It does not prove what is polarized or isn’t. Silly.
I wish someone would simply poll players.
It would not be conclusive but it would be as equally suggestive as this is.
Polling is fine but it would answer a different question.
You absolutely can, since polarization has mathematical signature in a population of data that can be measured, quantitatively.
Imagine a hypothetical deck with a normal distribution of win rates versus a 100 other decks, centered around a 50% win rate. Now imagine making a histogram of that data
(I.e.
On the x axis is a series of 5% ranges of win rates (e.g. 45-50%) win rate, on the y axis is frequency of matchups that fall in that range)
In a non-polarizing deck, most of those win rates would fall near 50%, so you’d have a single, bell shaped curve (if you draw a line over the tops of the histogram). With rare matchup in the 75% win/loss area.
A polarizing deck would have a u shaped curve, with few matchups around 50% and many more around 75-80 or 20-25. It could have this distribution even if it has a 50% overall win rate. Just to drive home the fact this isn’t about power, a unipolar distribution would indicate a deck is very bad or very good (depending on which end of the curve).
I also believe that players who abandon matches would suggest far more accurately which decks are the most polarizing.
Friend, by any chance would you happen to be upset that Ping Mage is listed among the most polarizing decks in the opening post?
Who says I am upset? I have laughed through most of this conversation. I think the data proves nothing.
I think people generally care about winning decks with a polarizing winrate rather than 40% winrate decks that lose against a lot, dramatically different though. Does anyone care if a deck randomly loses 10-45% of the time and lose sleep over it?
Besides priest i guess and legitimate second nerf quest mage. I still think the quest mage nerf while people still worry if slow 5 mana cap mana at 20 cards are gonna see play in sw if people are worried about instantly dying on turn 6 is a unvalid complaint though. 49.94% previous 1k legend winrates or not aside.
Any deck that gets a 49.94% legend winrate still usually has 60% winrate fodder lobby climbing lists for hsreplay, since even pre nerf quest mage still had a couple #100 players in it for a while during the lock and shaman time and still a bit after.
No one said you were upset. It was a question, not a statement.
I think this entire thread is dumb. Go play Word Chess with someone else.
When I want to know what is the most polarizing deck, I don’t have to use mathematics.
It’s Quest Mage. Even now, and that isn’t up for debate.
Have a good evening.
This is too broad a statement to be meaningful. The analysis is fine in terms of answering the question of whether a given deck has polarized matchups. As to what the impacts of polarization are, well that’s a whole set of questions you’d need to frame properly to understand.
For example - how much do people enjoy polarized matches, what % of matches are polarized, etc. To answer those sorts of questions you really need to have access to things like #games played, active players, and player spend, as those are true indicators of satisfaction (polling can provide insights, but has significant flaws and shouldn’t be relied upon).
I beg to differ. When I see people complaining about decks with winrates over 50%, I don’t see them complaining about how polarizing they are — I see them complaining using completely different arguments. Seems to me the “but it’s polarizing” line only comes up in response to a claim, supported with winrate data, that the target of their hatred is in fact Tier 3 or lower.
Or at the very least, attacking a deck archetype for its polarization tends to have maximum rhetorical power when attacks accusing it of raw power have already been soundly defeated.
Might I recommend using VS instead of HSReplay?
It is well known that VS is more accurate than HSReplay for several reasons.
When Blizzard released their data once, it way more closely resembled the data coming from VS rather than HSReplay.
Just a suggestion to see how that data compares to this.
Also, what do you mean by this:
How did you calculate this?
Because, while Quest Paladin may not be popular, it shouldn’t be “weighted” in a favorable fashion just because it’s not being popular, if that’s what you did. If Quest Paladin loses to a popular deck 80% of the time, the data should be 80% and not be toyed with due to popularity I would think.
Perhaps it would have been better to say “Polarization means very little for decks with losing winrates.” I could see how a deck that’s Tier 2 or even worse Tier 1 could have some negative consequences, because I can see people running those decks because they’re trying to be competitive on the ladder. In contrast, a deck with a losing winrate is not generally one people gravitate to if their aim is competitive efficacy. If someone is piloting such a deck, either they’re misguided about the deck’s viability, or they play it because it’s fun (or both). It doesn’t make sense to say that the pilot is suffering from the unfun effects of polarization when they’re deliberately playing that deck over more competitive alternatives for fun factor.
