Mobafire League of Legends Build Guides Mobafire League of Legends Build Guides
's Forum Avatar

The Truth Behind Solo Queue?

Creator: Yukimaru
March 19, 2012 5:09am
23 posts page 1 of 3 Forums » General Discussion » The Truth Behind Solo Queue?
Permalink | Quote | PM | +Rep | Commend by Yukimaru » March 19, 2012 5:09am | Report
I dont take any credit for it, but while surving the Solomidforums, as i do every Monday, i came about a interesting Thread with Logical Mathexplanation inside:
http://www.solomid.net/forums/index.php?/topic/22191-the-truth-behind-soloq/
Quoted:

Posted this on official forums. Reposting here just to see the reactions.

---

his is just for fun btw. Not trying to start a troll thread but..
I'm new to league and lets see if I have this figured out..

This is based on pure statistics, and not accounting for personal skill and champion counters/selection/carries (ie. OP champs tryn carry soloQ hard right?). Assuming you are of average skill level, and the player base consists of POOR -> AVERAGE -> PRO players. (Or 3 categories of skill level)

There's 10 players on the map, 5 on each team, you are 1 of them.

Your performance accounts for 1/10, or 10% of the game.
4/10 players on your team, or 40%, is chance
40 divided by 3 = 13.3
13.3% chance of poor teammates
13.3% chance of average teammates
13.3% chance of pro teammates

5/10 players on the enemy team, or 50%, is also chance
50 divided by 3 = 16.6
16.6% chance of poor opponents
16.6% chance of average opponents
16.6% chance of pro opponents

Lets balance that out and say that:
pro teammates = poor opponants
poor teammates = pro opponants
average = average

16.6 + 13.3 = 29.9 (lets round it to 30)

30% chance that you will win due to matchmaking
30% chance that you will lose due to matchmaking
30% chance that it will be even and will come down to coordination in specific plays
And remember:
10% chance that you will affect the outcome yourself.

TL;DR version?
60% chance
30% teamwork
10% individual skill

So here come the hard questions in terms of soloQ:
Is it a skill based game? No. Only 10% relies on your own skill
Is it a teamwork based game? No, only 30% chance relies on teamwork

Is matchmaking broken? 60% chance of a win/lose based on pure luck?

SoloQ = Lottery?

Lol, I guess it is a troll thread after all. But its true nonetheless.

---

I think that riot's justification of the current matchmaking system is that even with a 60% win/lose variance, as you slowly climb elo due to generalization (which is another statistical term) the impact of the 60% variance will become lower and lower.

This phenomena is called "regression to the mean"
Basically the more games you play, the closer towards the "average" elo you get to purely based on the amount of people that play this game. At the point where you reach the "average" elo (Which for this game I think is around 1400) your individual skill begins to matter a lot more than teammates since all teammates you end up queuing with are also at an "average" elo.

---

Im speaking objectively here. You're actions only account for 10% of the actions in an entire game. However, how you use your 10% differs from how other people use there. If you use your time making safe kills and taking objectives then obviously your worth more than someone who goes 0/10. However, if you go 10/0, that person who went 0/10 sufficiently negated any good you did in the entire game. See how that works? Yes you can carry a game hard and go 20/0, but if two of your teammates when 0/10 and everyone else broke even then even though you're fed and carrying you're work has already been undone.

---

BUT

statistics =/= real

statistics is only "what will most likely happen probably if conditions are consistant" which they never are

---

Ps. Actually a bunch of posts stringed together which brings it all together. This was just for fun mostly. Statistics isnt real after all, its just probability, and doesnt account for specific champions, situations, or players.
Yukimaru's Forum Avatar

Awards Showcase
Show more awards
Yukimaru
<Member>
Posts: 1370
Joined: Jan 14th, 2011
Rep: Memorable (56)
Profile
Permalink | Quote | PM | +Rep | Commend by lifebaka » March 19, 2012 7:03am | Report
Statistically, that's pretty much garbage. He's assumed a uniform distribution of players across skill levels, when a bell curve or a skewed right distribution are the most likely. Either there are a lot of people bunched up in the center, at "average", with very few people being either very bad or very good, or there are a lot of people bunched up on the left, at "poor", with fewer people playing "average" and very few being very good. Certainly there aren't that many people who are very good, or else there would be more people at stellar ELO. Since playing LoL is a skill, I'm going to assume that people begin in something that's roughly bell-shaped in terms of how good they are, but then can move to the right with practice, leaving us over time with a massively skewed right distribution.

