I’m not here to rant about anything, I’m just here to state my honest opinion as a top legend player in NA.
To help support my opinion, I am usually in top 50 legend, sometimes top 20 and top 10 as well.
I’ll start by saying that I think Hearthstone takes tremendous amounts of practice, memorization and skill to climb to the highest ranks. However, after a certain point and especially in certain metas-- skill is often overshadowed by luck.
I think that at the highest level of play where both players make the most optimal plays every turn and think of every possible scenario, the strongest determining factor becomes card draw, randomly generated cards/ effects and random scenarios in board based interactions such as hits.
For example yesterday vs NoHandsGamer where he played turn 6 Tsunami, all 4 of his water elementals lost a 50/50 and hit my face instead of the stealthed Hidden Gem I had on my board (a 6.25% chance interaction) and it cost him the game. In this example there is nothing either of us could do to determine the outcome of that game other than he either loses the 6.25% chance or I win it.
Another example, I started playing on the EU server recently with fresh MMR and so far I am 3-3, or in other words I am at a 50% win rate against completely new players with unoptimized decks.
My first loss was to a Mage who played Yogg In The Box which then full cleared my board, dealt 5 damage to my face and also casted Pack The House to summon him a full board of minions-- one of which included AFKay to further buff his entire randomly generated board by +2/+2, costing me the game out of nowhere.
Yes, we all know that the game is dictated by RNG I am not saying anything new, but I just wanted to offer my take on how even in the case of a top player, the game is still primarily decided by RNG.
In being very good at the game you can mitigate bad RNG by playing for those outs, but at the end of the day anything can happen no matter how many precautions you take in Hearthstone.
Hearthstone’s complexity isn’t difficult to understand.
It’s just time-consuming due to the overwhelming number of possible plays in each game. Instead of adding depth, it feels unnecessarily convoluted, which doesn’t make it more intricate or interesting, just a mess.
It’s trying to do too much for what’s supposed to be a casual card game.
Discover has ruined the game, along with cheap card draw and game-ending effects packed into single cards without any steps or setup in between.
The OP does the mistake to not mention at all the factor of time (i.e. the amount of games played). If you played only 3 games a day then the RNG factor is ABSOLUTELY BRUTAL; the skill varies between most 2 matched-players but the RNG is often bigger than the difference of their skill on a single game; given enough games then the RNG factor diminishes.
The Mathematics of it are very simple; the RNG factor is COMMON between 2 players in a match because even if 1 of the decks is more RNG then the opponent is also sharing the same game but their skill DIFFERS (especially if their MMR is not identical and especially if they don’t play too much (unrefined MMR); hence RNG is a zero SUM game eventually.
As you said previously, the higher the skill, the higher the RNG factor
In the long run, though, continuously making better decisions SHOULD result in ranking up (although I don’t know for sure, being stuck at around 200 for 6+ months xD)
There are also complications due to deck strength vs deck mastery/skill
Some top 20 players, like Norwis, keep spamming 1 deck, so it’s safe to say they’ve mastered it, but the deck isn’t really top tier, so he’s struggling nevertheless
The RNG is the same no matter the skill. You probably want to express another concept, that if you play a lot then your MMR will be very refined(stable) so you’ll be playing with very similarly skilled people; hence the RNG will decide a lot of games because your skills will be similar between opponents; even then though given a lot of games: the RNG becomes a zero sum game because it’s a common variable between opponents in 1 game when the skill is not a common variable between opponents in 1 game (even when there are small differences if the MMR is very stable).
I think he meant that in a match between 2 highly skilled players, RNG becomes the biggest determining factor in the outcome of the match considering both players don’t make mistakes. He was referencing that point which I made in the post. The same point you are making now too. In other words, we are all saying the same thing lol.
Their level of skill does not affect the RNG; if lower skilled players do mistakes then that just means they have …lower skill; they’re just the same RNG matches with players on different skills.
It’s more improtant that if you play for a hours a day in a top 20 then you play with almost exactly the same skilled people so skill is closer to a zero sum game than others.
