Horribly inconsistent could do it I guess, I can't really see that guy being below gold though unless ADC is the only thing he can do.

Every bad ADC I support for has the same problem, they don't trade and they can't last hit for shit, he did both better than I can despite doing it wrong and not leading with AA's, and I nearly always win lanes when I ADC because it's literally retard ville down here.  160 cs in 20 minutes without a lane opponent is probably an optimistic goal for most.

Here's a really good kalista tip that I did not know. I know thunder is kind of into her right now, so this should be useful.  Its a quick guide on setting up the spirits so they map a big loop providing very good vision of dragon and the possible gank routes.

https://www.reddit.com/r/leagueoflegends/comments/355a3n/trick_with_kalistas_ghosts_w/

Id also like to play some team games. Its been a while. I am also in a bit of a slump mid lane so I would gladly play else where for 5's.

Champion mastery: So far I find it pretty silly. It seems the points gained are pretty much a flat rate on a win or loss with a champion. I carried really hard as brand in a game last night and I got 13XX points and Brad had a rough game and ended up with exactly the same points. So basically it boils down to did I win game? I did ok heres 1000 more points than you get when you lose.

It should really be calculated on stats such as KDA, creep score, Kill participation and objectives taken and have nothing to do with the outcome of the game (or very little maybe a few hundred bounce points).

This would make a lot more sense and actually make players want to play better in losing games since they would still be getting a reward toward there champion mastery. The current formula has far to few variables in it contributing to the total number of points.

Pretty sure they have plenty of computer scientists who understand basic linear algebra and matrices just fine

More likely the issue is that the people who designed the system were psychologists and marketers.

Well, I can't connect to chat tonight.  Thats fun.  I believe that you can try and manually invite me to games, so go ahead and try that if you want.

Nah.  Anything other than what they did would have been a horrible idea.

Give people more points or only award mastery of a champ for stats based on KDA and total CS?

People already freak out over KSing.  Riot should give them more reason to?  There are already enough people out there that refuse to group and win games.  No reason to award them for trying to pad their own stats by split pushing.

Only give mastery of champs for an individuals best champs?

I'm better at Leona than Thresh.  Because Leona is easy as hell to play.  I know that.  Don't need master Riot equationists to inform me of what I already know.  And maybe I play certain champs in certain sorts of games.  I'd probably a Lee Sin Master on my smurf account from the insane KDA I had when playing with my IRL friends.  That would be stupid.  And so would Riot motivating you guys to friend request my IRL buddies so you could stomp bad players to easily become 'masters' of your favorite champs.

Give people champions mastery only if they perform at a certain skill level for a champ?

Well, I want that skill level to be Gold V.  Pacov wants it at Plat V.  Cow wants it to be Diamond V.  Wherever you put the line at, Riot is putting a lot of effort into making a large portion of their consumer base feel bad because they can't be 'masters' of any champions.

So yeah, Riot missed a golden opportunity to raise the toxicity levels in games and alienate casual players.  Instead they introduced a fun little thing for us to show off what champs we like to play most.

I'm not sure how useful LSA would be in this case, but yes, there are a lot of options available here.  I'm not suggesting that Riot is using the simple starting point of a PCA.

The problem that I was pointing out is that they want to create a metric that measures people's contribution to winning without changing people's gameplay (to the negative anyway).

So how do you differentiate correlation and causation?  That is, how do you differentiate between somebody who has good stats because they happen to be on the winning team, and somebody whose good stats happen to cause the team to win?  My guess was that this is very difficult - after all, we've all been in many games with bad players who happen to have good stats because of the team they are good (cleanup Master Yis, etc).  Having time resolution on your numbers would help (so that you could create time based statistics, which would likely help here), but throwing too many different stats into your analysis will make things more difficult.

The other big issue is how you also prevent degenerate behavior.  Riot just posted about this, but you want to prevent people from doing stupid stuff to just whore stats.

posted what and where?  

also.. this

apparently the t2g thing is crap. 

yeah my in depth research mechanisms were not in place as I was busy confirming the illuminanti and nickleback connection.  I bet you didn't even look at the photograph.  Usually total nonsense doesn't get so far up on reddit - just mostly nonsense things. 

I did learn something quite interesting.  I listened to this podcast (kind of dry btw):  https://www.youtube.com/watch?v=hmSbggZK-pw

Riot Lyte shared some info about 1 thing in particular - placement matches (ex the games you play at the beginning of the season that put you into a tier (plat 1, gold 4, etc)).  He said that any dodge in the placement series is counted behind the scenes as a loss and impacts where you'd place - BUT, it does not count toward the 10 games you need to complete to get ranked.  So, dodging during placements is a very bad idea.  It's also a very good way, however, to tank your elo if that was something you are into as the placement matches have more weight than normal games. 

I kind of think ashes updated hawkshot is going to have some strong potential.  Probably very helpful to keep an eye on the jungler - and it grants vision as it passes everything.  If used to scout out jungler locations periodically, that could be something...