problems?