Here’s what I’ve been noticing on ladder.
I’ve played against every deck here, as well as playing every deck myself with exclusion of not having personally played the Druid decks-- so I have a good feel. I’ll leave my opinion on the strength of each deck as well with a number rating from 1-10.
Coin Flip/ Scam Decks:
Big Spell Mage (8.4)
Dungar Druid (7.0)
Control Decks:
HL Rainbow Death Knight (8.1)
Odyn Warrior (8.0)
HL Druid (6.8)
BBU Death Knight (7.5)
APM Combo/ Otk Decks:
Spell Druid (8.3)
OTK Sonya Rogue (8.5)
Overheal Priest (9.1)
Fatigue Warlock (8.2)
Aggro Decks:
Pirate Demon Hunter (7.7)
Pirate Shaman (7.9)
FFU Death Knight (7.3)
Mech Rogue (8.1)
Pain Warlock (7.7)
Midrange/ Unique decks:
Cliff Dive Shaman (8.1)
Excavate Wishing Well Rogue (7.9)
Pipsi Paladin (8.7)
Incindeous Shaman (7.1)
If I didn’t list a deck you like or have seen on ladder its because I either haven’t seen it or have seen it extremely rarely which doesn’t justify ranking it. HL Druid and BBU Death Knight were very close to not being included.
Yeah it’s literally every single post an argument, matter of fact statement or just putting someone down. I don’t understand.
Pretty much every single comment or post I’ve made since coming back a few weeks ago they have replied to me like this. Way too much time on their hands.
I just posted something positive in another thread. It must be a coincidence. Anyway there is a technical issue with the concept of your thread; the sample size is just too small; do you have at least ~80 games played against each deck?
Yap Yap Yap. Problem with this, problem with that.
There’s no problem with anything, you just want to create problems out of everything. This post is my opinion on the decks I’ve been playing against and I said it in the header before rating them.
I see. You’re just doing hypocrisy. You supposedly want constructive criticism and when I give you specific technical arguments related to the sample size you just drop a regular ad hominem.
He’s a top 20 player. One of the best in the world. And you’re trying to explain to him that his opinion doesn’t matter!
We should be all be honored to hear it first hand. The sites are biased, you know it yourself. A top 20 player, as biased as they come, is a top 20 player. He literally does what he says, successfully.
Curious how you get the decimal, is it your just ranking stuff from your head? In that case, I’d expect single digits. I’m guessing your calculating something?
I’m using decimals to show marginal differences. I could have just used a 1-100 ranking system to show a whole number but I felt that a 1-10 ranking is more culturally relatable with internet tier lists and everything.
I wouldn’t say I used any calculations when making this as I didn’t add variables together to get each outcome. My rankings may offer a different insight than website data since they use statistical evidence which brings down the win rate of harder to play decks, despite them being clearly stronger than others when played at a high level-- like Overheal and Sonya Rogue for example.
Can you actually provide an explanation why 80+ is a sufficient sample size without simply claiming that you have some superior understanding of statistics. You provided a number, so I would hope you can shed light on how you derived it.
This is probably the most accurate reflection of the meta with player skill being taken into consideration. Overheal Priest is seriously the best deck in the meta that most players cannot play correctly.
Not sure you want an answer since your post wreaks of mockery (I guess they got you too with the laughably unscientific authority fallacy of “you’re not allowed to speak because he is top 20”) but in case you do,
it depends on the variance even if it’s general convention that 40 to 80 samples is where it starts being relatively reliable.
This game appears to follow similar convention since hsguru stats start stabilizing after ~60 samples.