Hearthpwn

Battletag: Lovefall#2255

Region: EU

Yes u first

N/A

Battletag: Lovefall#2255

Region: EU

Trade Only?: Yes, you go first!

Done

My run to legend from the last month.

I don't see my winning streak being followed by any losing streak.

If anything, after climbing to legend, my win rate only improved.

The four pictures show:
- my last 22 games before reaching legend, with 19 wins (I think)
- my win rate before reaching legend (67%)
- my win rate at the end of the month (68%)
- my win rate at legend rank only (71%)

So whatever theory you have of how the new system is punishing you, it seems to elude me, so maybe, just maybe, the fault doesn't reside in the system, but in your approach to understand it.

I really hate anecdotical examples, but since I saw a lot of people posting them in this thread, I decided to also post my anecdotical example from this month... it's really funny how it just confirms my general statement and proves that, at least in my case, there was no voodoo too break my win series or stop me right at the "last boss": the road to legend - 66 games (66.67% win rate), and in last 22 games (from rank 5 to legend) a win rate of 86.36% (19 out of 22 wins).

https://imgur.com/a/yYbk2vk

Quote from Macinstosh >>

 This whole conclusion is an example of someone who doesn't know how to read the data or how to ask the proper questions. The data shows that you are being kept at a 50%-61% win-rate, excluding seven outliers, one of which is at such a small sample size the data becomes redundant. This is not proof that "you are being kept at a 50% win-rate", but that you are either garbage at the game or your match-ups are being doctored to bring you closer to a 50% win-rate. I.E. you're being matched with an unfavorable opponent/class/deck in order to bring your win rate more in-line with the preferred percentage. If this wasn't the case, many players in legendary would have win-rates in the 75%-90% range. 

By definition, Hearthstone is a zero-sum game, meaning that the average win rate across all players will always be 50% (actually a bit less because of the draws, but they're so rare, that we can safely ignore them), so I really fail to see how you can say that someone who has, by definition, higher than average rating (since 61% which is my average is definitely higher [I would dare to say even way higher] than 50% which is the overall average across all players)...

Also, the reason why you don't see people with 90% or so win rate is not that the system is cheating you, but just because you are matched with people with a similar record as yours and you eventually reach a point where you face only people with a similar skillset and you simply can't continue to win much more than 50%. Initially, at the start of the month, (especially in the old system) when everyone was resetted at the same level, it was typical for the pros to climb very fast, many times with a very high win rate, 'till reaching legend (or high legend), but once you get there you'll face people of similar rank/skill and you simply can no longer maintain such a win rate.

I'm really mindblown how you can't grasp such a simple concept which you see everywhere, even in real life... think at a tennis competition (or any other sport)... the pros usually have easy wins in the first few rounds where they face weaker opponents, but as they progress farther in the tournament they just start facing harder opponents (which advanced there too) and they can't keep winning all the time...

Quote from Hooghout >>

 Working hard to celebrate Blizzards excellent A.I. ability to make it all ook like non-rigging. Reminding me of the flatlanders 'proving'  that flat is all there is. Anyways the only credible way is an independent investigation into Blizzard MMR/RNG mechanics. And with independent I mean independent.

 When you have 24 months worth of data of both ranked and arena games all converging to about the same (+/-3%) win rate, which is quite different than the average 50%, I think it's quite obvious that the system is not rigged to keep us at 50% win rate (because otherwise, it would do a terrible job letting me constantly score over 57% win rate.

Done

Quote from Anatak89 >>

 

 It's probably mainly because I played wild and I assume you played on standard.

 Ah yep that would be it exactly haha.  So I do still have to quit my job if I want to strive for that top 10 standard finish I guess =P

 Edit my previous answer to you with a bit more context, so you may want to recheck it, but yeah, in standard competition at high legend is definitely fiercer.

I usually entered legend ranks (and start climbing from) at about 2000 place in wild and ended it at about 100 place, while in standard (when I've done that instead) I usually entered at 10000 place, and I think my best ranking was somewhere around 490th place.

Quite a bit different. :P

Quote from DMH_Meta >>

dude even streamers that make to legend each time in like the first 24 to 48 hours of the season claim that its rigged af and for ppl that only play like 2 hours per day its very hard to reach legend and naturally they get stuck  in the pre-legend ranks aka rank 5 to 1 diamond.

