EPThttp://www.solomid.net/livestream.php?s=5194
[Patch 1.0.0.150: Shadow Isles] General Discussion - Page…
| Forum Index > LoL General |
|
justiceknight
Singapore5741 Posts
http://www.solomid.net/livestream.php?s=5194 | ||
TheYango
United States47024 Posts
On November 11 2012 14:43 Irave wrote: I now understand why Loco maxes glitterlance first on Lulu. Secured myself a triple kill with her because of it. This whole time I thought the taking kills was bad as a support. Go CLG! The only scenario in which you would believe kills on supports are bad is if you also don't believe supports should ever be allocated time to farm lanes ever. If you do believe that supports do deserve some farm allocation in any form, then a support kill is essentially a wave of creeps that someone else will farm instead of them--and since they typically farm slower than the other heroes, it saves time farming the wave. Obviously TOO many kills goes beyond a support's farm needs, but generally a good number of kills on a support is fine. | ||
|
zodde
Sweden1908 Posts
On November 11 2012 14:26 Craton wrote: The idea would be to push new players out at an Elo that represents the 50% mark. 1250 was (and probably still is) top 25% and I rather doubt (with a 99.9% degree of certainty) that 25% of players are between 1200 and 1250. It doesn't make much sense to start people out at an Elo much higher than "average." Riot could use the normal Elo of the account affect the start rating in ranked games. I'm not saying that a player with a normal Elo of 1900 should start at 1900 ranked Elo, but perhaps at 1350 instead of 1200. That way you would get a much wider spread of the new ranked players. You could also put the new players in a completely different bracket and let them play say 10 placement games that decides where they end up, kinda like SC2 placement games. | ||
|
TheYango
United States47024 Posts
On November 11 2012 14:54 zodde wrote: You could also put the new players in a completely different bracket and let them play say 10 placement games that decides where they end up, kinda like SC2 placement games. The game already does this to some extent. For the most part it favors placing new players with new players, and those first few games' Elo gain loss is heavily affected by accelerated Elo (+-50 Elo, rather than +-12/13). The concept is partially muddled by the fact that oftentimes new players will duo queue with players who aren't new. | ||
|
Craton
United States17233 Posts
On November 11 2012 14:32 Cr4zyH0r5e wrote: It takes awfully long. it took me 4 days to climb from 1000 to 1250, and I've been stuck all day around 1250 'cause of random disconnects; people who only play 1 champion; people who say its their first or second ranked game, or people who have no idea what they're doing and who get carried to 1200s, from wherever they belong (I had a game where me and a draven carried a 1/18/12 eve who built ionic spark). teamcomps come into play a lot too, if you're trying to carry a bad team and the enemy team will outscale you there's only so much you can do before the inevitable. Now, I'm not saying i'm super pro and should be diamond or anything, but I think I should be able to reach at least 1400s and hoping I can make it to 1500s. Consider that you may be where you belong. | ||
|
Irave
United States9965 Posts
On November 11 2012 14:54 zodde wrote: Riot could use the normal Elo of the account affect the start rating in ranked games. I'm not saying that a player with a normal Elo of 1900 should start at 1900 ranked Elo, but perhaps at 1350 instead of 1200. That way you would get a much wider spread of the new ranked players. You could also put the new players in a completely different bracket and let them play say 10 placement games that decides where they end up, kinda like SC2 placement games. Problem solved soon™ with the upcoming patch. I don't recall where I heard it from, but I think they are considering doing something like. 1200+your current elo divided by 2 will be your starting for the next season. | ||
|
Craton
United States17233 Posts
| ||
|
zodde
Sweden1908 Posts
@Irave, what's happening in the new patch that solves it? Edit: Yeah I know, but that's a similar solution to a different problem. That change is used for each new season, doesn't help with the problem we're discussing :/ | ||
|
petered
United States1817 Posts
On November 11 2012 12:32 UniversalSnip wrote: I won't sugercoat this. As unfortunate as it is to inconvenience new readers, this isn't going to be a credible source of information if the guides suck dick. It's going to look to intermediate players like we endorse any old crap as long as our community produced it, mostly because that is exactly what we'd be doing. The new players will want quantity over quality because they can't distinguish bad advice from good, so you're not exactly doing us favors arguing for a structure that caters to them. It's not a soluble problem, there just are not enough good players who want to spend their time on this stuff here. If our community doesn't produce enough article format content to cover every champion people want to talk about, fine, let's not pretend it does. Let me give you an example. When I first started playing I liked to play shen, so I looked at the giant list of guide links in the OP of the lol megathread and found the player written guide for shen. The guide was, basically, "build leviathan, find people, kill them, when you are 20 stacks solo their nexus." However, unbeknownst to me, leviathan had been nerfed from broken to it's current garbage state since the guide was written so I would buy leviathan, get nothing done, feed and be like "man I suck." When I figured out what was up I was like "oh lol. guess that post was old." If it had been presented to me as Team Liquid's Recommendation for What You Should Build on Shen I would have said "these guides are fucking awful. I guess these guys aren't so good after all." Likewise if the guide had been updated, but I figured out it was bad. Is it really that much work to have an OP with an up to date solid mastery, runes, and core build suggestion? I mean, not every thread has to be a post from a 2K player with a dissertation on how to play the champion. We should either update the guides with good information or simply delete the old guide. It does nobody any good to have the mass of out of date guides that we currently do. Your example only supports what I am saying. We should have good information or nothing at all, then you would never have been tricked into building leviathan. Since others have found worth in having guides, I am leaning towards actually trying to update them with just basics. If it is too much work though at least lets delete out the old ones. | ||
