Rowanmantle wrote: ↑Wed Jan 11, 2023 11:37 amI wholeheartedly agree, we need a 7 day lockout. Bigger bonuses for picking a realm/faction and staying with it.JohnnyWayne wrote: ↑Wed Jan 11, 2023 7:05 amAll of these are heavy premade players with a ton of simps backing them. Also xrealmer. Statistically they would be removed as outliers.
Also no premades allowed in sc's or open rvr. Only pug warbands without comms. Then, finally RoR will be a safe space to enjoy again.
Data Mining the Killboard Scenarios | Jan 2023
-
- Posts: 182
Re: Data Mining the Killboard Scenarios | Jan 2023
Ads
-
- Posts: 182
Re: Data Mining the Killboard Scenarios | Jan 2023
Yes you clearly know how to read statistics. This is the exact reason why you use medians and not unfiltered max values. I really like how people with no clue jump into conclusions here and smartass someone who knows his way around data....inoeth wrote: ↑Wed Jan 11, 2023 8:35 amlol "statisticly" but in reality you have to deal with these ppl and also these classes are chosen because they are superior to other classes...JohnnyWayne wrote: ↑Wed Jan 11, 2023 7:05 amAll of these are heavy premade players with a ton of simps backing them. Also xrealmer. Statistically they would be removed as outliers.
how much more proof do we have to bring till something is done about this plague?
Btw, you dont have to deal with these "classes" but these "people" you see in the max values. They'd just jump to the next best thing and the very same discussion starts anew.
One approach would be to get the 0.95 percentile and then average key values over all classes. I'd still expect order to overperform there and you will have outliers on both sides. You should also be aware that you can't compare ratios directly with the other faction as inner balancing is skewing them. What I mean is, that one overperforming class will decrease the ratio for other classes on the same side or vice versa.
- Evilspinnre
- Posts: 368
Re: Data Mining the Killboard Scenarios | Jan 2023
Or maybe try running with a competent group and learn how to play your class and you will succeed more often in scs, imagine that
Evilspinnre - Nightmare/Daydream - Xrealmers Anonymous
80+ AM, WL, WP, BW, SL, SM, 50+ RP, SW, IB
80+ BG, Sorc, WE, Mara, Choppa, SH, 60+ Zeal, Shaman, Blork
https://www.youtube.com/user/EvilspinnrePvP
https://www.twitch.tv/evilspinnrepvp
80+ AM, WL, WP, BW, SL, SM, 50+ RP, SW, IB
80+ BG, Sorc, WE, Mara, Choppa, SH, 60+ Zeal, Shaman, Blork
https://www.youtube.com/user/EvilspinnrePvP
https://www.twitch.tv/evilspinnrepvp
Re: Data Mining the Killboard Scenarios | Jan 2023
Yes that's good. Get great at the game or stop playing it. Not everyone has friends in the game yet and some people play for fun. How is this statement good for new players?Evilspinnre wrote: ↑Wed Jan 11, 2023 3:30 pm Or maybe try running with a competent group and learn how to play your class and you will succeed more often in scs, imagine that
The game should allow for new players to not get worked so hard by vets.
- agemennon675
- Posts: 505
Re: Data Mining the Killboard Scenarios | Jan 2023
Empathy my friend they dont have itFFNation wrote: ↑Wed Jan 11, 2023 8:07 pmYes that's good. Get great at the game or stop playing it. Not everyone has friends in the game yet and some people play for fun. How is this statement good for new players?Evilspinnre wrote: ↑Wed Jan 11, 2023 3:30 pm Or maybe try running with a competent group and learn how to play your class and you will succeed more often in scs, imagine that
The game should allow for new players to not get worked so hard by vets.