 

As previously said (and proved my logs/screenshots) I almost never played more than one hour a day (six games per day) and that didn't prevent me from reaching legend, sometimes even high legend almost every month, so it's definitely not true that you need to play for hours and hours to reach it. Six games (one hour per day) with a 55% win rate is enough to reach legend every month.

Quote from Anatak89 >>

I'm amazed you finished top 5 legend that one month with a 83-41 record.  My highest finish was 800 legend, and my record that month was similar to that.  I know I went from rank 5 to legend (in the old system) in 60 games, then continued to rake in legend, and I think the highest I hit was 742.  I might need to track down my stats that month (as I use that decktracker program as well), but I remember thinking to myself "wow, winning 2/3 of my total games and I still can't even break top 500."  I just assumed I didn't play enough games to grind to the very top, but now I just don't know haha.

 It's probably mainly because I played wild and I assume you played on standard.

Another reason is probably the fact that I was only playing six games a day, so I didn't face too many great players early on because most of them probably got on legend very early on while I took my time, so I manage to got to legend in only 47 games (compared with your 60) with a 34 - 13 record (72.34% win rate) as you can see from the print screen (total and legend stats; to get the non-legend stats just subtract the numbers).

Quote from Highwayman37 >>

"Q: Why are there two matchmaking pools?

A: Since each player starts at the bottom of the ladder at start, we want fair matches for everyone (aka. win-rate close to 50%)."

First, nice work on trying to prove your point but the game developers are on record saying they want a win-rate close to 50% for everyone.  They didn't elaborate on what that means (perhaps they will at some point in the future) but it's quite an odd statement.  

 

 That is something all game developers want.

If you win too much you get bored, if you lose to much you get frustrated, in order to be entertained you need to feel both the challenge and the achievement and that's happening when you have a lot of close calls and you end up anywhere around 50% win rate.

The thing that I'm saying is that you don't need to rig the system to achieve that, you just have to try to find opponents with similar skills.

It is happening in the swiss tournaments by matching you with someone with a similar record with you.
It is happening in the arenas by matching you with someone with a similar win rate with you.
It is happening in the battlegrounds by matching you with people with similar MMR with you.
It was happening in the old rank system by matching you with someone with a similar rank with you.
It is happening the new rank system, in the casual games, and in the tavern brawls my matching you with someone with a similar MMR with you.

If you climb too much, you will get tougher opponents that will push you back down where you belong.
If you drop too much, you will get easier opponents that will push you back up where you belong.
In the end, in all game modes, after an initial climb/drop, to about your level, you will maintain a more or less steady win rate, but not because the system is rigged (because in this case, it will keep everyone at 50%, not different people at different levels), but just because it tries to match you with "a worthy opponent".

Again, I think that if the system would be indeed rigged to punish you for win streaks and force everyone to about 50%, then in all my records (24 months of tracked ranked games [about 5000 games] and arena runs [about 500 runs, which average to about 4000 games]) I wouldn't show consistently about the same win rate, which is different enough than 50% to prove that is not just the error margin (not having even a single month with a lower than 55% win rate in neither ranked or arena). And, remember, I'm just a regular player, not a pro player or anything like that.

Quote from Shadowrisen >>

 

I guess to be the ultimate devil's advocate, since I have certainly shit on a lot of "rigged" posters, in order for your data to really point the direction you'd want, you'd have to have tracked the deck type you played, and now that I'm thinking of it, it may have been necessary to just play one deck each month to truly reach what some of these people are claiming and disprove.  I don't know; I'd have to think about it and try to figure out exactly how the experiment would be framed.

In any case, kudos.

 Hey and thanks for the nice words!

If you'd check the print screens you'd see that deck types (or at least classes) were already tracked.

