Greedy Goblin

Thursday, February 2, 2017

Now that's weird

My "league of legends is rigged" page is not getting the traction I expected. It's surprising considering how successful the "World of Tanks is rigged page" is.

It's also clear from the various "criticisms" on various reddit and official forum threads that people didn't even read it. No one even mention "new champion user" in the "criticism", which is the central point. The one-line summary is "if you have new champion users in the team, you'll easily win, if the enemy has new champion users, you'll lose". They simply ignore this and come to criticize my performance in games which supposed to be irrelevant in the research. Their point is "you just make stuff up to excuse why you lose" which has nothing to do with what I claim to find. Even if I'm in bronze 5 and lose every game except when I have new champion user teammates, the results are valid.

It's the same blog and I'm the same blogger who write the World of Tanks is rigged post that has over 100K hits and its main statement is considered consensus opinion, even if most people didn't heard it from me but from some popular gaming site (who heard it from me):

On the other hand, while the public opinion isn't high on League of Legends, it's not considered rigged:

One reason I can see is that the media changed a lot in the last 5 years. Back then there was blogging age: the "community" was around a few opinion leaders who read each other. If just one took the time to understand and check the results and found them worthy, they linked and mentioned it so it got to awful lot of people. Currently there is the social media age when everyone follows lots of little guys like himself. The few who bothers to read before commenting simply can't reach enough more people to start a chain reaction. Most of them are "lol didnt read mustbe scrub making excuses".

This is a real world problem. After all when crazy ideas like Hillary Clinton running a child abuse ring in a pizzeria or Donald Trump taking part in pee-sex parties can get traction, it's not surprising that something like "new champion users in LoL get free wins" doesn't get spotlight.

I don't really know what I should do. Simply more research, data collection won't help, people who didn't read the current page won't read the page with more data. Maybe reaching Platinum and pitching it as a guide "how to reach platinum while being low" would help? I don't know. But this is something that bothers me a lot and have to find the solution. Unfortunately as Riot knows which is my account (and also my machine ID and my IP) it's easy for them to stop me from reaching Platinum, even if I put in the effort of playing this rigged game. Not to mention that previous dodging can have a remaining flag on my account even if it's not manually sabotated.

I really don't know what to do. I mean I can fight people who say "your research is wrong" but definitely can't fight "didn't read lol xd" ones. In the "good old times" I could ignore them as I only needed the ears of fellow bloggers and forum/reddit celebrities to spread ideas, but now one seems to have to convince ordinary folks to reach other ordinary folks.

21 comments:

maxim said...

Alternatively, you can consider the radical notion that your research might actually be wrong in the sense that your initial assumptions took you as far as they did, but they can't take you any further. Not the first time this happened in history, either. When Francis Bacon invented science, what drove him to do that was a purely religious bias.

Riot probably doesn't give enough of a damn about you specifically to stop you from anything. You didn't do enough to warrant their attention.

The funniest thing to me here is that it is entirely possible that you can't reach Platinum without significantly improving your mechanical basics. You, however, are all too ready to attribute any potential future loss not to your lack of skill, but to Riot.
The losses didn't even happen yet, and you are already blaming them :D

maxim said...

Which, incidentally, puts me in "scrub making excuses" crowd.
Let us not forget the original Sirlin definition of word "scrub" as a "a person dreaming up rules by which the game must in his mind operate, when it really doesn't"

Anony said...

Walk away. You have done your research. Made your post. Either the internet will pick it up and run with it or they won't. It may be a slow burner, or people simply don't care enough to fully read and respond.

It's out of your hands on now.
Move on.

Alessandro said...

In my Circle of friends, it got a lot of traction! But the general sentiment was "That's sad... but I'll keep playing until I'm so good that I beat the system without any shenanigans".

Also, I translated and summarized your post - attaching a link to your post. But I don't know if they followed the link.

Maybe it needs more "marketability/propaganda"? Or maybe the LoL community just shrugs on the possibility of rigging, "whatever, I play for fun".

