![](data:image/svg+xml,<svg xmlns="http://www.w3.org/2000/svg" width="35" height="35"><rect fill-opacity="0"/></svg>){kind=small}
In January I wrote a post after long research how League of Legends is rigged. It wasn’t well received, because my claims were pretty outlandish and most people don’t bother to read charts when they can just say “lol conspiracy”. Why do I repost? Because my “outlandish” claims were patented by Activision and were reported by Rolling Stone. So my claim went from “Riot does this crazy thing that only I described” to “Riot uses Activision technology”. At this point it’s irrelevant if this usage is licensed or stolen or Activision stole and patented Riot technology, though the court porn would be delicious.
At first, let me summarize what the Activision patent declares:
- it purposefully matches a non-buyer with a someone who is expertly using a P2W item to make him envy the powers of the whale and spend money
- if the player is focusing on one type of gameplay (like sniper), they don’t just let a whale assassin pwn him, they send a whale sniper to make it clear that it’s not the class that’s op, but the item and make him buy the op rifle
- recent buyers are placed into games where they can devastate their opponents so they are happy about their purchases
OK, but how do I know if a player is a recent buyer in League of Legends without insider information (just having a skin means nothing, it can be old)? There is a very good way to guess it: someone uses a new champion. You can still see my old stats on third party sites like op.gg and see what champions I’ve used. You can also look for a champion and see my stats (or lack of) with it:
I’ve never played Aatrox, so if you see me play it, after you’ve seen such empty stats, you can assume that I just bought Aatrox.
If you check the stats of your teammates during pre-game, you can determine if they play a champion they haven’t played yet. Sure, they could win it or could be in their stables for ages, but most likely using a new champion means buying a new champion. My first test contained 27 games that I classified into four groups, based on the amount of new champion users – amount of average ones (see definition below) in the team:
That’s pretty convincing that the games are rigged and the matchmaker does what the patent says: give easy wins to paying players. Then I played 298 games and classified the teammates into 4 groups:
- New: using a champion he haven’t used before, or used only once, recently: likely a buyer
- Good: using a champion he has 60%+ winrate
- Bad: using a champion he has 40%- winrate
- Average: using a champion with 40-60% winrate
I’ve found that games with buyers are usually having good players too and these games have very high, 77% winrate:
This is obvious: buyers+boosters who carry them.
I’ve found that games with no buyers are practically always lost, but I’ve also figured out a way to turn it around and it shed a light to how exactly they rig: Instead of playing “properly”, I turned the laning phase upside down with forcing teamfights early on. The rigging works by matching the boosters with players they can defeat 1v1 (the average and bad ones), causing the infamous 0/10/0 at 12 mins Yasuo, while the buyer is facing a bad player who can’t farm him that bad. But if I don’t let them 1v1 in peace but force a 3v3, then I actually force a 3v2+buyer which won’t end well for them, as the buyer who is learning the basics of his champion will perform horribly in a messy teamfight:
I tested this exploit mode in a very aggressive manner: queued in as support and picked Warwick, smite and went for second jungler. Of course I lost almost all “easy win” and “fair” games. But I won most of the “Sure loss” games since everyone was running around as headless chicken and fights happened in random locations where my average teammates defeated the wallet warriors.
Finally, lets see the rest of the games, which had 53% winrate:
These have lot of buyers and bad players. There is two explanation for these: either these “buyers” were not buyers just someone who pulled and old champ out of the closet, or az unexpected champion pick. You see, the matchmaker selects players and lanes, but not champions. So they can assume, based on statistics that if they place Joe mid, he’ll likely play Katarina and wins, so they place him as booster. But Joe feels like playing LeBlanc today and he isn’t that good, so he can’t roflstomp.
Anyway, see it for yourself! Pull up the stats of your teammates and watch how the zero-experience, new champion teammate predicts victory. Or buy one yourself, jump into a ranked game without practice and see how easily you’ll win, despite having no idea what you’re doing.