Destruction: 40-BG / 40-DoK / 40-Chosen / 37-Mara / 37/Sorc / 36-SH / 36-Choppa / 24-Shaman / 16-WE
Order: 40-SW / 40-SM / 40-WP / 40-WL / 39-Kotbs / 38-BW / 33-AM / 22-WH / 16-RP / 12-Slayer
Order: 40-SW / 40-SM / 40-WP / 40-WL / 39-Kotbs / 38-BW / 33-AM / 22-WH / 16-RP / 12-Slayer
Re: Data Mining the Killboard Scenarios | Jan 2023
Sure it would be ideal to look at the top 5%, 10%, 20% of players and compare class by class then realm by realm. However as it wasn't put together certainly you can look at the best maximum performance of a class and judge from that. It is not just one outlier match either and having basically 8/10 of the best maximum performance allocated to 2 classes is a good hint that something is going on. Generally I think it is not healthy for a game to reach anything over 90% winrate on anybody.JohnnyWayne wrote: ↑Wed Jan 11, 2023 1:46 pmYes you clearly know how to read statistics. This is the exact reason why you use medians and not unfiltered max values. I really like how people with no clue jump into conclusions here and smartass someone who knows his way around data....inoeth wrote: ↑Wed Jan 11, 2023 8:35 amlol "statisticly" but in reality you have to deal with these ppl and also these classes are chosen because they are superior to other classes...JohnnyWayne wrote: ↑Wed Jan 11, 2023 7:05 am
All of these are heavy premade players with a ton of simps backing them. Also xrealmer. Statistically they would be removed as outliers.
how much more proof do we have to bring till something is done about this plague?
Btw, you dont have to deal with these "classes" but these "people" you see in the max values. They'd just jump to the next best thing and the very same discussion starts anew.
One approach would be to get the 0.95 percentile and then average key values over all classes. I'd still expect order to overperform there and you will have outliers on both sides. You should also be aware that you can't compare ratios directly with the other faction as inner balancing is skewing them. What I mean is, that one overperforming class will decrease the ratio for other classes on the same side or vice versa.
-
- Posts: 182
Re: Data Mining the Killboard Scenarios | Jan 2023
Yes i can tell you what is probably going on. Certain people play a certain way. These will use the best at their disposal to win. That means, and this is the biggest advantage, premades only, ideally farming pugs. It goes to such an extent, that they stopped queueing once they were forced to fight similar minded premades. That is also the reason why ranked is dead, a mode that was designed for them to compare performance with others. The match maker is designed to make such numbers possible, even cater to players to reach such win ratios.Valfaros wrote: ↑Thu Jan 12, 2023 9:50 pmSure it would be ideal to look at the top 5%, 10%, 20% of players and compare class by class then realm by realm. However as it wasn't put together certainly you can look at the best maximum performance of a class and judge from that. It is not just one outlier match either and having basically 8/10 of the best maximum performance allocated to 2 classes is a good hint that something is going on. Generally I think it is not healthy for a game to reach anything over 90% winrate on anybody.JohnnyWayne wrote: ↑Wed Jan 11, 2023 1:46 pmYes you clearly know how to read statistics. This is the exact reason why you use medians and not unfiltered max values. I really like how people with no clue jump into conclusions here and smartass someone who knows his way around data....
Btw, you dont have to deal with these "classes" but these "people" you see in the max values. They'd just jump to the next best thing and the very same discussion starts anew.
One approach would be to get the 0.95 percentile and then average key values over all classes. I'd still expect order to overperform there and you will have outliers on both sides. You should also be aware that you can't compare ratios directly with the other faction as inner balancing is skewing them. What I mean is, that one overperforming class will decrease the ratio for other classes on the same side or vice versa.
For statistics and data science, it is exactly these outliers you want to get rid off. They do skew your analysis and will show fraud patterns in your data. From the data we got, you can certainly tell that something is going on here. The what is pure speculation so far and I gave you my judgment on what we could do to find answers. Take it or leave it, but stop that unbased witch hunt people are trying to start here.
