VS Data Reaper #303: Shaman Supremacy Edition

Multiplayer Discussion
82

Im not sure where you got that I said shaman is tier 7. I don’t see rainbow shaman on the ladder. I see a lot of big shaman which makes sense because its probably a better deck. I also said that nostalgia shaman is probably a better deck than rainbow shaman as well (although I dislike the list on the VS site)

I mean I experimented with the deck back in july - Shamans, how are we?

My experimental deck was obviously different than the deck they have posted today, but the cards were different back then too and a weapon OTK deck isn’t that impressive then or now in my opinion let along the best deck in the game.

83

Schyla wins this one. Your homebrew doesn’t count.

84

The version from the previous report didn’t have Greedy Partners in it - that’s the version I played and that one was still tier 1, best deck according to them

Hence, this doesn’t work as an argument.

85

You still imply that you play the deck perfectly, which contradicts that collected stats show it high winrate; the same stats clearly indicate that the deck has a very high skill cap; that’s because it’s very unusual that it shows a HIGHER win rate the higher the rank.

It appears to be one of the decks that needs to consider MULTIPLE alternative plays; that’s because each round multiple spells it has are usable and good but one of them must be probabilistically the BEST or close to that; hard to 100% its skillcap.

86

I never implied that in my whole life and hearthstone career

I don’t even play sludgelock perfectly, or I’d be better than I am

So, no, not perfectly, but good enough to recognize if a deck is good or not

You don’t have as many options as it seems because most of your plays are combos, so you don’t get to just drop cards on curve and it call it a day

I think the greedy partner list should obviously be better, but that much better, no, it’s not possible.

And you’re forgetting that in both reports it had playrates of 0,28% and then 1,50% which is abysmally low when compared to other decklists and the sheer fact the sample is so low means it’s uncomparable to other decks.

The variance is too high and the fact that 10-2 is 83% winrate while 20-12 is 67,5% don’t help when you try to compare such a rarely played deck to one of the meta tyrants. That means that the lower the playrate of a deck, the higher its’ winrate is just because numbers work like that.

It’s absolutely unprofessional to put it in tier 1 let alone call it the best deck, but I’ve said this already multiple times and if it’s still not changing your opinion, I can just say “good luck” to anyone who picks up the deck and hopes to go positive win rate with it.

87

I’m not picking up the deck. I find it above my skill level; it’s not a matter of “brain power” only; it appears to be one of the decks that require very good knowledge and prediction of what the OPPONENT will do and I’m still at best average at that.

Albeit shaman is obviously the most interesting class right now; consider the most major advantage it has; absolutely nobody has any clue what variant of the 5+ you are lol…

88

Not to me personally, but it’s pretty fun, yeah

That’s true, but you also have the same situation with Druids, DK-s and Warlocks, so no obvious benefits while picking it up over one of those classes

In fact, you’re probably much better off picking any of the Warlock decks instead

That works for any deck, lol, so if you’d like to play this one and can afford it, give it a spin

As I said, it’s fun, but that’s where the advantages end

89

I said “very good” so I implied it’s especially true for this kind of deck. E.g. a flood paladin or a pain look will usually do its thing anyway; even a control warrior is mostly doing its thing; the tools for each situation are more clear.

These decks with multiple spells that work in various situations similarly are more complex; a lot of “good” choices but one is theoretically the “best” and the best players leverage that; and they escape Devs’ early nerds.

90

Are you talking about that idiotic magicdeck that recrycles cards endlessly and throws you a fireball (8) strength every time? This deck ignores the timer…You just sit waiting for the stupid recycle thing to end…IT’s got so many secrets that you can’t interact with it eventually. I must have spent 30 minutes in that game. I had the b*stard down to 2 points but everything gets removed, stealth, you can’t hit the hero…ridiculous deck.

91

Us above? No just the rainbow shaman that has as a main win condition to spam the board with Razzle-Dazzler.

I forgot to mention “feeding” the Razzle-Dazzler (well) is not the most “beginners friendly” job in the game.

92

https://i.imgur.com/lM3Sd5q.png
You are utterly shameless.

2 Likes
93

Rogue is dead. :cry:

I stop playing at the right time.

94

You really need to stop projecting your own misdeeds on me

The screenshot you posted is a clear attempt at dishonest manipulation.

Let me prove it.

Open that same report and read whats under the “Cutlass Rogue” - its playrate is 3,88%, which is around 2.5 times more than rainbow shaman’s 1.50

I dont know why you chose to ignore archetype called Cutlass rogue and decided to post a screenshot with only weapon rogue’s playrate unless your only goal was a dishonest manipulation.

Maybe you could shed some light on that?

But yeah, so far its you who looks shameless to me.