So the Riot rigging does exactly what the Activision patent suggests:
- it places someone who has a champions for a long time (but not particularly good with it) in a lane against someone who plays their counter to produce 0/10/0 laning phase result, both to lose the game and to make the player buy the counter
- the laning mechanism guarantees the second point, ADC mostly faces ADC and not top
- recent buyers are placed into games where they win so they are happy about their purchases
I suggest not to play League of Legends. Or at least demand them to make a clear, binding statement that the matchmaker does not get purchase nor asset ownership info as input, just MMR and preferred lanes.
.
Update: I made a reddit link for this post and it got shadowbanned fast. It’s still there if you look by the link, but you can’t see it on /new:
If they’d consider it trolling, they’d just ban it openly and give me timeout from their subreddit. They know it’s true and have to be buried. Spread the word, link it, tweet it, contact journalists! We are talking about the single biggest seller video game that poses as an e-sport, but actually just a rigged moneygrab.
![](data:image/svg+xml,<svg xmlns="http://www.w3.org/2000/svg" width="42" height="42"><rect fill-opacity="0"/></svg>){kind=avatar}
> Or at least demand them to make a clear, binding statement that the matchmaker does not get purchase nor asset ownership info as input, just MMR and preferred lanes.
But they are a business. They’re doing it to make money, not to be some noble cause that lets people have fun at their expense. Of course the matchmaker will continue to get recent purchases as input as it’d be crazy not to do it.
LikeLike
@ST: ant it would be crazy for us to play the game where the outcome of the matches depends on spending. If enough people do that, they’ll either have to cave and make only money from vanity purchases (player wants pirate Katarina, regardless of winning) or change their business model to subscription.
LikeLike
Just as before, the reason i’m saying all of this is pointless is not because i disagree that the ladder is being less than honest, but because i honestly don’t believe it harms the game in any meaningful way.
Ladder is not the same as actual competitive game. It is not a gateway to competitive game and is rarely used as any sort of screening for player quality, except in most rudimentary of ways.
LikeLike
@Maxim: 99% of the players only play matchmaked games and not “e-sports”. These players have their matches rigged, the outcome pre-determined, based on paying. Do you think they are not harmed by it? Would you play a game where the match is already decided on loading screen?
LikeLike
Next month Riot will be handing out bonuses to the team that brilliantly leaked the Activision Patent as they count all the money from the players that bought new champions because of all the press.
It’s genius! Not ONLY do they avoid looking like the “bad guy” for “developing” the code themselves… but they take it to the bank too. What’s Activision going to do? Sue them? Make them pay royalties? I know what ISN’T going to happen… players aren’t going to stop playing. If anything, they’ll embrace the legal “cheating.” Win win!
LikeLike
@Gevlon
Even accepting that some harm is done, your hyperbole would only be correct, provided:
(1) The outcome was determined in that way in a significant potion of the matches.
(2) There was no strategy to counter it.
Both are wrong. You yourself found the strategy for (2).
However, the truth is that there isn’t even any true harm done. Being put in 1v1 on a lane against a stronger opponent is something any ladder-climber should learn to deal with at one point or another. This is also the only real way to gain the requisite twitch skills outside of professional esports training regimen.
LikeLike
@Maxim:
1: it affects 60% of the games
2: the counter expects you to look up players on 3. party sites
Mostly it’s NOT you who are facing a stronger opponent. It’s one of your teammates who gets 0/10/0 feeding his opponent, who will be unstoppable at teamfights. The counter is exactly to stop the laning phase and be third wheel in someone else’s lane.
LikeLike
@Gevlon
1. This sounds a bit high to me.
What data are you deriving this from? Is it available publicly?
2. This is a relatively high level counter, involving team composition. That kind of stuff needs to be looked up on 3rd party sites as a general rule, there is nothing wrong about this.
LikeLike
@Maxim: it came from my own analysis, look at the charts, the scale is teammates, which is proportional to game sizes. I had 55 easy wins 60 exploit mode games and 92 fair games.
