Why must toxic players ruin my games?

Comment below rating threshold, click here to show it.

PogoPogoPogoPogo

Senior Member

10-31-2012

Quote:
Originally Posted by AtheistGuy1 View Post
Purely? No. Largely? Indeed. Sometimes people get lucky; but a competent team can minimize the threat that poses.
I understand that luck is involved, but over a large sample size, say thousands of games, the luck will swing both ways, right? And after a thousand games, if the two teams truly are equal skill, then they'll each win roughly 500 games. There's of course some variance, but it will be very close to 500 wins for each team.

Now, the importance of this is that we're basically guaranteed that over a large sample size, if I'm playing at say 900 Elo, the 9 players besides me will have an average Elo of 900 Elo. If I'm better than 900 Elo, my team will have an advantage, and out of 1,000 games, we'll easily win more than 500 (the better I am, the larger the difference in wins will be), and the same can be said if I'm significantly worse than a 900 Elo player, my team will lose games.

Right now, we're still working with the assumption that there are 0 trolls, AFKs, DCs, yada, yada, etc. You can agree to everything I've said so far, right? Any individual match may be determined by luck, but over a large enough sample size, luck evens out and the wins basically represent the relative skill level between the two teams, right?


Comment below rating threshold, click here to show it.

AtheistGuy1

Senior Member

10-31-2012

Quote:
Originally Posted by PogoPogoPogoPogo View Post
Yes, I did. I'd rather simply discuss it on the forums so that it's at the viewing pleasure of others.

If I'm going to go through an elaborate, over-explained proof, then it may as well be for public viewing.

Then you'll have no problems telling me the kind of math you've taken?

Also, it's not to discuss this elsewhere, I just think having a more personal conversation will ease these hostilities.


Comment below rating threshold, click here to show it.

AtheistGuy1

Senior Member

10-31-2012

Quote:
Originally Posted by PogoPogoPogoPogo View Post
I understand that luck is involved, but over a large sample size, say thousands of games, the luck will swing both ways, right? And after a thousand games, if the two teams truly are equal skill, then they'll each win roughly 500 games. There's of course some variance, but it will be very close to 500 wins for each team.

Now, the importance of this is that we're basically guaranteed that over a large sample size, if I'm playing at say 900 Elo, the 9 players besides me will have an average Elo of 900 Elo. If I'm better than 900 Elo, my team will have an advantage, and out of 1,000 games, we'll easily win more than 500 (the better I am, the larger the difference in wins will be), and the same can be said if I'm significantly worse than a 900 Elo player, my team will lose games.

Right now, we're still working with the assumption that there are 0 trolls, AFKs, DCs, yada, yada, etc. You can agree to everything I've said so far, right? Any individual match may be determined by luck, but over a large enough sample size, luck evens out and the wins basically represent the relative skill level between the two teams, right?
Let's say "yes" for argument's sake.


Comment below rating threshold, click here to show it.

PogoPogoPogoPogo

Senior Member

10-31-2012

Quote:
Originally Posted by AtheistGuy1 View Post
Then you'll have no problems telling me the kind of math you've taken?

Also, it's not to discuss this elsewhere, I just think having a more personal conversation will ease these hostilities.
A few years ago, I took Finite Mathematics (which is the math being used here), and I'm currently taking (and passing with an A) Calculus I.


Comment below rating threshold, click here to show it.

PogoPogoPogoPogo

Senior Member

10-31-2012

(At this point, I don't think there are any hostilities, because you seem to be willing to try to understand the point I'm making and have stopped accusing me of not getting your point.)

Quote:
Originally Posted by AtheistGuy1 View Post
Let's say "yes" for argument's sake.
I mean, it's pretty hard to say no to that first point I was making. The problem with the first point I was making is that it requires a dream world where there are no trolls, and we both know that simply doesn't exist. Regardless though, we know that SOME portion of matches we play have zero trolls in them (even if it's a ridiculously small portion), and we've now established that, assuming we have equal skill to the average for our Elo level, we will have a 50% win rate in matches in which there are 0 trolls.

So next we're going to look all the various scenarios where there are trolls in your match.

In my first explanation, I only showed the math for when there is 1 troll in the math, but that same math can be extrapolated to any number of trolls (1 to 9, since you're the 10th and not a troll). But right first, I want to look at 4 special cases and establish win rates for these 4 special cases.

The four special cases of possible trolls are when there are an equal number of trolls on both teams.

Scenario 1: Each team has 1 troll.
Scenario 2: Each team has 2 trolls.
Scenario 3: Each team has 3 trolls.
Scenario 4: Each team has 4 trolls.

If we can agree that in the scenario where both teams have 0 trolls, we have a 50% win rate, then can you also agree that in all four of these scenarios where both teams have the same number of trolls, we also have a 50% win rate?