This single mistake sorta' tears apart his whole argument, since it invalidates all of his math. Seriously, all of it.

Anyway, I have more objections, and will post them later, but I need to get ready to go to my classes for the day.
OTGBionicArm wrote: Armored wimminz = badass.

My posts may be long. If this bothers you, don't read them.
lifebaka's Forum Avatar

Awards Showcase
Show more awards
lifebaka
<Scout>
Posts: 1126
Joined: Dec 12th, 2011
Rep: Prominent (151)
Profile
Permalink | Quote | PM | +Rep | Commend by Crazy Smurf » March 19, 2012 8:04am | Report
Ef this all.
All you need to win is to have a lot of swag.
- Witcher 1 and 2 Collecter Editions - Best RPGs in tha World -
Crazy Smurf's Forum Avatar

Crazy Smurf
<Member>
Posts: 745
Joined: Dec 18th, 2010
Rep: Notable (47)
Profile
Permalink | Quote | PM | +Rep | Commend by Wayne3100 » March 19, 2012 8:31am | Report
dem logics

Thanks to Hoppermh for the signature!
Wayne3100's Forum Avatar

Awards Showcase
Show more awards
Wayne3100
<Community Manager>
Posts: 7148
Joined: Aug 3rd, 2011
Rep: Esteemed (481)
Blog   |   Profile
Permalink | Quote | PM | +Rep | Commend by omagma » March 19, 2012 8:46am | Report

Ef this all.
All you need to win is to have a lot of swag.


Damn straight boy!
Thanks to me and Elleeeeeka (love that name) for my sigs.
If i made a helpful comment on your guide/build, then hit that +rep button. or don't, its not like i care.
ALL HAIL QUEEN CHRYSALIS!
omagma's Forum Avatar

Awards Showcase
Show more awards
omagma
<Scout>
Posts: 1677
Joined: Jul 27th, 2011
Rep: Prominent (175)
Blog   |   Profile
Permalink | Quote | PM | +Rep | Commend by Darcurse » March 19, 2012 9:12am | Report
Pretty much BS trying to put this in statistics.

But Regarding your own ability and how much it actually changes the game, that's how it works.
Darcurse's Forum Avatar

Awards Showcase
Show more awards
Darcurse
<Member>
Posts: 1632
Joined: Nov 2nd, 2010
Rep: Notable (38)
Profile
Permalink | Quote | PM | +Rep | Commend by DuffTime » March 19, 2012 9:52am | Report
What a crock of ****.

This is why the general forums are an abysmal wasteland.
Learn the true S3 op, Tryndamere!
http://www.mobafire.com/league-of-legends/build/duffs-season-3-hamdamere-296968
Best Aatrox in The world, with a Video Guide!
http://www.mobafire.com/league-of-legends/build/duffs-336706
Like my official DuffTime Facebook page! http://www.facebook.com/DuffTime
Follow me on Twitter, @RealDuffTime
Come hang out when I'm streaming! http://www.twitch.tv/dufftime
DuffTime's Forum Avatar

Awards Showcase
Show more awards
DuffTime
<Member>
Posts: 12302
Joined: Oct 31st, 2010
Rep: Renowned (670)
Blog   |   Profile
Permalink | Quote | PM | +Rep | Commend by Beeswarm17 » March 19, 2012 10:10am | Report
"Truth" my ***. It's like this guy has never played the game before in his life.
Beeswarm17's Forum Avatar

Beeswarm17
<Scout>
Posts: 1192
Joined: Jun 6th, 2011
Rep: Established (109)
Profile
Permalink | Quote | PM | +Rep | Commend by Luther3000 » March 19, 2012 10:14am | Report
100% chance that this guy is pulling meaningless statistics out of his ***.
Luther3000's Forum Avatar

Awards Showcase
Show more awards
Luther3000
<Sharpshooter>
Posts: 8046
Joined: Jun 24th, 2011
Rep: Esteemed (440)
Blog   |   Profile
Permalink | Quote | PM | +Rep | Commend by Wayne3100 » March 19, 2012 10:29am | Report
If my team faces CLG EU, we can either win or lose. That means we have a 50% chance of winning, right?


Thanks to Hoppermh for the signature!
Wayne3100's Forum Avatar

Awards Showcase
Show more awards
Wayne3100
<Community Manager>
Posts: 7148
Joined: Aug 3rd, 2011
Rep: Esteemed (481)
Blog   |   Profile

Quick Reply:

Guest commenting is currently disabled, please log in or sign up to respond!