It’s not about TEAM composition, it’s about player composition. Two players, both pick the same champion, both have the same MMR and visible rating (like Silver 1). One of them never played this champ, he signals a win, one of them had 20 wins 19 losses with that champ, he signals a loss.
LikeLike
When will we get a DONT PLAY LOL permanent post?
Maybe I was unprecise in my last comment, but Gevlon did answer it correctly. I was pointing out that p2w is not harmful if it is clearly transparent. What should not be tolerated is an obscure mechanism that puts the majority of players in situations where they suffer from unfair advantages, only to make payer feel unstoppable…
If a f2p player is served as cannon fodder, he should be told so!
LikeLike
@Gevlon
Fair enough
It seems i need to fall back to “where’s the harm” defense at this junction
Just for clarification, can you name a game in the MOBA genre with what you consider properly functioning non-rigged ladder and detail how exactly this game is superior to LoL?
LikeLike
@Smite: there is a saying about consenting ADULTS. So I’d say P2W games, even fully transparent should be 18+
@Maxim: I don’t know the MOBA scene too much
LikeLike
Well, I can agree to P2W games being 18+ as there are a lot of cases where minors spent hundreds of dollars/euros or more to buy shitty digital nonsense in shitty digital games (hearthstone, i.e.).
Ofc it is also their parents fault to let them run a mobile or pc account with the possibility to spend money as a minor….
LikeLike
BTW, I do not see “Don’t play League of Legends” among your permanent pages at wordpress. Is something amiss?
LikeLike
@Souldrinker: laziness. Will fix it.
LikeLike
You should definitely check Heroes of the Storm then. As an Actvision-Blizzard game the algorithm is 99,9% in it.
The matchmaking is laughably bad, 2 healers 3 assasins games vs 4 pushing specialsits ect.
it has also a lot more cosmetic items not just skins, like mounts and sprays.
LikeLike
Also it handicaps matches so enemy deal more damage
LikeLike
@Maxim Preobrazhenskiy
Dota 2 completely F2P
LikeLike
Glad you reposted this theory, I didn’t get a chance to respond to it the first time.
Basically your theory relies heavily on the fact that people who play a champion first time in ranked are most likely buyers, but you don’t have significant evidence to back this up. You played most of your games in the Silver Elo, where 60% of all players are either in this elo or below it, in Bronze. Silver Elo is known for players who do bad strategy and make hasty picks according to the meta or their own personal preference, but you’ve completely skipped any of this and made the hasty assumption that most new players are buyers.
Reasons for playing a new champion in Ranked, that you’ve not considered:
1. Champions that are strong/popular in the meta: Silver players are known for doing whatever is considered easy to abuse currently, Champions which are strong for some time, until the meta (The Overall Strategy of the game) shifts. Even champions that aren’t strong, but are huge favorites of the community (I.E. Yasuo, Riven, Lee Sin, etc.) are going to be picked up by noobs looking for easy wins.
2. Personal Preference: Silver players are known for doing illogical picks and choosing champions that they’ve never played before because they have some hope in them being strong. Lose enough times to champion and you might consider playing that champion and seeing if it’s as good as you presume.
3. Lootboxes/Incentives/Large Champion Pool: Playing a champion that you own and gaining an S-, S, or S+, grants you a chest that can give free champions and/or skins. If a Silver wins a skin/champion for a champion he hasn’t played before, he might be inclined to play that champion or use that skin. Furthermore this system as a whole encourages playing champions you own to gain more chests (As you cannot gain another chest per season on a champion you’ve already got one on), plus when you already have plenty of champions you’ve never played, why buy a NEW champion? You could just play one of your own.
4. Smurfs/Eloboosters: Silver is the elo where nearly everyone starts, so it’s the starting line for smurfs/Eloboosters to climb from. These players can play different champions every game to throw off booster bans, and still win regardless, because they’re at an overall higher level of skill than their competition.
