Buddy, you just don't understand what they mean. It's more about how they match you with people. Once you find where you're supposed to be in terms of your ranking then you will likely have a win rate close to 50% with some variance because the game mode has a lot of luck involved. This is because you will simply be matched with similarly skilled players ("Worthy Opponents").
My rank is around 6500. But if I was able to play players with a ranking of 2500 I guarantee my win rate would be 80% or higher and then taper off as my ranking rises and close in on the 50% mark. Simply by playing against people closer to my level. Not by some forced algorithm.
As a developer, I DO understand what it means. The Luck stat itself is literally rigging the game for or against you depending on how well you're doing. Then the pay to get 2 more choices increases your ability to get a better hero by 50%. It's not nefarious, It's not an insult to the game or anything sneaky. It's something they themselves have discussed and a norm in auto-battlers. I don't though, understand how you guys can't understand it.
Paying to get a 50% chance to get a better hero increases your chance to win exponentially, which is rigging. Adding a luck stat to keep you at 50% is rigging. Even pushing you to play against people better or worse than you based on your w/l and rating is rigging. It's literally all forced algorithms.
I think you're not demonstrating properly the effect of having 4 rather than 2 choices.
You move from the average "best" hero in a 2 choice being the hero at the 67th percentile, to the hero at the 80th percentile. You go from an expected best pick being about the 23rd best hero to the 14th best hero. About 9 ranks better. That's good, but it's not really game-breaking.
In terms of not getting a very bad hero, you go from a 25% chance of two heroes in the bottom half of the distribution, down to 6.25%, which is really noticeable. You'll very rarely get screwed with the pass (which is really low gold cost, too).
However, let's look at the chance of getting a hero in the top 10% of heroes (that is, the top 7). For a 2-pick, you'll have a 19% chance (1 - [0.9 ^ 2]), so about 1 in 5 games you'll get a top-7 hero. For a 4-pick, this is about a 34% chance (1 - [0.9 ^ 4]). Once in every 3 games. Noticable, but not horrifying, and the gold cost is pretty easy for anyone who plays even a little.
There's real Pay-To-Win issues in some games. There are major abuse issues at Blizzard. Battlegrounds is not a game toxic with P2W.
Battlegrounds has no real rigging (some folks just don't understand how Elo and MMR work). There is real rigging in loot boxes/packs--things that actually cost people money.
This thread is the most whiny distraction I've ever seen.
High School math really needs to include some basic statistical literacy. Magical thinking and conspiratorial belief can be really dangerous to society, and a basic understanding of probability might be able to put a dent in at least some of it.
Uncommon things happen all the time, but there are all sorts of uncommon things. Think about poker, how just about any specific set of cards is really unlikely. But the odds of getting no more than a high card? About 50%. Half the time, you'll have a pair or better. If you've got a 7 card hand in poker, the chance of not getting a pair drops to just over 17%. All those unlikely things like flushes or straights or full houses? Odds are something is going to happen.
I was playing dragons the other day. Couldn't get an Eliza or a Kalecgos until my third triple for a 6* unit. Seems likely, right? But with about 13 or 14 minions in the pool, you only have around a 40% to get one of them, and a 60% chance to miss (approximately... the number of each 6* minion already accounted for from the pool would impact things, but this is close enough). Following a basic geometric distribution, it's not at all uncommon not to have a success until the third trial.
Ever get eliminated on a 5% lethal? Ever play Dungeons and Dragons and roll a natural 1?
//
So the OP said:
I have found that if you intentionally concede a bunch of games you suddenly start to get more favorable matches.
Well... that's exactly how MMR or Elo ought to work. Concede and you lose ranking and the matchmaking system will match you against weaker opponents. Being matched with weaker opponents means games are easier. Then you'll win games, and raise your Elo and face better players. Players better than you. Players who win.
Professor Arpad Elo came up with a really good system for ranking comparative strength. Using something like it for matchmaking is rather effective.
//
I guess what irks me is that there are actual times things are rigged. Slot machines, scratch tickets. They're rigged to have a fixed payout rate, but also to have small wins at small intervals to keep people addicted and playing. Loot boxes in games like Hearthstone (they're called "packs") and Overwatch and Heroes of the Storm are likewise designed to prey on some folks to extract money. "Collect them all" traps, "must act now" seasonal offers that exploit a fear of missing out (ever buy anything--from groceries to electronics to whatever--because the sale was about to expire?). There's actually nasty stuff going that doesn't involve rigging RNG in specific games, and could behave exactly as statistically described and still be exploitative.
But what probably isn't happening... is what the OP describes. The system almost surely isn't manipulating the fine details of individual fights. There just isn't a point. The combination of matchmaking and high RNG will do the trick well enough. Meanwhile, there's like no upside to doing it for BGs--the really flat purchase curve means it just doesn't matter.
