Captain, there’s an anomaly in the system!
My attempt to get a ranked win with all 75 champions is going well, at least for what I expected. I’ve managed to pick up a few first-attempt wins, which feels really good. Karthus was a one shot, as were Eve and Rumble. I expected the latter two to be easy, but the Karthus win was a pleasant surprise.
Ryze and Alistar continue to give me trouble. I’ve had solid starts with both champions in several games, but my teammates just can’t find a way to work with another. It’s painful.
I also had one extremely frustrating game last night in which I was paired with an unranked Malphite, even though he hadn’t duo queued with anyone. You can see the teams in the screenshot above. My own team’s rankings, in order from top to bottom, is as follows:
The enemy team, again from top to bottom, was:
The last two players on the enemy team also duo queued together, which I’ve always seen as an advantage, especially when the two players are so close in ELO. This sort of thing just shouldn’t happen in the matchmaking system. Even if that unranked player was lying for some reason and actually did duo with another player, we were outmatched as far as ELO is concerned. I know ELO isn’t a very good gauge of player skill, but when a player is dumped into a match 120 points above his ELO, against higher-rated duo queue to boot, well, I expect about the outcome I got. He bought Ninja Tabi against a very limited physical team and “initiated” by ulting one of the enemy players at a time, most often Xin Zhao. It was a mess.
Like I said, though, this is going better than expected. I just lost another with Ryze, bringing my total attempts with him to three. We’ll see if I have more tim this afternoon to try again.
Riot adds “Unskilled Player” to reportable offenses
I’m sure you’ve been in plenty of games with players saying things like, “report this Shaco for sucking,” or “report Eve for being a moron.” Of course you wouldn’t report them. After all, isn’t matchmaking supposed to weed out the unskilled, matching them against similarly unskilled players? Well, Riot wants you to give matchmaking a hand by reporting the unskilled just like you report the trolls.
I’m going to put this in italics because I don’t want anyone to miss it: The “unskilled player” offense is NOT bannable. No one will be catching a ban because they aren’t as good at this game as someone else. In fact, Tamat said the expected impact is small at best but “worth implementing all the same.” My question is, how?
I’m trying to think of the ways that Riot can pull valuable data from this metric and I’m coming up short. For starters, the kinds of players that typically want to be able to report unskilled players tend to be unskilled themselves, just searching for external reasons a game is going poorly. As an example, I recently played a game with an Udyr who was calling for reports on his team’s Shaco, who ended the game something like 4-12. Thing is, Udyr himself was 4-13, picking up his kills late in the game. To top things off, their team actually beat us (they had a solid Amumu/Kat/Corki and we had no tank). I know that examples don’t prove anything, but this is exactly the situation I see skill based reports being asked for most. Even if decent players are using this feature, there’s going to be a LOT of unreliable data, and that’s not good for anything.
Isn’t the real issue here that matchmaking and ELO simply don’t work well for assessing a team game. Solo queue is a nightmare to balance simply by its nature. There’s not a reliable statistical method for evaluating player performance as it relates to game outcome when you’re matching 10 random players. There just isn’t. The best we’ll get is an approximation. Throwing a set of unreliable data at the problem won’t do us any good.
Posted in: Current Affairs, league of legends, News
Tags: elo, elo change, elo doesn't work, elo improvement, matchmaking, post game report, report a player, report bad players, report for being bad, reportable offense, unskilled player
My luck has officially run out
Earlier this week I was really excited to be playing ranked solo 5v5. I was making my way up the charts, thanks in part to good luck with teammates. My allies were cooperative and ready to coordinate, and chose teams that had a good mix of magic and physical damage, tankiness and crowd control. I made it from the high 1300s to 1530 in just a couple days. It was great.
That all changed at the start of the weekend. I played a few games for some Miss Fortune testing on Friday and it was loss after loss. Teammates weren’t paying attention, they were making bad champion choices, arguing about who would take middle, never bothering with dragon. I plummeted, all the way down to 1412 in something like seven straight losses, and I haven’t been able to pull myself back up.
I’ve pictured one of my most recent losses above. As you can see, my teammates were feeding early and often. Singed was 0-7 by the 18-minute mark. It wasn’t that he had a tough lane – he really didn’t – but he was running into fights he had no chance of living through. After engaging Warwick for a spell, I watched him start to run away at 200 hp just to go back in because Fling was off cooldown against a 500 hp Warwick. He died, of course, and it wasn’t the first time.
It’s strange how this game runs in streaks. I want to say that it points to a problem in the matchmaking system, but I don’t know if that’s accurate. I do think matchmaking needs to account for ELOs that drop dramatically in a short period of time. You don’t want to artificially inflate a player’s ELO so that he’s in a bracket he doesn’t belong in, but it’s also absurd that I can drop more than 100 ELO in a couple games, largely because of underskilled teammates. What if matchmaking tried to make you the lowest ELO player on a team if you lost, say, four games in a row. Losing that many games in a row is indicative of a problem with the MM system. Either a player who shouldn’t have been winning so many games was winning and he’s beyond his means, or a player is losing more games than he should, likely because the other team overmatches his own. I think the system should try to correct itself to better provide the service it’s designed to provide.