As far as Tier 1 decks having a problem if they’re polarizing, I still think it’s mostly a moot point — they arguably deserve a whack with the nerf bat just because they’re Tier 1. I do believe that good nerfs should make the targeted archetype’s good matchups more even instead of making its bad matchups worse, and it would take a combination of design skill and metagame knowledge to make that happen as planned.
That said, I still don’t think polarization is a reason to enact balance changes at all. I just think that it can be a consideration in how to nerf.
Also, wouldn’t your data favor really, really bad decks? If a deck is at, say, 30% win rate because it’s trash…it would end up showing as “more polarizing” than say a deck that has a 65% win rate…which isn’t exactly fair.
We want to gauge good decks, not really bad decks.
I think that’s why your list is full of really bad decks…aka Quest Paladin, Nzoth Shaman, Ping Mage.
Of course those decks are polarizing…they suck, they are their own worst enemy. That’s not exactly fair.
Which decks do people point to as problems for polarization, OG Caverns rogue (Kill before turn 5, or die, turned meta into kill before quest or die.
Quest mage (Most successful decks are aggro, nearly any deck archetype that went slower like quest shaman which was decent speed and decent endgame, but just the 1-2 turns slower but with better anti aggro tools to beat aggro, but die to mage and handlock hyper fatigue 70% of the time or just out late gamed if it couldn’t kill the game before raise dead 2-6+ 8/6 + 8/8 + 8/8 Bodies came out was another.
I think this problem feels like it’s more trying to redefine a definition behind why people disliked a deck. “I feel like quest mage/quest rogue/ Wild seedlock is unfun, because it’s polarizing to have 25% wr or 75% loss matchups entirely around a deck, at mulligan, because it feels like my choices didn’t even / don’t matter unless they topdeck badly or screw up the deck” isn’t a good feeling.
Meanwhile i’ve yet to see someone complain about some unknown Nzoth shaman. I feel like this is all trying to redefine quest mage is bad to try and redefine a definition. Not going to lie, it kinda reminds me a ton about some religious apologetics for a group that say, might believes that all indians had fought a war in the garden of Eden, which was also declared to be in Southern Missouri, near a founder’s nearby childhood house. Without leaving any records.
And when asked why millions of supposed indians died without a trace, in their backyard, without leaving a mark when we have incidents of single 10 people battles leaving behind hundred’s of arrowheads, bones and remains, with the founder’s book depicting the indians fighting Horseback with roman chariots [Horses were quickly adopted, but not in America Before Columbus. Much less 14 centuries earlier in 0 BC]. They’d just shrug and go. “The book says it is true, therefore it is true, and what’s important isn’t the evidence but the conclusion. It’s more important to believe the book is true, then know if it’s true. Religious science should be taught side by side as a serious alternative to biology as part of a biology degree, even if the part’s might not make sense.”
Followed by shrugging, and then when you ask about why there weren’t remains, it would jump up into “Our lord wanted our faith to be strong, so he manually removed every single piece of evidence of the north american arrowhead fights and personally carried them away to space, by hand, in the 14th century so Columbus and geologists couldn’t discover them. Then he created dinosaur fossils for the devil to plant within the earth to deny creationism.”
Just not gonna lie, a lot of this seems to be conclusion first (quest mage is good), with lots of confusing smoke, psuedo babble, all just to say Quest mage was good, followed by other quest mages each saying ‘OF course it was polarized’ and 'no it wasn’t!" in one giant, circular headache.
I’ll be honest, there’s nothing wrong trying to protect a deck we play, (we all have to otherwise whenever i get fair and reasonable i get my deck nerfed with quest shaman and 1 mana hp raza that killed the deck because i found and publically shared Spawn of shadows combos and growing better wild pools, followed by the nerf a week later with blizzard citing potential combo concerns and growing better wild pools)
But a lot of this just feels a lot similar.