And someone whining about losses when they're unable to understand probabilities just obscures the real issues.
//
One last point: if Bliz was rigging things, people would know. Someone has to program these games. We know that people have revealed a massive internal structure of abuse and exploitation of workers within the games industry--Blizzard included. And yet there haven't been leaks or whistleblowers about rigged RNG.
Because abuse is real, and rigged RNG isn't.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
⚙
Learn More
Artwork
Cosmetics
Related Cards
Card Pool
✕
✕
PopCard Settings
Click on the buttons to change the PopCard background.
Elements settings
Click on the button to hide or unhide popcard elements.
I think you're not demonstrating properly the effect of having 4 rather than 2 choices.
You move from the average "best" hero in a 2 choice being the hero at the 67th percentile, to the hero at the 80th percentile. You go from an expected best pick being about the 23rd best hero to the 14th best hero. About 9 ranks better. That's good, but it's not really game-breaking.
In terms of not getting a very bad hero, you go from a 25% chance of two heroes in the bottom half of the distribution, down to 6.25%, which is really noticeable. You'll very rarely get screwed with the pass (which is really low gold cost, too).
However, let's look at the chance of getting a hero in the top 10% of heroes (that is, the top 7). For a 2-pick, you'll have a 19% chance (1 - [0.9 ^ 2]), so about 1 in 5 games you'll get a top-7 hero. For a 4-pick, this is about a 34% chance (1 - [0.9 ^ 4]). Once in every 3 games. Noticable, but not horrifying, and the gold cost is pretty easy for anyone who plays even a little.
There's real Pay-To-Win issues in some games. There are major abuse issues at Blizzard. Battlegrounds is not a game toxic with P2W.
Battlegrounds has no real rigging (some folks just don't understand how Elo and MMR work). There is real rigging in loot boxes/packs--things that actually cost people money.
This thread is the most whiny distraction I've ever seen.
High School math really needs to include some basic statistical literacy. Magical thinking and conspiratorial belief can be really dangerous to society, and a basic understanding of probability might be able to put a dent in at least some of it.
Uncommon things happen all the time, but there are all sorts of uncommon things. Think about poker, how just about any specific set of cards is really unlikely. But the odds of getting no more than a high card? About 50%. Half the time, you'll have a pair or better. If you've got a 7 card hand in poker, the chance of not getting a pair drops to just over 17%. All those unlikely things like flushes or straights or full houses? Odds are something is going to happen.
I was playing dragons the other day. Couldn't get an Eliza or a Kalecgos until my third triple for a 6* unit. Seems likely, right? But with about 13 or 14 minions in the pool, you only have around a 40% to get one of them, and a 60% chance to miss (approximately... the number of each 6* minion already accounted for from the pool would impact things, but this is close enough). Following a basic geometric distribution, it's not at all uncommon not to have a success until the third trial.
Ever get eliminated on a 5% lethal? Ever play Dungeons and Dragons and roll a natural 1?
//
So the OP said:
Well... that's exactly how MMR or Elo ought to work. Concede and you lose ranking and the matchmaking system will match you against weaker opponents. Being matched with weaker opponents means games are easier. Then you'll win games, and raise your Elo and face better players. Players better than you. Players who win.
Professor Arpad Elo came up with a really good system for ranking comparative strength. Using something like it for matchmaking is rather effective.
//
I guess what irks me is that there are actual times things are rigged. Slot machines, scratch tickets. They're rigged to have a fixed payout rate, but also to have small wins at small intervals to keep people addicted and playing. Loot boxes in games like Hearthstone (they're called "packs") and Overwatch and Heroes of the Storm are likewise designed to prey on some folks to extract money. "Collect them all" traps, "must act now" seasonal offers that exploit a fear of missing out (ever buy anything--from groceries to electronics to whatever--because the sale was about to expire?). There's actually nasty stuff going that doesn't involve rigging RNG in specific games, and could behave exactly as statistically described and still be exploitative.
But what probably isn't happening... is what the OP describes. The system almost surely isn't manipulating the fine details of individual fights. There just isn't a point. The combination of matchmaking and high RNG will do the trick well enough. Meanwhile, there's like no upside to doing it for BGs--the really flat purchase curve means it just doesn't matter.
And someone whining about losses when they're unable to understand probabilities just obscures the real issues.
//
One last point: if Bliz was rigging things, people would know. Someone has to program these games. We know that people have revealed a massive internal structure of abuse and exploitation of workers within the games industry--Blizzard included. And yet there haven't been leaks or whistleblowers about rigged RNG.
Because abuse is real, and rigged RNG isn't.