But RNG is even MORE of a zero sum game because it’s 100% shared for players in the same 1 match so after a lot of games the skill shows because it’s not common.
I don’t understand how you keep saying the exact same thing as the point I have made 3 times now and Altair has said as well. We are literally saying that, we are not saying that skill effects rng.
Not sure why you think you said the same thing. You said that high skill makes RNG a bigger deciding factor which means that low skill makes RNG a smaller deciding factor.
No it does not; low skill keeps RNG the same factor; if 2 players have the same low skill: RNG still decides similarly; they both do mistakes; it’s not like one does fewer mistakes.
You are kind of arguing with yourself at this point because literally nobody said low skill makes RNG a smaller factor other than you. I didn’t say that, nobody said that. Altair was just quoting what I said in the post in his own words, not saying that low skill is lower RNG. Literally to quote myself I said.
Anyways this is getting extremely annoying now. Lets just stop.
I’m not annoyed at all myself. I was answering to a clear statement that “the higher the skill the higher the factor of RNG” which is false. It seems there’s cognitive dissonance there if you find that agitating.
TL;DR The nature of probability dictates that any game of both skill and luck will scale very poorly as luck’s impact is allowed to grow unchecked. For details and a precise quantification of this statement read on.
If you understand basic concepts in probability, what we colloquially call RNG is usually called variance, and “skill”, in and of itself, is not a concept that can be readily described in probability theory’s nomenclature.
The measure of “skill” via proxies such as winrate over a period of time, however, is something that can be thought of as a statistic, and fits nicely into this line of thinking. Note the difference. One is some loosely defined concept called “skill,” the other is the measure of what (we assume) can be used to quantify “skill” via something that can be measured, such as winrate.
So what’s the point of all this? Well, sparing you the intro to probability course, for any random variable with finite variance, we can assert the law of large numbers (CLT) holds, allowing us to apply a lot of powerful concepts from probability theory developed over centuries by mathematicians.
Here are the key takeaways:
The standard error gives us a means to quantify the error we expect on your measured statistic, that is, in this case, how accurate does the observed winrate match a “true” or population winrate.
The standard error is defined as sqrt(sigma2) / sqrt(n), where n is the number of samples and sigma2 is the population’s variance.
Holding variance (colloquially known as RNG) constant, to halve the error in your estimate of your actual winrate, you need to play four times as many games.
This is a quadratic relationship and scales extremely poorly in theory and in practice.
Even small increases in variance (i.e. RNG) will require players to play many more games to achieve the same level of confidence in their true winrate being reflected in the raw data they collect.
We all have finite windows to play the game, at the very most you have one month, therefore, Blizzard should be very cognizant of this effect. They appear to be not.
What you are saying makes sense. After about top 20, the total quantity of games played becomes extremely important as you are essentially tossing a coin over and over on whether you win or lose.
The players with the most games played at this rank will often be closer to rank 1 as their rank is a closer reflection of their skill than the players with a higher overall win rate and less games played, as any losses to RNG are much more impactful to them.
The amount of skill allowed in years past was a lot higher than in the present… mostly because the amount of RNG in years past was far less than the amount of RNG in times present. Not to mention the mana cheating cards in this game have gone off the deep end.
It’s an incomplete summary of what you said, because I was too lazy to scroll to find the direct quote xD My bad
He was right to correct me, although it doesn’t matter. That person goes around “correcting” people all the time, even when those people are experts in what that person is trying to correct them at and is generally a very toxic user, so I suggest to everyone to put him on ignore.
This behavior should be taught in psychoanalysis schools. I was right to correct him but since he is supposedly an expert I should be on ignore.
Self awareness zero.
PS I like how yellovvsnovv analyzes how standard deviation and standard error leads us to need a much higher sample size to reduce the perceived standard error, and Qwark reads whatever he wants to hear again, “yeah it’s about the top 20 that makes it important”; it was never about your rank; you (just like Altair) can’t stop bragging about your rank and you can’t possibly accept those concepts are the same in all ranks because that would be too mundane for “experts”.