Re: Data Mining the Killboard Scenarios | Jan 2023
If the best always tend to use the same classes because they yield the best results its not an outlier and you can't just cut it from your data because you choose to do so. Not good practise to just dismiss it. This is not some kind of measurement error that would justify it. It's real people abusing a class to such an extreme that shouldn't be possible to begin with, no matter what awful tactic they use to get there. It is further an actual problem that shouldn't be just dismissed and rather needs addressing as they have a large negative impact on a big portion of the sc community.JohnnyWayne wrote: ↑Thu Jan 12, 2023 11:35 pm
Yes i can tell you what is probably going on. Certain people play a certain way. These will use the best at their disposal to win. That means, and this is the biggest advantage, premades only, ideally farming pugs. It goes to such an extent, that they stopped queueing once they were forced to fight similar minded premades. That is also the reason why ranked is dead, a mode that was designed for them to compare performance with others. The match maker is designed to make such numbers possible, even cater to players to reach such win ratios.
For statistics and data science, it is exactly these outliers you want to get rid off. They do skew your analysis and will show fraud patterns in your data. From the data we got, you can certainly tell that something is going on here. The what is pure speculation so far and I gave you my judgment on what we could do to find answers. Take it or leave it, but stop that unbased witch hunt people are trying to start here.
Ranked had to fail from the start. I can't possibly think of any gameplay more boring than 6v6 in which you fight each other for 10 minutes without anything close to dying. Or the other extreme, complete stomps. But if at any point those overly strong defensive mechanics within game get challanged because they cause this increadible power difference between pug and premade people retort to form a group yourself. Which is ridiculous as the premades themselves have proven they also heavily dislike playing against like minded groups.
Ads
- GrishnakBlagtoof
- Posts: 129
- Contact:
Re: Data Mining the Killboard Scenarios | Jan 2023
Yep, we generally feel the same way about a lot of this stuff, but you can't tell me 58% is a table-turner in battles where Slayers and White Lions are literally in the above 90th Percentile in wins. Might be 58% Destruction on, but how many of them are actually jumping in SC's, or Open RvR? Just Questing or PvE? 58% isn't enough in any way, shape or form to swing a data proven underperforming Realm BACK to a win. Order's held the domination today/tonight all night - war bands make a difference and regardless of the % on it shows, Order, on average, has 3 war bands roaming at any given time vs Destro's ONE. And Order is flat out dominating. It's 12:49am NA PST, so generally where we start seeing the EU players come on.....and Order is rolling Destro across the map at their will.Paxsanarion wrote: ↑Tue Jan 10, 2023 10:29 pmTypically I agree with you....however I can assure you Order is not dominating during EU prime time......it really depends on when you play more than anything else.......and who logs Order and who logs Destro at a given moment For example right now it is 530 pm NA east coast time.......and T4 is 58% DestroGrishnakBlagtoof wrote: ↑Mon Jan 09, 2023 8:27 am
Order is just flat dominating. And years ago when it was Destro "dominating" people went to Order....and none have come back...why?
The numbers don't lie....
Wanna try and tweak the xrealming? 7 Days is too long. Make it ONE City Siege Cycle....if you log onto Destro for the first time during a siege cycle, then that's it til the next siege, you're Destro! Sure, it can be toyed with, sort of. But not to how horrible it is now, Order players pushing Destro Forts, only to swap out, log into their Order mains, and roll Forts til their heart's content. Yay free gear for gaming the system!! Woohoo, you're a hero!!
There's another game that rewards lack of ability and all farming over skill....starts with a W and ends in arcraft. This is Warhammer....where real WAR lives!!!
More often than not, Order has up to 7 Fort wins compared to Destro's 3 or 4 by the time city siege rolls around. How else are they farming Sov gear to roll newbies? In unbalanced fort attacks and City Sieges while doing Sov Premades rolling over people just trying to catch up. If you really don't see there's a very real and now proven problem with the balance here, I don't know what else to say.
The Data Mining doesn't lie.....numbers are unbiased and don't carry favour with anyone. And the numbers are telling everyone there's a huge issue that should't be ignored anymore. The evidence is no longer anecdotal where someone could just wave it off and be like "That's only how YOU feel, it's not the real numbers" because the real numbers show it's not imagined.