A key point that is different from World of Tanks is that League of Legends has a lot of reputation as an "E-Sport" - probably the one with more money involved. They might think "Meh, at least the World Champion Matches are not rigged!".

Raphael said...

First, I think you should ask readers to replicate your results and share them in subsequent posts to further prove your conclusions.

Second, and what I want to focus on: Even assuming people 100% accept your conclusions, I think you overestimate how troubled people will be that some games in silver and gold are tilted towards recent skin buyers. As long as they are 25 easy win/25 likely loss/50 fair, people may just be OK with that. And you figuring out the meta and how to respond to it is good play on your part, but not something that undermines the experience of playing LOL enough to trouble them.

Compare this result to your "WOT is rigged" result. There, you proved "matchmaker is rigged, skill is not rewarded, play mindlessly to get more rewards." Here, even assuming that they accept your proof, you proved "matchmaker is rigged, metagaming skill is rewarded, you must adapt your play to the situation you find for more rewards." See the difference?

Cathfaern said...

You can try to reword it, as you wrote. Because "who cares if Riot is doing some shenanigans? Every company is for profit and etc.", but "guys look at this guide which helps you to climb fast to gold!".
Or maybe it also would not help. I don't play LoL but ocassionally check LoL reddit and forum threads. And to my suprise many-many player don't play any ranked (or whaterver it called in LoL). So maybe there is not enough player who cares what's happening in ranked?

Antze said...

Well, LoL community is generally known to be a bit... ugh... well, you came to the conclusion of muting everyone in the game yourself.

Still, looks like advertisement these days is not the same it was before. I've seen it not the first time - someone makes something great and noone cares, unless you run after them and put it under their noses.

In the comments of that reddit post there was at least one person who wrote "hey I actually tried this crazy stuff and it seems to work", so maybe hope isn't lost and you just have to wait a bit. But I wouldn't bet on that.

Or maybe most people don't care. I am not a dedicated LoL player but when I played, I played normals, and I think I bought a skin once. Even though I consider you totally right on the matter, your research doesn't stop me from starting LoL again and play a couple of normals, because all that rigging doesn't make my personal gaming experience any worse, I never expected much from the game and never will. Probably your work will stop me from paying Riot again, though.

Who knows, maybe there are more LoL players with similar "OK, it exists but it's irrelevant" motives for some reason?

Tithian said...

People often do not have the mental capacity to think about complex stuff, and this is even more true to LoL players than anything else. They criticize your performance because that was the only thing they could understand from your entire post. Aka, they are morons.

Do not waste your time on trying to educate morons that do not want to be educated, it's futile. On the other hand, if your research somehow 'attacked' them in the social level, the reaction would be differennt, because everyone instinctively reacts to social engineering.

Thus your next project should be more about social interactions rather than the technical stuff. It gives everyone, even the asocials, more things to talk about.

Anonymous said...

Your primary as in only datasource is heavily biased and the results are dependend on the mindset of the source.

This is all someone who read a study before needs to dismiss your work.

Where is you scrapping through op.gg and testing your hypothesis against games of thousands of users.

And btw the league community went through thousands of claims like yours, resistance did build (elo hell etc)

Esteban said...

Have to agree with the other commenters here: even if the research were perfect (and let us be honest, it is not - the correlation between new champion in ranked and recent buyer is assumed, and the games are a veritable hell of uncontrolled variables with too few datapoints) I just don't think people would care. In the end, it is possible to develop sufficient skill to climb high anyway, even with the occasional unfair feeding game, and as Alessandro's friends exemplify, 'skill' as opposed to 'clever gaming of the system' is the power fantasy sought after here. Can't divert the Volga with a stick.

Peer review would go a long way, though. Specifically, I would really, really like to see a complete novice player use your classification of games and the algorithm to climb. Like back in 2009, in your WoW-goblin days, when you sponsored apprentices to succeed using your gold-making methods. Thought that was pretty cool because it shifted the emphasis from your person to your ideas.

Gevlon said...

Scrapping through op.gg is problematic because I need information if the player is new champion user at the time of the game and there is no such info on op.gg. You can see who are new champion users right now.