Posted in: Editorial, league of legends
Tags: bad comp, bad teammates, best comp, fail, feeders, good comp, losing, losing streak, losses, matchmaking, worst comp
Late nights with LoL matchmaking
I’m moving in about a week so my days have become increasingly hectic and I’m starting to lose sleep. It’s not that I’m worried, there’s just a lot to think about so I end up awake fairly late at night twiddling my thumbs. The obvious solution? You guessed it.
I make the distinction between night and day because night games are a very different animal. While daytime I usually expect someone on either team to know what they’re doing, nighttime that cannot be assumed. Late night games are this crazy hodgepodge of people playing toons they don’t know, crazy level one teamfights, complete and utter indifference to teammate requests for help, the same indifference toward map and buff control. I know, it sounds a lot like normal ranked games, and it is, just turned up about a hundred decibels and driving a ninety miles an hour toward a flaming brick wall. Heimer doesn’t get banned. Are you starting to see what I mean?
In a way it’s been a lot of fun. I stopped placing much stock in ranked queues a while ago, so it doesn’t bother me much that I’m seeing players go well into the double digits on kills again, or that games swing violently back and forth between one side’s advantage and the other’s. It’s been hectic in an enjoyable way.
Riot should let 3-man premades play 5v5 solo
I’ve been trying to spend a good bit of time in solo queue over the course of the weekend to see how things feel. There are still a lot of the same problems – bad players have been carried, good players are in brackets well below their skill level, there is so much animosity among teammates that it’s nearly impossible to organize people. My problem is that I want to see the stats that ranked matches provide, but playing the queue as a solo is painful.
I really think Riot should open up the solo queue to allow 3-man premades. There is really no reason not to. Among the queues, solo seems to be taken the least seriously. Even with three people, the chance for your other two players to be atrocious is high enough that it really shouldn’t be much different from the two man system.
I know there are normal games, but the fact that normal stats aren’t being tracked gives players too much flexibility to leave games without consequence. I still want to enjoy a competitive game with my friends on Summoner’s Rift and, unfortunately, the current system doesn’t allow for that.
Should ranked games have free champions?
Here’s something that came up today, mostly because I was seeing a lot of Mordekaiser players. Then I realized he’s free this week (I only looked because of the miserable quality of some of the Kaiser players). The question pretty much asks itself. Why do we have free champions in ranked matches?
The question I always like to ask regarding ranked matches is, “will this enhance or encourage more competitive play out of more players?” I have to say, I think the answer here is no. Most champions only come up on the free rotation once every few weeks, some even longer than that. Should ranked matches really be full of champions that a player might use for a few days at a time, once every few weeks? It seems to me that if you’re going to use an ELO system that doesn’t account for individual player performance, I would think a primary goal would be minimizing the attrition from bad players and maximizing a good player’s ability to carry a team. Having teammates who don’t know the champions is a huge problem in that system.
Granted, there are exceptions to this rule, but I think the only thing holding it back from being a great idea is the current 14-champion requirement to enter ranked play. Reduce it to ten, which will be an increase for some players, hopefully forcing them to play more games at the level cap rather than less before jumping in to ranked play.
Level 30 will now be required for ranked play
I am actually astounded it took this long for Riot to make this change, particularly when you consider the kind of data it had available. Granted, these stats were probably taken after the launch of Season One, but it wasn’t exactly uncommon to see players below level 30 matched with those above prior to the Season One launch either.
As Pendragon stated in a post last night, the change was made because players below level 30 tend to leave 4.2 percent of games, versus the 1.1 percent of games from level 30 players. The really interesting stat, though, is that sub-30 players have a staggering 35.2 percent win rate, while those over 30 are seeing an average 52 percent win rate.
Again, I’m shocked it took this long, and the data serves as a pretty embarrassing statement toward Riot’s foresight on this. You should start to see the change today.
Posted in: league of legends, News
Tags: level, level 30, level 30 ranked, level 30 requirement, level 30 stats, matchmaking, ranked matches, ranked play, season one, statistics
Season One hasn’t fixed matchmaking
I played a few solo 5v5 ranked games today. That’s been my method of late. A few. Not four. Sometimes not even three. Just one or two in a go, and they rarely last longer than 30 minutes anymore. It’s a strange way to play the game, kinda like taking a step back more than six months to when my ELO was fresh and I was just learning to play.
The one game that stuck out today, though, was a game in which the other team didn’t ban Shaco and I got to have a little fun. It started in their woods when I polished off Warwick at level two. I ganked him again as soon as he respawned and made it out of base, netting him three deaths in the first six or seven minutes of the game. My opponent’s other lanes were struggling as well – one because Sion was incredibly over-confident. I wondered what was going on (why we were winning so easily) until I got to the lobby. That Sion? He had one win. One solitary win, and his ELO was more than 150 points below most of my team.