|
Craton
United States17233 Posts
| ||
|
thenexusp
United States3721 Posts
On November 11 2012 14:26 Craton wrote: The idea would be to push new players out at an Elo that represents the 50% mark. 1250 was (and probably still is) top 25% and I rather doubt (with a 99.9% degree of certainty) that 25% of players are between 1200 and 1250. It doesn't make much sense to start people out at an Elo much higher than "average." 1200 is supposed to be the average as in arithmetic mean elo of all players, not the 50th percentile elo. (this would be definitely true if it weren't for elo decay mechanics.) It's entirely possible that 1200 is the average elo and yet 65% of the population is below that elo. Starting players off at the 50th percentile elo has its own host of problems, mainly, if the elo distribution isn't perfectly symmetrical then there will be elo inflation/deflation over time. | ||
|
Craton
United States17233 Posts
| ||
|
Sufficiency
Canada23833 Posts
On November 11 2012 14:54 zodde wrote: Riot could use the normal Elo of the account affect the start rating in ranked games. I'm not saying that a player with a normal Elo of 1900 should start at 1900 ranked Elo, but perhaps at 1350 instead of 1200. That way you would get a much wider spread of the new ranked players. You could also put the new players in a completely different bracket and let them play say 10 placement games that decides where they end up, kinda like SC2 placement games. inb4 tryhards in normal everywhere trying to get out of "elo hell". | ||
|
thenexusp
United States3721 Posts
Anyway, my point was that there's nothing contradictory between the statements "1200 is the average elo" and "70% of players are under 1200 elo" | ||
|
Sufficiency
Canada23833 Posts
On November 11 2012 14:43 Irave wrote: I now understand why Loco maxes glitterlance first on Lulu. Secured myself a triple kill with her because of it. This whole time I thought the taking kills was bad as a support. Go CLG! Maxing Q is the standard skill order anyway. I scream when I see a Lulu maxing E first. It's not the right way to play. | ||
|
Alaric
France45622 Posts
On November 11 2012 13:25 Sermokala wrote: Next time you go coop vs ai with nid take surge it makes everything crazy for nid. are you making your heal or what? if your going ad you should really max heal first. throw your spear and make sure that they get the effect on your trap both will make your dps really nice. in cat mode your q is fucking nuts for an execute move. It scales up your damage by 3% for every 1% that they're missing and adds an amount of ad to the attack as well. Also rageblade. Yeah, I knew all that theorical stuff (what baffles me most with Nidalee when theorycrafting is her itemization, AD Nid seems to have so many viable options due to her being ranged and the nature of her kit @_@) but the execution was still much harder than what I'd have thought. Felt helplessly pouncing after bots and trying to damage them while Tryndamere and Garen just ran up and killed them.  | ||
|
Irave
United States9965 Posts
On November 11 2012 15:30 Sufficiency wrote: Maxing Q is the standard skill order anyway. I scream when I see a Lulu maxing E first. It's not the right way to play. I'm fairly new with her. I typically go r,w,q,e. Get the longer poly/slow then a more gentle followup glitterlance. New favorite stream. King Neilyo, 1900 play good commentary http://www.solomid.net/livestream.php?s=2786 | ||
|
Dgiese
United States2687 Posts
On November 11 2012 15:24 thenexusp wrote: Say the natural elo distribution is naturally right-skewed. The average elo of the population is 1200 but the 50th percentile is 1000. You start new players at 1000, this brings the average elo of the population down. After a bunch of new people were added the average elo of the population is now 1150 and the 50th percentile is 950... Anyway, my point was that there's nothing contradictory between the statements "1200 is the average elo" and "70% of players are under 1200 elo" I'd be interested in seeing real data on lol players, some of the stats they provide are too ambiguous to derive any meaning from. 1200 is the top 25% for example. Top 25% of what? All players? Only ranked players? At least provide some more background information, If they are feeling extra generous sprinkle some pictures of the distribution on top | ||
|
Mondeezy
United States1938 Posts
On November 11 2012 15:38 Irave wrote: I'm fairly new with her. I typically go r,w,q,e. Get the longer poly/slow then a more gentle followup glitterlance. New favorite stream. King Neilyo, 1900 play good commentary http://www.solomid.net/livestream.php?s=2786 He was a really famous WoW player (rogue). Glad to see he's pursuing league now | ||
|
chalice
United States1945 Posts
On November 11 2012 15:38 Irave wrote: I'm fairly new with her. I typically go r,w,q,e. Get the longer poly/slow then a more gentle followup glitterlance. New favorite stream. King Neilyo, 1900 play good commentary http://www.solomid.net/livestream.php?s=2786 "i'll fake afk, hopefully someone will dodge" is not the good commentary im looking for when im watching a lol stream | ||
| ||