Do you know a way to test the exploit mode without someone who believes it's needed is testing it?

Also, peer-reviewing my work isn't hard. All you need is to fire up op.gg and check the number of new champion users on the team and see that you win those games with high percentage. Actually testing my work is less effort than reading about it.

People resitant to rigging claims simply don't read rigging claims. What I don't understand is someone putting in the effort to read it and comment on it, without even mentioning its main point.

Cathfaern said...

If someone would like to try this method from scratch (who never played LoL) what would be the best hero to rule out the personal skill development?

Anonymous said...

Would it be possible to scrap todays matches and analyze them using current "new champion" data?.

Repeat that for a few days and you should have accumulated a shed load of data.

Basil said...

You have two problems. First: your post should contain a way for the reader to see this for themselves. Second: you walk through your entire method starting with your hypothesis and all your false starts, leading to your point too far down in the article. I understand why you wrote it this way, but to have the greatest impact, you should start with the punch line, and then use your data to prove it, working backward.

To give the post more traction, I would advise that you rework it as follows:
-Statement that it's rigged
-The way it's rigged, including a way to see the rigging for yourself (spend a week checking for paying players and decide whether the match seems unequal)
-The story of how you hypothesized and proved this

Gamjatang said...

In OP.GG you can click the champions tab to see how many times an account has played a certain champion. Go find that really bad unranked player and see how the total win/loss ratio on champions he or she has only played once compares to your sure win/fair game/sure loss ratio.
Find more unranked players and do the same. Repeat until you get a good sample size.

Your reddit post might have gained more traction if you didn't write like a conspiracy theorist. You also say that pros recommend learning every champion, which literally, no good or even average player will ever say.

You make too many assumptions.
You go from...

one player with positive win rate on one champion plus one player with no record on one champion

to...

rigged.

No data in between. No effort to rule out correlation. In a rigged scenario, you MUST assume causation. You are still in tinfoil hat territory just as the reddit comments say. Saying that Riot knows your account and can now prevent you from climbing makes it worse.

Assumptions made:
1 new champion users are bad
2 new champion users don't deserve to win
3 matchmaking should be perfectly fair

1 if new champion users are bad, what is their kda at the end of the game? Check champion.gg. It shows a winrate curve on games played with a champion. Why aren't all of these "more games played = more wins"? In fact, it often makes a negligible difference.
2 if new champion users don't deserve to win, what is their ranked win/loss ratio?
3 if matchmaking should be fair then why should someone be allowed to climb with a 50% win rate?

Perhaps compare it to fencing. A new fencer is unpredictable and has that as an advantage.
There might be an advantage to playing a champion for the first time. I think the majority will play less aggressive and thus be less likely to feed. A good player is also less likely to feed. Thus, including yourself, your team can only have two feeders whereas the other can have five. That's a 2.5x higher chance on the other team. If feeders alone determine your chance to win then 2.5x equates to a 71% win rate on those games. I've gotten a similar win rate number as your 'sure win' without the use of causation and head scratching assumptions.

Gamjatang said...

Turns out OP.GG data on winrate on games played is old/broken.

However, I found something even better that everyone who believes in your argument must address.

If the game is rigged then the overall winrate on a champion played for the first time should be greater than the overall winrate on that same champion played the second time. Less clear math would suggest that the winrate should also be above 50%.

This is simply not true.

Go to lolking.net and look up stats on a champion. You can set it to filter for only silver games, which you were in.

How can a first timer be rigged to win if he is in fact more likely to lose?

Bill said...

You assume that an argument can be made with facts. I'm not saying that to be facetious. In fact in some cases presenting facts to people who currently disagree with you will reinforce their false beliefs.

Your best bet is to create a decent TL;DR summary of your method at the top, put the explanation underneath for those intellectuals interested in the explanation and title it "How I beat League of Legends match rigging". This feeds into other people's feelings that the game is rigged, and it suggests a way to get around it. Very appealing. Some people will try it and quit because either there is no challenge OR they hate playing a rigged game OR they don't quit and undermine the paying players that subsequently stop paying.