5. Incomplete Stats and Biased Graphs: Not only have you made your theory on the grounds of most new players bought their champions, but you’ve also trusted the website, OP.GG, without understanding that it requires constantly MANUAL updating. Unless you made sure to update every player’s page before you check their champion picks, you might have missed a player who just never updated, or even more likely, never used OP.GG (A third party website) to begin with. As for biased Graphs, I refer to the 3 blue graphs showing the different teams you faced/were paired up with (Newbie Teams w Good players, Few Newbie Teams, Newbie Teams with Bad Players respectively). The first graph that gives evidence to your point the most is displayed first and in your words shows : “…Buyers+Boosters who carry them.” Fair enough, you’ve gotten circumstantial evidence that you admit to heavily bias towards Buyers, but if you continue to the Few Newbie Teams and the Newbie Teams with Bad players, you just state that you played a clown fiesta (A game without strategic thought, just a rush for kills until it’s over) in the games with Few Newbie Teams. In the the games with Newbie Teams with Bad Players, you immediately state that this is probably because some new players might just being playing a champion they already own or that the Bad players are players trying a different champion than usual. You don’t even consider that maybe this is proof that your graphs aren’t correct, but immediately spin towards your favor.
Not only have you created your theory on a shaky foundation, but you theory simply doesn’t have enough evidence to even be considered sound. You used 300+ games as your evidence, but that’s just a drop in the bucket for a system that’s so influential, yet so hidden that apparently nobody has created a theory even close to this one.
This theory is just too half-baked to work.
You want to complain about being Shadowbanned, but you deleted my comment…
LikeLike
@SeriousZ: I did update op.gg every time. 300 games with so outlying result is a 99.9% confident (it has less than 0.1% chance that pure luck created the results)
You wrote awful lot (dev or GM on duty?) to claim that a new champion user can be non-buyer. However that would mean defeat instead of win. The ONLY reason why a new champion user wins is rigging. If idiot kiddie pulls Yasuo first because he saw on a stream, he will be 0/10/0 by 12 mins. The whole idea came from a queue dodging plan: I wanted to AVOID playing with new champion users because they MUST BE instant loss and I’ve found that my winrate dropped. Good luck finding ANY reason how can a new champion user win besides rigging!
LikeLike
@Gelvon
Thanks for actually replying. I’m no Dev or GM (I’m not even sure what that means in this context), I just remember your old League of Legends post and how it completely bewildered me. It really made me want to respond to it, but I only saw it months after it was posted, so I decided to forget about it until you brought it up again.
“The ONLY reason why a new champion user wins is rigging. If idiot kiddie pulls Yasuo first because he saw on a stream, he will be 0/10/0 by 12 mins.”
This kind of logic is the heart of this misunderstanding about LoL. You seem to think it’s impossible for someone to be more than a one-trick pony (Which in case you’re unaware, it’s when you main a very small pool of champions). I understand where you’re coming from, being a one trick myself, but there’s plenty of players who can play a different champion every game and actually win sometimes. In fact most of the top tier players switch to champions they never play, but instantly do good on, because their basics are so polished from playing other champions before. I’m sure if I climbed higher I would hit a ceiling where my one main wouldn’t be capable of carrying me higher, then I’d have to learn more champions in my main role or maybe learn another role entirely.
Just because those newbies win on a few first time champs doesn’t make them geniuses at the game, in fact it’s usually self-harming if they want to climb or improve. They’re just normal players who can’t settle on one champion, the fact that they keep buying champs is proof enough that they aren’t sure what they should play or what they should do.
