Hearthpwn

Tickatus, the "OMG, I can't play control anymore, thus it's bad design" card of 2021

The "OMG, I can't play control anymore, thus it's bad design" cards over the years: Original Jaraxxus, Justicar Trueheart, Elise Starseeker, N'Zoth, Kazakus, Jade Idol, Un'Goro Warrior Quest, Dead Man Hand, Coldlight Oracle, Bloodreaver Gul'dan, Voidlord, Rin, the First Disciple , Baku Odd Warrior, Dr.Boom Mad Genius, Eternal Servitude, Lesser Diamond Spellstone.......

You can't outgreed the greediest out there, change your gameplan.

Just BHGed a Mountain Giant, feels like home man

Not quite, Vintage and Legacy are currently and have been for quite some time the most balanced formats. While Standard and Modern are burning to the ground every single season. The problem of wild is the lack of tools to fight hyper aggro and combo. In MTG you can Thoughtseize their combo pieces, Force of Will their busted stuff, Tax greedy cards, Sword to Plowshares as a universal catch all removal. While here you get to Loatheb once on turn 5, Dirty Rat attempts and everyone is capped at 30 life, no overhealing, making healing cards worse because you can get their full value all the time, only when you were damaged.

Oh man, I thought the discussion was over, poked in and want to gouge my eyes out. I promise that I will never bring this subject up again just to not stir up the conspiracy folks.

Quote from Pizzacats >>

Quote from Shed >>

I understand what you are talking about, what you seem to fail to understand is that I'm not trying prove that it works that anyway and that it LIKELY works that way, it's impossible to 100% absolutely prove it without looking into the code, even with a bajillion games worth of data it is still just anecdotal evidence.

I understand the gist of discussions and you are using quite a lot of casual fallacies yourself, if you go back and read your posts you realize that you are simply trying to discredit me with these fallacies. You are threading in the "you can't prove god does not exist, therefore it exists" realm

Tackle the subject at hand and prove me otherwise with actual game design arguments and logistics as to why such system is not into the game instead of grass iguanas and red herrings. See you tomorrow.

 The burden of proof goes to the accuser. I do not work for blizzard, I do not have to prove anything.  If you want to prove that you are right, that's on you.   And I'm not asking for "a bajillion games"  If you had even 1000 games that proved this hypothesis, it would be quite credible, and beyond anecdotal evidence of 30-50 games.

As for the "you can't prove god doesn't exist, therefore it exists." statement, it's quite literally the opposite of that analogy.  I am saying you can't prove it exists.  No therefore needed.  Except in this case, I never even said you can't prove match making rigging isn't happen.  It is definitely possible to prove it with enough data. It is definitely possible Blizzard gathers archetype information.  However, it is not my job to prove that, it is the job of the person who accuses that such is the case.  If you wanted to prove that god existed, it would be the job of the person who claimed that he existed  to do so.  SO no, you are literally taking my argument the opposite of what it actually means, another fallacy.

 Well, I brought up enough information on the matter that should bring everyone to the same page.

Like I said before there is no proving without looking up the code. But everything points towards it's likelihood:

*One of the core features of the system is the ability to predict outcomes correctly.
*It's necessary for the deck to be taken into consideration if the system wants to offer fair matches.
*The lack of such feature would lead to exploitation of the system, which in turn would lead to an unbalance on the system. https://win.gg/news/7500/riot-created-pro-player-accounts-to-be-disabledafter-unfair-mmr-gains
*I explained a similar implementation for League of Legends, if you are too good you are paired with bad teammates to keep balance. https://leagueoflegends.fandom.com/wiki/Matchmaking
*The system does match you with different difficulties to prevent staleness.
*Not having it also lead to some extreme cases of bad player experience(caused by coincidence), which could be prevented otherwise.