Another reason to boil down your research into a simple game guide: players will try out those methods just to see them work. You win converts that way. Down the road they may be more open to what you say.

Anonymous said...

A bit tongue in cheek but here it is: People cant be bothered to read anymore. You can see it also with game reviews: video reviews are soaring but written reviews largely ignored...

Randomus271 said...

My explanation of this result can be broken into 4 main categories:

#1: You were heavily biased from the beginning and made very little attempt to hide it. Not only did you express this bias in your early posts regarding LoL - before you had a working strategy to prove it...You re-affirmed your bias right from the beginning in your official results summary page. You literally only made it 2 sentences even pretending you ever thought the game was fair...And that is about how far most people kept reading. Opening on a false note/lie is never a good idea when presenting an argument.

#2: Due to the lack of clear leader-boards or statistics for LoL - as well as to the intricacies/complexities of the actual matchmaker's process which still don't line up perfectly with your simplified formula (I do realize you literally don't have access to the data required for a perfect formula - you did the best you could under the circumstances - but the end result is that your formula remains imperfect nonetheless) - your results end up feeling fairly incomplete and inconclusive. Certainly you've demonstrated a statistical anomaly over the course of several hundred games...But given your own ups and downs and your limited win-rate even following the formula (since it is necessarily imperfect) the fact remains that you simply do not have enough hard data to conclusively disprove simple random chance being the deciding factor.

#3: Your claim that it is "easy to prove for yourself" is ridiculous. Even if the average LoL player was willing to alt-tab in/out of game constantly to pull up information about their opponents on an external site to run through your formula - the average LoL game takes anywhere from 20 minutes to 2+ hours - so the average LoL player only has time for a few games in a single day, probably only a couple days a week... Replicating even your latest run of results would take years for the average player - and it is ridiculous to expect them to do that just based on your word.

#4: Ultimately the biggest one for me personally: Even if you are right - the rigging doesn't actually affect the average player in any meaningful way. Since there are the same number of "Easy Win" and "Sure Loss" games - they balance each other out, and the average player will still win exactly the same number of games as if the system was entirely fair and balanced. The only impact on them is that 50% of their games take 2+ hours and are long-drawn-out struggles vs evenly matched opponents, and other games only take 20-30 minutes and either they completely crush their opponents or they get crushed. Since without using your formula they have no way to know which games are which - this really just serves to add some variety/randomness to the games for the average player - rather than constant long grinds. This actually allows them to play more games before burning out, and is overall probably a beneficial thing for them. You've certainly identified that an anomaly in the matchmaker exists...And a way to exploit it...But despite you pointing out how it could be used to support your bias that the game must be rigged to favor paying players - the only player you actually have conclusive evidence of the system unfairly benefiting is...Yourself - a non-paying, antagonistic player...

Conclusion: If you really, honestly want to know why your "WoT is Rigged" blog entries were so thoroughly accepted and your "LoL is Rigged" blog entries have received a far more negative response - I recommend that you take a step back (and a deep breath) - and objectively read through your entire series of both WoT and LoL posts side by side. Pay attention to the tone you took in writing the WoT material vs the LoL material - as well as the type and quantity of quantifiable, provable information you presented in each. You are a smart man - I'm sure you can see the differences for yourself.

Trees said...

The response you are getting is the same as if your argument was "Dota is better than LoL". The people who play one lane-pusher/wizard-battler are always going to want to believe their time-waster of choice is best, and will argue with anything that challenges that.

It might also be that WoT is a game many people already accepted as rigged (gold ammo) but didn't have the time/data/exposure to prove it. Where as LoL has community managers, PR people, and devs actively denying that it's systems are rigged, and the people who play LoL choose to believe their lies because they want their success in a 'fair' game to be verified.

Alrenous said...

It's because LOL is far more popular. Currently 20x as many viewers on twitch than WOT.

Result: psychological investment. Having identified with the game, they don't want to believe it's rigged, because that's low status and would make them low status.