Yes, there's still luck, and in some cases your team's trolls may be trolling harder than the other teams trolls. But just like before, we know that in the long rung, luck will swing both ways, and the enemy team's trolls will also be the harder trolls in some cases, so over a sample size of something on the order of at least 1000 games, each team would win roughly 500 games, right?

Are we still agreed up to this point?

So far, I think everything I've explained is pretty easy to agree to (although, so far, I also haven't demonstrated how one team gets more trolls than the other, not yet, but we're getting to that part).


Comment below rating threshold, click here to show it.

PogoPogoPogoPogo

Senior Member

10-31-2012

Quote:
Originally Posted by AtheistGuy1 View Post
I've done linear equations, graphing, done quadratic equations, factoring, and will be learning to solve rational expressions tomorrow. Yourself?
So by the way, just so I'm clear, this is like Algebra II, right?

Linear equations being: y=mx+b
Quadratic equations being y=ax^2 + bx + c

And by factoring you mean like 4x^2 + 2x = 2x(2x+1), right?


Comment below rating threshold, click here to show it.

AtheistGuy1

Senior Member

10-31-2012

Quote:
Originally Posted by PogoPogoPogoPogo View Post
I mean, it's pretty hard to say no to that first point I was making. The problem with the first point I was making is that it requires a dream world where there are no trolls, and we both know that simply doesn't exist. Regardless though, we know that SOME portion of matches we play have zero trolls in them (even if it's a ridiculously small portion), and we've now established that, assuming we have equal skill to the average for our Elo level, we will have a 50% win rate in matches in which there are 0 trolls.

So next we're going to look all the various scenarios where there are trolls in your match.

In my first explanation, I only showed the math for when there is 1 troll in the math, but that same math can be extrapolated to any number of trolls (1 to 9, since you're the 10th and not a troll). But right first, I want to look at 4 special cases and establish win rates for these 4 special cases.

The four special cases of possible trolls are when there are an equal number of trolls on both teams.

Scenario 1: Each team has 1 troll.
Scenario 2: Each team has 2 trolls.
Scenario 3: Each team has 3 trolls.
Scenario 4: Each team has 4 trolls.

If we can agree that in the scenario where both teams have 0 trolls, we have a 50% win rate, then can you also agree that in all four of these scenarios where both teams have the same number of trolls, we also have a 50% win rate?

Not really. After you throw trolls in (Troll being defined as someone who deliberately plays to the detriment of others and/or creates a negative experience with their words- intentional or otherwise), all hell breaks loose. There's no data to support any sort position. Any sort of assertions past this point are, at best, contentious.

(If you haven't noticed already, I'm a science major.)

EDIT:
Quote:
Originally Posted by PogoPogoPogoPogo View Post
(At this point, I don't think there are any hostilities, because you seem to be willing to try to understand the point I'm making and have stopped accusing me of not getting your point.)
There are plenty, but they seem to be going over your head. And my position has not budged an inch given what I've seen so far.


Comment below rating threshold, click here to show it.

AtheistGuy1

Senior Member

10-31-2012

Quote:
Originally Posted by PogoPogoPogoPogo View Post
So by the way, just so I'm clear, this is like Algebra II, right?

Linear equations being: y=mx+b
Quadratic equations being y=ax^2 + bx + c

And by factoring you mean like 4x^2 + 2x = 2x(2x+1), right?
Correct. Here we call it Math 139.


Comment below rating threshold, click here to show it.

PogoPogoPogoPogo

Senior Member

11-01-2012

Quote:
Originally Posted by AtheistGuy1 View Post
Not really. After you throw trolls in (Troll being defined as someone who deliberately plays to the detriment of others and/or creates a negative experience with their words- intentional or otherwise), all hell breaks loose. There's no data to support any sort position. Any sort of assertions past this point are, at best, contentious.
As a science major, I'd think you'd understand about sample sizes.

I'm not arguing that in ONE match, if both teams have an equal number of trolls, they have equal chances of winning. I'm admitting that for singular games, variance will be extraordinarily high.

I'm talking about over the course of a very large sample size. That's what all statistics (and science, by the way) deals with.

If you can't agree that over an extraordinarily large sample sizes, two teams identical teams will average out to the same number of wins, then you're going to have to explain to me how this is not the case.

Anything that can or will happen to one team causing them to lose that individual match is just as likely to happen to the other team, right?


Comment below rating threshold, click here to show it.

PogoPogoPogoPogo

Senior Member

11-01-2012

Quote:
Originally Posted by AtheistGuy1 View Post
Correct. Here we call it Math 139.
Just so you're clear on my familiarity with math, this is the level of math I was doing 10 years ago. That doesn't mean the level I'm doing now is 10 years ahead of you, because I've been in and out of school, but the math in this thread IS college level, while the math you're doing now is high school level, so it's a little beyond what you've done to this point, but I don't think it's that complicated (you just haven't been exposed to it yet), so I should be able to make you understand it.


That is, however, if you can either agree to the logical assumptions being made or prove them wrong. You can't just say you disagree without really proving them wrong.