Boards›Gameplay

Another example of Riot being deceitful

For Carthage·4/25/2016, 8:43:59 AM·5 votes·1,505 views

It was mentioned before, but I'm noticing a lot of people forgetting this:

Players don't win more games because they're positive.

Players are more often positive when they win games.

This is one example of data manipulation that Riot uses - basically, they make a statement that is only superficially supported by their data, but any further look into the matter would prove that the statement is not technically supported. Correlation is not causation, guys. Please don't be stupid. Reason out Riot's claims to make sure you aren't being conned.

And already people don't get it. What a surprise...

So allow me to not only elaborate but bring Riot's exact wording in their studies into this:

"Players who cooperate with their team win 31% more games." "Players who follow the Summoner's Code win 27% more games." "Players who break the Summoner's Code lose 23% more games." "Players who verbally abuse their teammates lose 16% more games." "Players who curse at their teammates lose 13% more games." "Players perform better if you give them constructive feedback after a mistake."

These are all deceptive. I will explain them all here one by one:

"Players who cooperate with their team win 31% more games."

  • This is a no-brainer. Since the game is a team-game, failing to cooperate with your team ultimately leaves everyone weaker - united we stand, divided we fall as they say. This, however, does not mean a team is being positive towards each other. This only means that players are cooperating.

"Players who follow the Summoner's Code win 27% more games."

  • This is a statement meant to mislead you. This does not have a thing to do with "negative attitude". On the contrary - the actual Summoner's Code is found here: http://gameinfo.na.leagueoflegends.com/en/game-info/get-started/summoners-code/. Negative attitude in-game falls under one - perhaps two - articles. In fact, the summoner's code covers a far wider range of topics than you'd think by the sound of it. Oh, and one of the things covered? AFK'ing. So I would wager to say that could very heavily skew the results.

"Players who break the Summoner's Code lose 23% more games."

  • See one above, it's basically the same thing.

"Players who verbally abuse their teammates lose 16% more games."

  • Players are more likely to verbally abuse their teammates if they're losing, that's self-explanatory.

"Players who curse at their teammates lose 13% more games."

  • Almost immediately taken from the one above, this is just a more specified point than the one above.

"Players perform better if you give them constructive feedback after a mistake."

  • And this can't be disproven, nor can it be proven. The only reason this can be "truthfully" said is because there is no absolute scientific way to prove one way or another how a human being will behave to stimulus from a myriad of different sources.

Now, before you all look to start poking holes, here is what we do not know about the data gathering that can influence the gathered data:

  1. How Riot gathered this data - anonymously or by following specific summoners.

  2. The sample group size that Riot pulled from in order to gather these numbers.

  3. Whether Riot pulled on a game-to-game basis or if Riot had players sorted into categories (basically, if a player who was verbally abusive one game but followed the summoner's code to a T the second game would be judged respectively).

  4. If this was a specific sample size number, if Riot filtered out extreme MMR's (very low and very high, gathering based on a bell-curve).

  5. If this only covers ranked games or if normal games were also included in the data.

None of this is shared. At all. There is not a hint of it anywhere in here, and all 5 of those points would heavily impact gathered data.

But here is the number one point that cannot be supported using any of this: That positivity causes more victories. You cannot prove it with the data. Their method of gathering is shaky at best and at worst it's downright unprofessional - it's skewed and worded to ensure that the best possible message can be taken from it all without actually being outright wrong.

Again, you cannot trust this kind of marketing. Correlation is not causation.

4 Comments

Blackquill4/25/2016, 1:08:44 PM3 votes

Why do you even try to call out Riot on this? None of us can prove or disprove whether these things are real statistics or just placebo but why does it even matter? The worst thing that can possibly happen is that these prove to be placebo but does this cause any harm at all? Can you really blame Riot for trying to influence people to behave better?

Trying to shame Riot for these statistics that are supposed to encourage better behaviour for the sake of shaming Riot is honestly pathetic.

And this continuing trend here on the boards and on reddit shows the immaturity of the community. Good day, Sir.

ModWulf Helhammer4/25/2016, 9:06:38 AM2 votes

So you're saying that the guy who rages at his team for every mistake, spams /ff after first blood, AFKs if someone dies, etc..... That guy wins a lot of games?

Ba22crow4/25/2016, 9:10:51 AM2 votes

You are partially correct, but there are some cases where it can be causation, just as there are plenty where they are only positive because they are winning like you said.

I have lost games because my team surrendered at 20 after some trash talk even though we were basically even across the board. I've been in the exact same situation and stepped between the people arguing and calmed the fight, and the game went on. sometimes we win in this situation, sometimes we don't, The point being when I don't step in there is no chance at winning, whereas if I do there is a shot even if it's 0.1%. Me being positive has raised my chances of winning.