Just like a newbie can pick up a fighting game and learn to use a cheap combo to beat a couple of other noobs, so too can a low elo player in League, get by on a new champion. Hell, learning to play a small pool is such a good way to climb, most of these low elo players are stuck where they are because they simply refuse to stick to a single, or even a few, champions. I’ve seen it myself climbing the ladder and know friends who stay stuck because they refuse to play what they’re good at, because they would rather “play what’s fun”, even though they hate to lose, which happens more often than not. These 2 friends both have champions with 60+% winrates but few games, but whether it’s because of they don’t want to climb or because they just don’t know that they can, they play champions whom they have lots of games on, but low winrates. It’s not even the champions’ that they play fault, it’s because they refuse to learn the basics before playing champions that require them. Both of them buy plenty of skins, much more than the few skins I buy for the champs I main, but they stay stuck in the elo they belong (Bronze 5 and Silver 3), because they don’t want to learn.
If no amount of noobs could win a game by playing a first time champion, then Silver or Bronze wouldn’t just make up 60% of players, it’d make up 99%. Winning with a new champion, whether through meta cheese, a strategic pick advantage against a certain team comp, or through sheer dumb luck, is how the majority of these non-tryhards operate and discover that sometimes a certain champion is really good that random circumstance they picked it into. A circumstance that requires the matchmaking to be able know things it can’t possibly know, like every players’ picks and their weakness, all without having a sure way that the nine other players will pick them.
When these noobs do win on these first time champions, it’s only after they’ve lost on plenty of others. They don’t climb not because they haven’t bought enough skins or champions, but because they have no diligence or patience, things you just can’t ever buy. When you climbed to Gold 5, you didn’t do it out of superior strategy (You admit yourself that you played chaotically often), not because you found a superior pick (WW was very weak at that time, I should know, it took me 500 games on him to hit Gold 5 my first season), not because of a theory with no definite evidence to back it up, but because you actually stuck to something and actually tried to climb unlike everyone else. You hit your goal because of your own hard work (And plenty of games/time), not because you found a previously undiscovered exploit of a supposedly Hal9000-esque matchmaking.
The only other thing I could explain is how in your old League post, you reverse-engineered the ELO system that the matchmaking uses and how you misunderstood that too, but that’s sorta off topic and this comment is (again) long enough already.
LikeLike
@SeriusZ: I don’t deny that noobs can get lucky. But 77% winrate is not “luck”, it’s God mode. Which makes more sense:
* Noob who is in Bronze-Silver with his main, pulls random champ with zero XP and becomes a 77% winrate monster
* Buyers get boosted.
LikeLike
@Gelvon
It’s easy to get a high winrate when you’re playing against players who are just as bad as you are. Noobs vs Noobs always have more kills per match than a higher elo match.
LikeLike
@Gelvon
Oh also, a 77% winrate is absolutely abnormal unless you’re a smurf. Anyone can get a 77% in ~30 games but you’ll never see that sustain for long. There’s a saying called the 30-40-30 rule: 30% of games you play will be automatic losses that are improbable to win. 40% of games will be normal games you can win if you play well enough, and the other 30% are automatic wins that your team will carry you to. Now you could say that this is more proof towards your Rigged Matchmaking Theory, but the relation between these two ideas is correlation. Of course there’s gonna be unfair games, whether they’re wins or losses.
It’s all because the Matchmaking is not rigged, but flawed, to create fair games between players, it must use flawed data to estimate each player’s skill. The ELO system was designed to be used between 1v1 matches of players, not Teams of Players, so trying to assign a number to a ‘skill value’ of a player is all theoretical. So to even out the numbers, it has to add all players scores to a team score and make sure that the scores are similar enough. The ELO system is flawed, but it works because it’s goal isn’t a completely fair game every match, but it works to try to put you at a 50% winrate no matter what, so you always have a fair climb. That’s why to climb all you have to do is keep a 51% winrate and maintain it. Of course if you’re good you can make that winrate much higher and maintain it in higher elo too.
LikeLike
@SeriousZ: of course you get 30-40-30. But the problem is that I could predict which one I am in before the game even started. If I saw 2 new champion users, it was a sure win. If I saw zero new, 2+ average, that was a defeat. If there is no rigging, just randomness, how can I see the future?
LikeLike