Now, aside from your usual dishonest argumentation, let me be clear on something else which is important to understand where im coming from.

  1. unlike rainbow shaman, you will actually see cutlass rogue in top 1k every day multiple times. Its a real deck
  2. its good, it has obvious advantages that can be explicitly stated - it counters some of the meta tyrants and its the best deck rogue currently has
  3. it was just a prediction. If you are being honest, you know i dont put too much weight on any predictions. I just like trying.
  4. remember Multiperspectivism from last night? Well according to that philosophy, ignoring my own subjective experience and data would be a form of bias, one i will never allow myself happen to me.

Unfortunately, it appears you have your own intuition and experience tuned down to minimum, as life or toxic people in your life kept devaluing them (just like you are trying to do to mine) until you just gave up from it and decided to put double the weight on “objective” data, not realizing or ignoring the fact that if objective data dont align with your experience, then they are not quite objective at all.

1 Like
95

Oh. I made a mistake then, my bad

No, it’s Tier 3-4

No. I could tell it was pure gibberish. You’re supposed to filter out your own bias as much as possible, obviously

1 Like
96

And you’re consistently failing to do so.

Taking my experience into account is not a bias if that experience is interpreted correctly/reasonably.

However, I won’t even try to dwell on this any longer. Maybe another day.

But what is really, really pissing me off is the fact you, who prouds on being “objective” to the bone, fails to take into account an inherent behavior of numbers when interpreting statistics with high differences in their samples.

If a deck is played 70 000 games, I can have a score 35 500 : 34 500 and it will be a 50,71% deck.

If a deck is played 700 000 games, I can have a score 350 500: 349 500 and it will be a 50,07% deck, which is 0,64% less even though the difference between the number of wins and the number of losses is the same in both cases (1000)

Now, here, deck 2 is played 10 times more than deck 1. In the example of Rainbow shaman from last report, when it was played 1%, it was compard to decks with over 15% playrate, which means that the difference in winrates is even higher than 0,64% just because of the order of the magnitude of games played.

How can you compare apples and oranges and say the results of your comparison are objective data? You can’t. The only thing objective about that is how wrong it is.

I don’t need personal experience to take this into account because I’m educated enough to know this. But if I didn’t know this, I would at least have my own personal experience to rely on to know that the report is objectively wrong.

You, apparently, lack both

P.S. And that inherent property of numbers is the reason why popularity * winrate is a more objective measure of a deck - this way you cancel out the division with multiplication and congrats, you’re finally ready to compare the data on a similar scale, for a change.

Other, more tedious method would be to put them all on a logarithmic scale. Either way, these data NEED to be normalized before comparisons are made. The differences in winrates between the decks are way too low for you to ignore a 0,5-1% deviation.

97

Bruh, you were literally arguing for deliberately retaining one’s own bias, under the banner of “multiperspectivism,” probably because you watched some motivational manosphere propaganda that told you to listen to yourself.

This is more Terrence Howardism. This is only fascinating to you because of your mathematical ignorance. There is nothing there.

98

Wow. I’m just gonna quote this so everyone can see it.

Just wow.

1 Like
99

Or at least nothing interesting. If someone points to a plain concrete wall and yells “look!” most people are going to say “what?” as if nothing is there, rather than acknowledging the banal truth of the plain concrete wall. I do not literally mean there’s nothing there.

The point is that there are no interesting comparisons or contrasts between those two ratios.

1 Like
100

I just can’t believe you’re this blind. You’re the biggest victim of Dunning Kruger on this whole forum, and somehow you’re the first one to say the exact thing to others.

101

I can rant for a bit about HOW you’re wrong.

If you have a sample of 70,000 games with a 35,500 to 34,500 win to loss ratio, then yes, that sample implies a winrate of ~50.714%.

You then imagine this sample growing to 700,000 games with a ratio of 350,500 to 349,500, because you have a perverse obsession with the subtractive difference remaining constant. But what you’re ACTUALLY describing is taking a new sample. Sample 2, which has no overlap with Sample 1, is 630,000 games large with exactly 315,000 wins and 315,000 losses for a winrate of exactly 50%.

Because Sample 2 is larger than Sample 1, Sample 2 gives a better estimate of the truth. And indeed, after Sample 1 and Sample 2 were the same size at 70,000 each, Sample 2 then correctly predicted the future 8 times in a row.

The fact that Sample 1 and Sample 2 point to different winrates shows that they are NOT the same. You are trying to make them the same, when they’re not, because they subtract to the same number.

To make a long story short, just as your fascination with popularity × overall winrate is an abuse of multiplication, your fascination here is an abuse of subtraction. You are under the misapprehension that mathematics is a freeform activity like finger painting.

1 Like