Hearthpwn

Lead dancer in a renathal deck seems horrible since it can summon renathal 

also, clears aren’t great in wild anymore on their own. You need to swing the board the same turn you clear, that’s what makes the murlocs so good

Yeah, but it’s also a board swing. And that’s huge. You also have to clear it or else lose on the spot since you cannot redevelop

Dirty rat isn’t good atm 

The combo op mentioned is useless but flurgl tox is broken as are the murlocs.

The problem is that it’s not an amazing deal and very slow

The comments here prove nothing. It’s like with all conspiracy theories: a mixture of anecdotes and personal experiences that are influenced by a lot of psychological mistakes (primacy/recency/halo effect to name a few, if you are interested in how real studies work, you can look them up) plus some data with a way to small sample size to ever be significant. To draw conclusions about matchups, you would need a) play at least hundreds better thousands of games per deck, b) adjust the data for class/deck frequency changes that happen all the time, c) include cofounders like mmr and deck wr into your calculations (since winning more leads to facing better decks resulting in less diversity the higher your mmr gets, since the amount of playable decks is reduced/losing more makes way more decks viable) and d) analysis of the data including a p-value (how likely is the data the product of pure rng), graphics and significance tests while e) that becomes very difficult since no algorithm in the world produces true randomness, that is to the date only possible with analoge randomness generators.

Since no one in such posts even complete step a) - not to speak of the whole statistics thing, for that you need to have experience from sience/maths university classes- there is no proof anywhere here. The one guy claiming he has data since beta doesn’t even share the raw data - not that it would help much, he would have to play all his games with a tracker and you would need a bunch of data about the respective metas, deck frequencies, etc pp.

The „experiment“ of playing 10 games with each deck is no experiment: sample size way to small to ever be significant, none of the data necessary to determine the „natural“ rate of a deck, not repeatable by anyone since it’s mmr dependent etc pp

Discover is way to weak to give it a stat penalty. Vanilla stats are a penalty by themselves 

Quote from MixedJuice >>

Discover is only a problem when the card pool is very limited for a specific class. At that point you just have an extra copy of your strongest stuff.

But the mechanic itself is not a problem.

 Yeah, the solution might be to make discover pull from all cards including wild.

Discover is never good in wild since the variance is so big. That wouldnt help with a new class and it’s class specific discovers like spells but it would fix the problem of discover being so consistent after rotation

That’s not even real discover, since the player controls the pool. So completely different from what was discussed in other posts here. Rewind is a fine card, you have decision points wether to play a 4th spell or not and risking to dilute the pool, and your opponent knows which cards you could get. So no rng blowout here

rewind for mage is not really random and not really discover either 

you mean situational cards like… blightfang? A card that will be nerfed tomorrow for being played…

It’s one of the reasons to play hs

Yeah but if I have the choice between them earning money through cosmetics or actually gameplay relevant stuff that’s an easy decision 

spending money on cosmetics is a bad idea in the first place.

Alternatively they could print some very specific hate cards that suck in all other matchups but win the non game one like 1 mana 0/1 remove neptulon from the game (wherever it is). But that’s an unfun counter against an unfun and not that good deck. Alternatively just make the highroll one turn slower -> illuminate reduces by 2. the second one seems the better idea