AND AGAIN I'M NOT SAYING THAT GAMES ARE RIGGED FOR YOU TO LOSE OR TO BUY PACKS OR ANY SORT OF TINFOIL HAT TALK. If it is rigged for you to lose, someone else is winning. I'M TELLING THE SYSTEM IS BIASED DUE IT'S OWN NATURE, IT SLIGHTLY TIPS THE ODDS EITHER SIDE SOMETIMES TO MATCH YOU AGAINST BETTER/WORSE PLAYER OR MATCHUPS.
To further elaborate on that, picture this: You click play and the game starts to look for a match, it chooses a 60-40 match against you. That opponent either have a better skill rating with a neutral matchup, a neutral skill rating with better matchup or slightly better skill rating and matchup. It simply runs the calculations and give you a random guy based on the odds.

Now regards to OP post, deck A and deck B have different matchups, thus the pool of fair matchups is different, playing a counter deck slightly offsets your chance of finding the one you are countering. It's the nature of the game, most of the time it's fine, sometimes you are on the short end of the stick, sometimes your opponent is.

I understand what you are talking about, what you seem to fail to understand is that I'm not trying prove that it works that anyway and that it LIKELY works that way, it's impossible to 100% absolutely prove it without looking into the code, even with a bajillion games worth of data it is still just anecdotal evidence.

I understand the gist of discussions and you are using quite a lot of casual fallacies yourself, if you go back and read your posts you realize that you are simply trying to discredit me with these fallacies. You are threading in the "you can't prove god does not exist, therefore it exists" realm

Tackle the subject at hand and prove me otherwise with actual game design arguments and logistics as to why such system is not into the game instead of grass iguanas and red herrings. See you tomorrow.

He clearly said that the algorythm needs to predict and choose the winner correctly to be good, in order for this to happen in Hearthstone, decks must be taken in consideration, 50/50 matchup in skill does not translate to a 50/50 game if it is not a perfect mirror match. It is also said somewhere in the video(in the beggining I believe) that you should tune the matchmaking to the needs of your game. And finally it is said that you shouldnt only match player into 50/50 scenarios as it gets stale over time and you should throw the player a multitude of different scenarios.

Here is a fun scenario: Begginer player makes his first competitive deck, however he was unlucky and this deck has a poor matchup against the most popular deck around, if decks arent taken into consideration this guy is gonna have a terrible player experience and probably stop playing. And as it was stated multiple times in the video, you don't want this to happen.

To prove that, you need dat, not generalized statements about how mmr and matchmaking systems work.

I was giving generalized statements because it is a complex subject that not everyone will understand right away if in-depths, however when I went in-depth and brought an expert you called bogus.

Quote from Pizzacats >>

Quote from Shed >>

Quote from Pizzacats >>

Quote from Shed >>

I'm tired now. I'll have Dr. Josh Menke, Senior System Designer at Activision do the talking.

https://www.youtube.com/watch?v=-pglxege-gU

 sharing a longass video and expecting us to draw any sort of conclusion from it about how it supports this hypothesis is preposterous.  If you really want to use a video like that, give timestamps and quotes and context as to how it supports this hypothesis.

 Well, the entire video is important actually. It's a very technical and complex subject, either or you watch everything and understand that there many intricate parts that compose the system as a whole or you dont understand anything

 Oh yes and once I “understand” it, the hypothesis will suddenly make sense! Great! You realize that’s what ALL conspiracy theories do. They say “this article/video/study PROVES X!!! And if you don’t think so, you just don’t get it!” That’s like flat earthers saying “The bottom of a shoe being flat PROVES the earth is flat!!! It would obviously be round on bottom if the earth was round!”

if you want to go beyond the logic of a conspiracy theorist, you actually have to put in work to show how they correlate.

You are the one clinging very hard to simple answers on a complex multilayered statistics and design problem

I'm telling that this stuff is required for the system to work properly, while you are on the "It just works" team

Well there you go:

42:30 starts Q&A and the second question is about fair matches. There is a comment listing all the questions and their time stamps.
33:00 for an explanation on several skill-based matchmakings, graphs included
13:45 Opening explanation on matchmaking, the first 3 slides shows well what it is supposed to do.

Just watch the entire thing already, it is nice regardless.

Quote from Pizzacats >>

Quote from Shed >>

I'm tired now. I'll have Dr. Josh Menke, Senior System Designer at Activision do the talking.

https://www.youtube.com/watch?v=-pglxege-gU

 sharing a longass video and expecting us to draw any sort of conclusion from it about how it supports this hypothesis is preposterous.  If you really want to use a video like that, give timestamps and quotes and context as to how it supports this hypothesis.