Also, while my typical behaviour was to switch decks on a daily basis to match them with my daily quest (just because I had limited time I could put into the game), there were also months when I did exactly that and sticked with one deck for the whole month (or at least for most of the month) as it follows:

December 2017: 61% win rate (83-52) - all decks played with a single deck (aggro paladin)
January 2018: 68% win rate (57-27) - played a variety of decks
February 2018: 68% win rate (61-29) - played a variety of decks
March 2018: 62% win rate (92-56) -  played a variety of decks, but 75% of games were with a single deck (aggro paladin) with a 63% win rate (overall +1%)
April 2018: 58% win rate (104-76) - played a variety of decks
May 2018: 57% win rate (79-59) - played a variety of decks
June 2018: 67% win rate (85-41) - played a variety of decks
January 2019: 66% win rate (129-65) - all decks played with a single deck (shaman)
February 2019: 57% win rate (112-85) - played a variety of decks
March 2019: 61% win rate (114-72) -  played a variety of decks, but 76% of games were with a single deck (shaman) with a 59% win rate (overall -2%)
April 2019: 61% win rate (81-51) - played a variety of decks
May 2019: 57% win rate (79-59) - played a variety of decks
June 2019: 74% win rate (34-12) - all decks played with a single deck (shaman)
July 2019: 67% (8-4) - ignored because too few games were played
August 2019: 57% (134-100) - played a variety of decks
September 2019: 61% (17-11) - ignored because too few games were played
October 2019: 57% (79-59) -  played a variety of decks, but 87% of games were with a single deck (mage) with a 54% win rate (overall -5%)
November 2019: 57% (75-57) - all decks played with a single deck (mage)
December 2019: 57% (75-57) -  played a variety of decks, but 77% of games were with a single deck (mage) with a 60% win rate (overall +3%)
January 2020: 57% (35-26) - played a variety of decks
February 2020: 58% (50-36) - played a variety of decks
March 2020: 61% (35-22) - played a variety of decks
April 2020: 57% (103-77) - played a variety of decks

As you can for yourself there were:
- four months during which I played a single deck, with the average win rate (ignoring the big outlier I previously told you about when I end up with a top 5 legend finish) was 61.33%
- four months during which I played a variety of decks, but one of the decks/classes were highly prevalent, with the average win rate of 59.25%
- 13 months during which I played a variety of decks, with the average win rate of 60.23%
- and, from the original post, the average for all 21 months was 60.1%

As you can see, the variations depending my playstyle were really minimal, which I think reinforces my conclusion that, in the end, it's not so much the deck you're playing, but your technical skill as a player and that there is definitely no compensation/rigging mechanism to keep you at 50%.

As a fun coincidence (or not), my arena's average win rate on my last 300 runs is also very close to my ranking games win rate (61.5%, with an average of 4.92 wins per run) - if required I can provide the exact distribution too.

Quote from Mugen8 >>

I also think people underestimate how hard it is to account for the metagame and deck selection parts of the game. While I haven't had a competitive run on the HS ladder in a year or so, and never played it 'really' competitively (aiming for high legend would be my definition of this), I did play MTG competitively back in the day - and for me, the hardest part of any tournament was deck selection. I'm sure some people find this part easier than piloting, but compared to piloting a deck on the day, actually trying to figure out the best deck to take when there were ~10 viable metagame options was the real difficulty.

And my point is that sometimes you'd choose what you thought was the right deck for whatever reason/s you had, and just come up against a few bad matchups and fail to make day 2 or the top 8. Doesn't mean the pairings were rigged - you just rolled some bad matchups when you may have been pretty close to breaking through (the analogy being that you might be getting close to legend, but then you hit some bad matchups and it can be super frustrating, especially if it seemed like you made the right metagame call up to that point).

 While you are absolutely right on what you're saying, I think you're overcomplicating things.

While being able to accurately predict the current local meta could definitely boost your win rate by a lot, it is definitely not needed in order to achieve a positive win rate and/or reach legend; as I said in my personal example I never played the top deck, but I just picked one of the good decks for the class that I had the daily quest and I would say that it worked quite decent for me, simply on the premises that I was a somehow better technical player than the average player.

And, in some cases, trying to read the meta and keep switching decks can become detrimental even if you manage to do the reading, for the simple reason that (almost) any deck requires a "training" period before you can say you fully understand its playstyle, potential, and matchmaking. Actually, for anyone except the top players that are playing at high legend meta, I would recommend to just pick a good (tier 1) deck and stick with it for the whole (most of the) month since they will probably be able to win more from the fact that they'll get to master the deck.