I love this game. Absolutely. I want to be playing this for many, many more years!!! I'm in NO WAY saying anything about getting better gear becoming easier. I think the time it takes is fine. It's the pure frustration people feel getting rolled, and now having the actual proof WHY this is happening, and people are actually defending it??
If the numbers were swapped and Destro had all this Domination Data, these forums would EXPLODE with Order players all day, every day DEMANDING insane things happen to Destruction classes and Realm as a whole. And nobody will convince me otherwise of that. lol
C'mon...how can anyone, with a straight face, defend that these numbers are completely cool? We need a separate item level bracket at 40/40 so everyone has a chance against people their own power levels to properly earn the gear they get.....but that's the rub ain't it? That's WHY people are trying to say these numbers are fine...if this gets fixed, they'll actually have to compete and not afk /fol / carry their way to rolling over people with ease. God Forbid they fight similarly geared people and ACTUALLY have a challenge, right?
I mean someone DID say Rated was dead from the get-go because it's boring having people with the same power levels fighting, because they last forever....I would think strategy would come into play when you're fighting against like-geared people. But because you don't have any strategy just further shows how it's detrimental to the game to keep it this way. Nobody knows any strategy coz who needs it when you're so overpowered you don't NEED strategy. Where's the fun in that? That's why we're here, right? Fun?
These numbers show how unfun it can get. That has to be addressed, and I have faith that the devs will address it. They're looking into solutions, an item level cap bracket may make it in, but they're also doing a balance pass to get more balance of power between characters, so yeah, it's coming, but these numbers should serve to prove the point and now the Devs actually have this hard data to see what really needs tweaking, fixing, nerfing or buffing. I have faith because I love this game and the friends I have made in it. I just want to see this place flourish and have 1500 people on, or 2000....the battles would be glorious!
So.....to be honest again, I'm fine making new characters and getting them to R40/RR40, and when the rolls in SCs start with Sov Premades or Open RvR is decimated by the same 3 war bands v 1, I'll just roll a new character. No (Green)skin off MY nose....
But I would LIKE to be able to go into SCs or Open World and actually stand a fighting chance, but right now that's not really possible. Soon as you hit 40, you're being farmed. So what do we do? Slam our heads into the wall until eventually, if we're good little nubs, after a few months getting plowed, we can catch up?
If not deal with the grind, do we quit entirely? Because I see more and more people either rolling Order to be "on the winning side" and "sick of getting farmed" or they flat quit the game. And the entire community suffers. Or do we Reroll new characters until you hit that R40/RR40 wall and have to make the choice again? We only get so many slots for so many characters...lol
Have a fantastic night!!! Get out there and FIGHT!!! WAR IS EVERYWHERE!!!
WAAAGH!!!!
WAAAGH!!!
Grishnak Blagtoof
<The Blagtoof WarTribe>
Est. 1996
Grishnak Blagtoof
<The Blagtoof WarTribe>
Est. 1996
Re: Data Mining the Killboard Scenarios | Jan 2023
the problem is the 90% winrate is less affected by class but more by other factors .... This game has as nearly all mmo's in the past a huge gap in the playerbase about teamwork and understanding of the game mechanics ...Valfaros wrote: ↑Thu Jan 12, 2023 9:50 pm Sure it would be ideal to look at the top 5%, 10%, 20% of players and compare class by class then realm by realm. However as it wasn't put together certainly you can look at the best maximum performance of a class and judge from that. It is not just one outlier match either and having basically 8/10 of the best maximum performance allocated to 2 classes is a good hint that something is going on. Generally I think it is not healthy for a game to reach anything over 90% winrate on anybody.
heck the difference is soo high that you can win 90% of your normal SC's an evening with decent Players. I can just imagine how easy it would be with 6 regular ranked Players .....
Warhammer Return of Reckoning
Verey / Ashoris / Devoras
WaR -dead-
Erengrad - Order - Frostbringer - Devoras
Verey / Ashoris / Devoras
WaR -dead-
Erengrad - Order - Frostbringer - Devoras
Who is online
Users browsing this forum: Bing [Bot] and 90 guests