 Well, the entire video is important actually. It's a very technical and complex subject, either or you watch everything and understand that there many intricate parts that compose the system as a whole or you dont understand anything

I'm tired now. I'll have Dr. Josh Menke, Senior System Designer at Activision do the talking.

https://www.youtube.com/watch?v=-pglxege-gU

Oh man, you guys either overestimate or underestimate the system and it's purpose.

It's not for profit, it's not match fixing and it only slightly skews the odds against or towards you, it is necessary for the ranked system to work properly, otherwise things would be insane and it is standard practice for competitive gaming. Let me explain how the League of Legends system works, since it is already widely known to be true.

There, they also want to keep every single player at 50% and it is required for a player to have >50% to climb, not only that, but when you are trying to promote to a higher league, you need to beat someone else into yours, taking it's place, they want that every single league have a fixed amount of player %. Now, how does the matchmaking works? The game will attempt assemble teams with roughly the same MMR rating, this rating comes from the players themselves and if there DUO players(Queuing as duo increases the MMR required, as it boosts the team strenght), this means that the better you are, the worse will be your teammates, sure the same thing can happen there, but it is more probable for them to have a more stable and consistent player line. Having a winstreak also plays a role, as it greatly increases your MMR, since the game wants you to rank up faster to prevent further disruption on low leagues and it assumes it will average out when get to a higher league will better players.

Hearthstone wants to do the same thing, if the system was simply based off MMR there would an increasingly higher number of players with higher and lower WR at lower ranks, creating a wonky situation where the player pockets would try to balance themselves creating even more disparity, they do not want 70% of the active player base trapped at rank 25 with the top 1% at rank 10. Having nothing to balance out besides MMR also create a Circular Meta, which is highly exploitable, as you are matched randomly, it is highly likely that you will be matched with popular decks, to give yourself a boost, play a counter deck, eventually it will be the most popular one, then play a counter to that one instead and on and on. Remember, all you need to climb is a >50%, you don't need to face the most popular deck 70% of the time to do this, if the counter is neutral across the board, if the popular one shows up only 3% more than other classes, it is already effective. This is why it is they have this parameters into play, they don't want +70%WR players running rampant, there is a very unhappy and unstaisfied player with -30% for every single one of those players.

Quote from Pizzacats >>

Quote from Shed >>

Quote from Pizzacats >>

 

Quote from Shed >>

AhQuote from Pizzacats >>

You really need a large sample size to give any credence to this hypothesis. It’s just confirmation bias. You play a few games that aren’t weapon rogue and suddenly they’re gone! Well I got news for you: sometimes, there are things called coincidences. You need to play 100+ games with weapon tech and then 100+ games without weapon tech, record the data, and THEN maybe you’ll have at least somewhat relevant data that isn’t just confirmation bias or conspiracy theory because you’re not winning. In fact, 100 might not be enough because there are still anomalies relatively likely to happen with 100 data points, so you should really play 1000+ if you wanted to be sure

 Thankfully I play since Beta and have used a tracker for the most part. But uncovering the data may be difficult I have played hundreds of decks over time against a even greater pool of decks and we have to take into consideration the metas and whatnot and nobody wants to do that.

But it has happened way too many times over the times to be merely a coincidence, also it's standard game design in competitive multiplayer games to try keeping players at 50%WR for balance reasons, but ELO systems alone do not work for Hearthstone due card variance. You can easily notice the absence of the system in tournament settings, as fine tuning your decks to beat what you think the meta will be or even flat out play anti-meta decks actually works.

And now a very small sample size that I can comment about:
One of the decks I played the most this season was Tickatus Warlock for a total of 73 games. Two versions: 59 Games with the first one and 14 with the second one(I dropped it due bullshit).

The 1.0 build was more balanced and featured very little healing, the main goal was to beat down your opponent with midrangy dudes like Void Drinker and had the Corrupt Package

The 2.0 version featured Miniset cards and way more healing and removal, plus two Acidic Swamp Ooze, dropping the midrange dudes and not going for the infinite value with the Envoy Rustwix package, although the card was played by itself.