I will say, most of my games have been with players of similar ELO, with a tolerance of maybe 100 points in the worst case. I can live with that. But when my queues are no longer than 25 seconds, it seems like the MM algorithm should be doing a better job than this. As more games are played, there will be an increasing gap between 1200 ELO players and 1450 ELO players. Why matchmaking would ever put those two together, particularly in solo queue, is completely beyond me.
Is character balance the real matchmaking problem?
I had this thought the other day when I got into a game with a Shaco player. That’s not really a story in and of itself – Shaco is appearing in almost every game since Zileas labeled him OP. The story is more in the way this Shaco decided to play.
Like most Shaco players, ours ran off to golem first thing to set himself up for the rest of the game. He was packing Smite, so I figured he’d spend a little time in the jungle and then, as most Shaco (and most jungle) players tend to do, hit up the lane that was pushed furthest for some early game killing. That wasn’t the case. Despite my pleading (and explanations that our lanes were suffering/being pushed and really needed to be ganked), Shaco continued jungling and only offered a phrase that completely blew my mind: “The fucking point of jungle Shaco is not to gank.”
I…was…floored. I understand that the primary advantage of a jungler is the exp. bonus for a teammate, but the secondary – and only secondary by a tiny margin – advantage is the element of surprise and opportunity to gank in any lane, thereby increasing the experience given to the ganked lane as well. My first reaction was, wow, I should not be paired with a player who so fundamentally misunderstands the game. I realized, though, that this could just as easily be a result of Shaco’s imbalance as it is the player’s ignorance.
I don’t know what your hero spread looks like, but mine is pretty wide. I play a lot of different champions, despite the bursts of new champions around patch time. I don’t think that’s the case for a lot of players, though, and I definitely don’t think that’t the case for the Shaco player in question.
This guy clearly misunderstood Shaco on a level that only someone fairly new to Shaco would. If he had been playing, say, Ashe – one of the strongest characters in the game – he could easily storm his way to a bunch of wins. That doesn’t make him good at the game, though. That makes him good at playing Ashe. The problem is that when he tries to transition to a toon like Shaco, most of his skills are lost. There isn’t the same kiting, the stun, the range advantage, the necessity to stay and farm. The game is almost entirely different, but the average player might not have that understanding about that game.
When I first started playing LoL I really liked that you could slowly build up and unlock champions. I’m starting to realize, though, that the champion unlocks coupled with the rune system means that most players will choose just a small number of champions to master and then fill out their rune pages accordingly. I actually played with a guy the other day that didn’t know Nidalee’s cat form skills don’t cost mana.
I would blame this on the character selection system, but I don’t think it would really be a problem if the characters were more balanced. Players would be less likely to change to a perceived OP toon and more likely to choose toons that they truly enjoy rather than the flavor of the month champions. Granted, not every toon can be balanced with every other toon, but when toons like Ashe and Ezreal stand as high above the pack as they have, it keeps players from being as widely familiar with the game as they could be, if only because the potentially or probably successful champion list is shorter than it could be.
Posted in: Development, league of legends
Tags: balance, broken matchmaking, character balance, easy characters, ezmode, jungling, matchmaking, matchmaking fail, shaco, skill ceiling
LoL: Throwing out the rules
I’ve never been great at moderation – when I get interested in something I tend to be completely absorbed, which is why I usually only give my time to a single game at a time. That also means I want to get the most out of whatever game I’m hooked on, if only because I don’t focus my time on any other titles. For LoL that’s meant developing a set of rules for my playtime to hopefully maximize my enjoyment. I’ve written about some of those rules here – avoiding TT when I don’t have a full premade, trying to choose the best team comp possible, playing toons with strong CC – but the rules don’t always work. I had my worst losing streak a few weeks ago and I’ve been struggling to bring up my ELO ever since. I took a week off to visit with my family, but having come back, I realized I had to try something new – I had to throw out all the rules.
So far, I haven’t seen huge success. I’ve still had the mix of good and bad players, both on Summoner’s Rift and Twisted Treeline, but I think I’m having more fun. I’m playing some toons I haven’t played in a while, including Shaco, who I used to be completely addicted to, and trying to play some new strats to pick up some wins. I had forgotten how much I love Twisted Treeline – the pace is just so much more enjoyable than SR – which was easy to do with that last losing streak.
Moral of the story is this, though. When the rules stop working, it’s time to get some new rules. I could beat my head against a wall all day trying to play the most OP toons in the most favorable situations but the bottom line is that there are too many variables outside my control for the variables within my control to make much of a difference. I know I’m a solid player, and as long as I’m playing smart every game and focusing on assisting my team and farming as heavily as possible, I’m sure I’ll be just fine.