Science traditionally starts, The data first proves the conclusion.
Circular reasoning usually starts, I will start with the conclusion, and THEN find evidence to support my conclusion.
Just in terms of like, Creationism (I know this is true, because it is true, because i know it is true. → Repeat)
Vs like science being
(I have a hypothesis the earth was created. Oops, here’s a fossil disproving it!
→ I reform my hypothesis, God created dinosaurs as well
→ oops, it conflicts with genesis!
→ I reform my hypothesis, based on what we can see, that dinosaurs work with the history of earth. Rework data into new hypothesis, BAM, science.
Psuedoscience / Circular reasoning:
“I believe god had a indian war in my backyard, and had the garden of Eden in my backyard in Missouri. And the reason why there aren’t bones is god must have also stolen all the bones, arrowheads, and then hid it to strengthen our faith. This data agrees with me, therefore my conclusion is true.”
Not to say that maybe there aren’t some polarizing decks nobody plays or 0.01% of players know about, but yeah, idk, it just feels hard not to feel like all these data based arguments are putting the data first, or finding a conclusion first, then fiddling with stuff after to find a desired conclusion from random outputs.
I said already. But let’s break down in excruciating detail what “weighted average by the opposing deck’s general popularity” means.
You said Quest Paladin, so let’s say I’m calculating Quest Paladin, T1KL.
First opposing deck: Face Hunter.
Matchup winrate is estimated at 37.4%, because there isn’t published data for T1KL so I use the general Legend matchup in lieu of better data.
That 12.6% from 50%, so that’s the polarization for that match.
Now that 12.6% is put into a weighted average where the weight is not the number of Quest Paladin vs Face Hunter games recorded at T1KL — indeed, HSReplay doesn’t release those numbers if they’re below a certain threshold — but instead by Face Hunter’s general popularity in T1KL. In this case, the weight is 16000 games, because a lot of Face Hunter in T1KL. So in a column on the spreadsheet, 0.126×16000, and in another column 16000. Repeat for all other opposing archetypes (unless no matchup data whatsoever). Sum the former column and divide by sum of latter column.
So doesn’t that just prove that polarization favors bad decks? The worse a deck is, the more likely it is to show up on your list of “polarizing decks” which doesn’t make sense. This is why your list is filled with terrible decks.
If I made a deck that had the worst minions in the game and it showed up on your report, it would be listed as the most polarizing deck in the meta…which is really bad.
It should be listed as “least polarizing” because it’s not doing anything positive.
That was indeed a factor. But as I posted above…
As you said, of course decks that suck are polarizing. But if I give decks that suck a handicap, I’m minimizing the concern that decks with winrates below 50% can be problematic if they’re polarizing. And the reason I began this research in the first place was because I wanted to get down to the factual basis, if any, behind such claims, as made by players who post here other than myself.
I was clear enough with the definitions that you caught on, so I think I did my due diligence here. And if I did add the handicap, well, that’d make Quest Mage less polarizing and that wouldn’t be fair to their argument, would it?
The problem is you are including decks that are like…Tier 4/5/6…lol
Your report should only be Tier 3+ decks. Everything else is muddying the waters. Terrible decks will always show up on your list more if you include bottom Tier decks because it’s easier to be 35% win rate than it is to be 65% win rate.
This is why I suggest you use VS reporting. It’s more accurate, and you get the Tier list that’s more accurate.
Quest Paladin, N’zoth Shaman, Ping Mage etc shouldn’t even be looked at because they are like Tier 5.
Just compare Tier 3 and above to each other. No need to muddy the waters here.
If the sole consideration is balance changes, then I agree that polarization alone is not a reason to nerf. Nerfing is just one tool to address player’s enjoyment.
In broader terms, it’s well established that ‘polarization’, or ease of experiencing a ‘win event’ vs experiencing a ‘loss event’ affects enjoyment in gaming. The rate of wins vs losses affects players’ enjoyment differently, and is mostly dependent on the experience of the player (new players are happier when they win easily, experienced players are happier when they feel they played well). Generally speaking, minimising polarisation as defined here is a good thing.