Version 1.0 with a solid 29-30W/L with a kinda even class distribution with Shaman leading at 17% and Priest at the bottom with 3%. Usually control at very low end of the spectrum, kudos to Warlock decks at 7%

Then Version 2.0 comes along with the abysmal 4-10 W/L. Warlock shot up to 33% with ALL OF THEM being infinite Rustwix, Warrior increased from 14% to 20% with a mix of Big Warrior and ETC Combo, those 3 decks don't care about healing, removal or Tickatus and easily outlast me, Rogues did have an increase to 13% from 5%, but all good control matchups dropped to 0%, no mages or priests. You know who else dropped to 0%? Paladins dropped to 0% from 10% despite being the most powerful class with multiple high tier decks, the Paladin matchup is THE BEST ONE for this deck, 62% according to HSReplay. Since there are so few matches I will list it.
Warlock: 5/14
Warrior: 3/14
Druid:  2/14
Rogue: 2/14
Shaman: 1/14
Hunter: 1/14
DH: 1/14

Ah, speaking of Paladins, Warlocks always competing for the top spot on my charts 9%(Rogues being the top at 12%) and 13%(Druids with 17%) for Pure Paladin and Ramp Paladin respectively, over 100 games combined

Your data for v 1.0 seems reasonable, but 14 games is not nearly a large enough sample size to make a conclusion.  You also lied/exaggerated about not facing rogues.  Previously, you stated about rogues, that if you added weapon tech, you would "never see them anymore."  2/14 is FAR from never, and is even HIGHER than when you didn't have weapon tech, so your original argument and the point of this post is completely invalidated.  The fact that you didn't go against your best matchups is just a coincidence and could be due to a meta shift or you just got unlucky, since it's only 14 games.  Plus, you should take mmr into account.  If you are losing repeatedly, you will change the types of decks you see, facing other decks which also might be greedy with lower win rates.  Prime example of this is Markmckz, who just stays at a rank floor and faces meme decks all the time.

 My argument was not over weapon removal makes rogues go away, it is over that the matchmaking is biased.

I agree that 14 games is too little, however I do have thousands of games registered spanning years here to backup this claim, however I won't bother compiling and analyzing all this data, it requires too much time and I need a HSReplay Premium subscription to get things that I either deleted or archived.

but here a quick freebie I analyzed here, the day I changed decks was already two weeks into the Miniset and I played some games with both and then switched to a Fel Demon DH, all of them had the odds stacked against them with mix of bad and neutral matchups to balance out.

1.0 : DH(W), Hunter(L), DH(L), Hunter(W), Druid(W), DH(W)
2.0(Only Losses this day): Warrior, Druid, Rogue, Warlock, Warlock, Warlock
Fel Demon DH(Currently at 57% with 56 games): Paladin(Null), Paladin(W), Shaman(L), Rogue(L), DH(W), Warlock(L), Warlock(W), Mage(W), Warlock(L), Druid(W), Paladin(W), Paladin(L), Shaman(W), Mage(L).

We can even watch the replays to list out archetypes, important card interactions and opening hands of each match to further analyze the matter. Did my bad MU lose because of brick hands, lack of key cards? Did I win my bad MU because I drew well, had the key cards? Who knows.

If youare claiming that blizzard has coded a way to rig your draw to make it so you lose a certain percentage of matchups, youre crazy.  Blizzard could barely code Zephrys to work, let alone an entire card game where every card drawn is coded like Zephrys, in that it's drawn based on the likelyhood that it is great for whatever situation youre in.

 God no, not only it is impossible to do, but dumb too.

What I meant is that by analyzing each game you can have a better grasp of the matchmaking algorithm, for instance, you have a good ELO/WR and the game will try to pair you with a bad matchup because it wants you to lose, but you win anyway. Why is that? Here some options:

*In an attempt to even out, the game matched you with a worse player WRwise and you outplayed him.
*You were matched against an even skilled player, he was unlucky and had a bad hand and/or missed key cards
*You were matched against an even skilled player, but on this ocasion you had your key cards to play out the matchup

Now, if you analyze every game and every piece of data available and observe stuff like this very frequently, it possible to tell that the matchmaking is trying to match you with opponents that will drive yours and theirs WR to 50%. AND I'M NOT SAYING THAT GAMEPLAY ITSELF IS RIGGED, the game don't choose your draws, the randomizer algorithm is very good actually.

Quote from Pizzacats >>

 

Quote from Shed >>

AhQuote from Pizzacats >>

You really need a large sample size to give any credence to this hypothesis. It’s just confirmation bias. You play a few games that aren’t weapon rogue and suddenly they’re gone! Well I got news for you: sometimes, there are things called coincidences. You need to play 100+ games with weapon tech and then 100+ games without weapon tech, record the data, and THEN maybe you’ll have at least somewhat relevant data that isn’t just confirmation bias or conspiracy theory because you’re not winning. In fact, 100 might not be enough because there are still anomalies relatively likely to happen with 100 data points, so you should really play 1000+ if you wanted to be sure

 Thankfully I play since Beta and have used a tracker for the most part. But uncovering the data may be difficult I have played hundreds of decks over time against a even greater pool of decks and we have to take into consideration the metas and whatnot and nobody wants to do that.

But it has happened way too many times over the times to be merely a coincidence, also it's standard game design in competitive multiplayer games to try keeping players at 50%WR for balance reasons, but ELO systems alone do not work for Hearthstone due card variance. You can easily notice the absence of the system in tournament settings, as fine tuning your decks to beat what you think the meta will be or even flat out play anti-meta decks actually works.

And now a very small sample size that I can comment about:
One of the decks I played the most this season was Tickatus Warlock for a total of 73 games. Two versions: 59 Games with the first one and 14 with the second one(I dropped it due bullshit).

The 1.0 build was more balanced and featured very little healing, the main goal was to beat down your opponent with midrangy dudes like Void Drinker and had the Corrupt Package

The 2.0 version featured Miniset cards and way more healing and removal, plus two Acidic Swamp Ooze, dropping the midrange dudes and not going for the infinite value with the Envoy Rustwix package, although the card was played by itself.

Version 1.0 with a solid 29-30W/L with a kinda even class distribution with Shaman leading at 17% and Priest at the bottom with 3%. Usually control at very low end of the spectrum, kudos to Warlock decks at 7%

Then Version 2.0 comes along with the abysmal 4-10 W/L. Warlock shot up to 33% with ALL OF THEM being infinite Rustwix, Warrior increased from 14% to 20% with a mix of Big Warrior and ETC Combo, those 3 decks don't care about healing, removal or Tickatus and easily outlast me, Rogues did have an increase to 13% from 5%, but all good control matchups dropped to 0%, no mages or priests. You know who else dropped to 0%? Paladins dropped to 0% from 10% despite being the most powerful class with multiple high tier decks, the Paladin matchup is THE BEST ONE for this deck, 62% according to HSReplay. Since there are so few matches I will list it.
Warlock: 5/14
Warrior: 3/14
Druid:  2/14
Rogue: 2/14
Shaman: 1/14
Hunter: 1/14
DH: 1/14

Ah, speaking of Paladins, Warlocks always competing for the top spot on my charts 9%(Rogues being the top at 12%) and 13%(Druids with 17%) for Pure Paladin and Ramp Paladin respectively, over 100 games combined

Your data for v 1.0 seems reasonable, but 14 games is not nearly a large enough sample size to make a conclusion.  You also lied/exaggerated about not facing rogues.  Previously, you stated about rogues, that if you added weapon tech, you would "never see them anymore."  2/14 is FAR from never, and is even HIGHER than when you didn't have weapon tech, so your original argument and the point of this post is completely invalidated.  The fact that you didn't go against your best matchups is just a coincidence and could be due to a meta shift or you just got unlucky, since it's only 14 games.  Plus, you should take mmr into account.  If you are losing repeatedly, you will change the types of decks you see, facing other decks which also might be greedy with lower win rates.  Prime example of this is Markmckz, who just stays at a rank floor and faces meme decks all the time.

 My argument was not over weapon removal makes rogues go away, it is over that the matchmaking is biased.

I agree that 14 games is too little, however I do have thousands of games registered spanning years here to backup this claim, however I won't bother compiling and analyzing all this data, it requires too much time and I need a HSReplay Premium subscription to get things that I either deleted or archived.

but here a quick freebie I analyzed here, the day I changed decks was already two weeks into the Miniset and I played some games with both and then switched to a Fel Demon DH, all of them had the odds stacked against them with mix of bad and neutral matchups to balance out.

1.0 : DH(W), Hunter(L), DH(L), Hunter(W), Druid(W), DH(W)
2.0(Only Losses this day): Warrior, Druid, Rogue, Warlock, Warlock, Warlock
Fel Demon DH(Currently at 57% with 56 games): Paladin(Null), Paladin(W), Shaman(L), Rogue(L), DH(W), Warlock(L), Warlock(W), Mage(W), Warlock(L), Druid(W), Paladin(W), Paladin(L), Shaman(W), Mage(L).

We can even watch the replays to list out archetypes, important card interactions and opening hands of each match to further analyze the matter. Did my bad MU lose because of brick hands, lack of key cards? Did I win my bad MU because I drew well, had the key cards? Who knows.

AhQuote from Pizzacats >>

You really need a large sample size to give any credence to this hypothesis. It’s just confirmation bias. You play a few games that aren’t weapon rogue and suddenly they’re gone! Well I got news for you: sometimes, there are things called coincidences. You need to play 100+ games with weapon tech and then 100+ games without weapon tech, record the data, and THEN maybe you’ll have at least somewhat relevant data that isn’t just confirmation bias or conspiracy theory because you’re not winning. In fact, 100 might not be enough because there are still anomalies relatively likely to happen with 100 data points, so you should really play 1000+ if you wanted to be sure

 Thankfully I play since Beta and have used a tracker for the most part. But uncovering the data may be difficult I have played hundreds of decks over time against a even greater pool of decks and we have to take into consideration the metas and whatnot and nobody wants to do that.

But it has happened way too many times over the times to be merely a coincidence, also it's standard game design in competitive multiplayer games to try keeping players at 50%WR for balance reasons, but ELO systems alone do not work for Hearthstone due card variance. You can easily notice the absence of the system in tournament settings, as fine tuning your decks to beat what you think the meta will be or even flat out play anti-meta decks actually works.

And now a very small sample size that I can comment about:
One of the decks I played the most this season was Tickatus Warlock for a total of 73 games. Two versions: 59 Games with the first one and 14 with the second one(I dropped it due bullshit).

The 1.0 build was more balanced and featured very little healing, the main goal was to beat down your opponent with midrangy dudes like Void Drinker and had the Corrupt Package

The 2.0 version featured Miniset cards and way more healing and removal, plus two Acidic Swamp Ooze, dropping the midrange dudes and not going for the infinite value with the Envoy Rustwix package, although the card was played by itself.

Version 1.0 with a solid 29-30W/L with a kinda even class distribution with Shaman leading at 17% and Priest at the bottom with 3%. Usually control at very low end of the spectrum, kudos to Warlock decks at 7%

Then Version 2.0 comes along with the abysmal 4-10 W/L. Warlock shot up to 33% with ALL OF THEM being infinite Rustwix, Warrior increased from 14% to 20% with a mix of Big Warrior and ETC Combo, those 3 decks don't care about healing, removal or Tickatus and easily outlast me, Rogues did have an increase to 13% from 5%, but all good control matchups dropped to 0%, no mages or priests. You know who else dropped to 0%? Paladins dropped to 0% from 10% despite being the most powerful class with multiple high tier decks, the Paladin matchup is THE BEST ONE for this deck, 62% according to HSReplay. Since there are so few matches I will list it.
Warlock: 5/14
Warrior: 3/14
Druid:  2/14
Rogue: 2/14
Shaman: 1/14
Hunter: 1/14
DH: 1/14

Ah, speaking of Paladins, Warlocks always competing for the top spot on my charts 9%(Rogues being the top at 12%) and 13%(Druids with 17%) for Pure Paladin and Ramp Paladin respectively, over 100 games combined

They keep tabs on decks, cards and player winrates and they will pair you up to the perfect opponent to keep your WR balanced. It is like this ever since the game started.