Understanding Tier Lists

Intro

Gentleman Gustaf here today to talk to you about Tier Lists. Tier Lists are a pretty divisive topic in many circles. Casual players think they ruin the game by making it seem too deterministic. Amateur competitive players may think the tier lists are stupid; they cause players to fixate on a few characters, or they may think that the tier lists are simply flawed; we never have perfect knowledge of a game, or they may think that the whole idea is pointless, as it still comes down to skill. Finally, pro players - well, we'll get to that in a minute. But first, what is a Tier list?

 

Tier Lists

When we approach competitive games where you have a choice of characters in some way, some characters are considered more powerful than others. This may mean that they are simply easier: they have the same potential output as another character, but it's harder to mess up. For example, if you made a new champion, Ryze 2.0, who was just like Ryze except all of his abilities were skillshots, they'd have the potential to go toe-to-toe: they'd have the same damage output, the same stats, the same everything. But Ryze 2.0 would have to land skill shots to put this damage out, whereas Ryze would not. Essentially, the mechanical complexity for playing Ryze 2.0 would be higher, and as such, the skill cap would be higher, and the room for failure due to errors would be higher. Both characters would still take skill. Both characters might still be viable. Both characters may even end up being tournament competitive, if the meta favored his kit enough. But why pick Ryze 2.0 over Ryze?

However, this is the simplest idea of a Tier List: which champion is best. In reality, this is hard to do, because champions are simply not as comparable as that. There is no Ryze 2.0, instead you have to compare Ryze to Morgana or Ahri or any number of other champions with entirely different abilities. And so Tier-Lists are less a function of 'good-ness', and more a function of viability/safeness. Essentially, a huge factor in Tier Lists is the ability to be countered. Let's say you have an AP Carry who beats every other AP Carry in the game. This AP Carry would a top tier AP Carry. Would it always be the best choice? Not necessarily. Let's say this AP Carry SLIGHTLY beats every AP Carry in the game. If they pick their AP Carry before you pick yours, you're better off picking a different AP Carry; one who beats their specific AP Carry harder. This, for example, is part of why Morgana has remained high-tier for so long; she isn't necessarily the best laner against any given champion, but it sure is hard to counter her in lane. Most of the champions who "counter" her in lane don't really counter her in lane so much as minorly inconvenience her and not lose lane to her. Champions who can fill multiple roles are even more extreme in this tendency. You can counter them in one role only to have them go to another role. In fact, if we take the top three champions from Elementz' (admittedly old) tier list, we see that the top 3 champions can fill multiple positions; when it comes to Tier Lists, versatility is king.

I do not mean to suggest that tier lists are built solely on matchups and difficulty of play. Matchups are very important, and difficulty of play varies wildly as a factor depending on the competitiveness of the tier list. Other factors include damage, cooldowns, and base stats.

Typically, Tier lists are broken up as follows

  • God Tier: Champions which have no glaring weaknesses, and thus no strong counters (except extremely niche champions).
  • Top Tier: Top tier champions are basically less strong god tier champions. While god tier champions stand head and shoulders above the others, top tier champions are simply at the top (and some champions have to be). They are typically low on weaknesses, and only countered by a few (probably niche) champions.
  • Mid Tier: These are your run of the mill champions. Mid tier champions tend to lose their matchups to most top tier champions (slighty), but to have a few favorable matchups against a few top tier champions.
  • Low Tier: These are champions that you probably don't want to pick, although if you are really good at them, they can still be reasonable picks. It's just that if you got that good at some other champion in Mid Tier or above, you'd do better. They can occasionally have niche value against opponents who aren't used to seeing them, but this is really only applicable to lower levels of skill
  • Trash Tier: These champions are extremely niche. Perhaps if used with a team built around them, and against a team they perfectly counter, they could be a viable last pick.

In fighting games, Tier systems include matchup information. So 'Graves vs Vayne: 6-4', means that in a typical 10 game set between equal skill players, the Graves player will win 6 games, and the Vayne player will win 4. As far as ranges go, the average matchup rating will be higher the higher your tier. It is this average value that is very important. A Low Tier champion may have 4 incredibly favorable (8-2) matchups, while a God Tier may have only 2, but the God Tier will have almost no 4-6 matchups, whereas the Low Tier champion will probably have mostly 4-6 matchups.

Why Tier Lists?

Tier Lists serve two very simple functions:

First, Tier Lists provide a good intro into the competitive aspect of the game for players. Higher Tier champions are not always beginner friendly, but they are often slightly easier for average players to excel with. For players who have mastered the basic skills of the game, the high-tier champions can at least give you an idea of what is valuable in a champion. For example, the Morgana case above shows the value of CC avoidance (and as such, ability to lane safely and avoid ganks), sustain, AoE damage, and CC.

Second, Tier Lists provide a good tracker of how play is progressing at higher levels of play (at least, when made by pro players). The champions valued at high tier are not valued because they're better, but because they are more valuable in/on typical teams.

But all too often, Tier Lists serve an all-too-different function.

Excuses

Gentleman Gustaf, it's not fair! Everybody always picks high-tier champions and wins!

People naturally form attachments to characters. And some people just like to play worse champions 'for the challenge'. Don't handicap yourself by establishing a stupid identity as some hipster who only plays champs who aren't cool. If you're playing to win, some of that is going to consist of playing champions that aren't your favorite. You wouldn't play an AD Carry that was just Ashe with no ultimate, or Ashe with 5% less. What lets people hold onto these illusions is that they like to think of each champion's strengths and weaknesses as balancing the game overall: sure, Morgana is good, but her cooldowns are pretty long and if she misses a binding, you can abuse that. While this may be true (that all champions have some weaknesses to abuse), that doesn't make all choices equally good.

Gentleman Gustaf, why are people such meta-sheep? I want to play the champions I like, not the same 6 champions every game.

People want to win. As such, they will pick champions they think will let them win. Often, these will be high tier champions (either because they've seen them do well consistently, or because they trust the authors of tier lists). You obviously want to win too (or you wouldn't be complaining about being unable to pick the champions you like. You also had the same chance to pick the OP champions too. So by not doing so, you effectively lost the first round of the game (champion select). You can't exactly blame the other guy for trying to win.

But the ultimate problem with the above is it puts tier lists as the scapegoats for your losses, making it hard to learn from losing. In a sense, it's the Elo Hell argument all over again, but with a different target. Everybody knows the feeling of losing to a really common, strong champion (or Black Cleaver), and thinking "I hate this high tier character! He's just a free win". And on the flip side, doesn't it feel ten times worse to lose to Karma? And thinking along those lines stunts your improvement by giving you a scapegoat (just the same as 'Elo Hell' does).

Common Complaints.

Gentleman Gustaf, what about player skill? The determining factor in games isn't champion selection, but skill!

This is an argument that gets thrown a lot, but is fundamentally flawed. For one, champion selection and knowledge of matchups is part of skill. If you're picking a champion that simply underperforms in your matchup, that reflects poorly on your skill. But more importantly, it sideskirts the issue. Yes, Azubu Blaze would probably beat Team Joe Schmoes even if they random all of their champions. But the game would be closer than if they had picked strong champions. And I'm sure that Tiger Woods could beat me even with a 10-stroke handicap and a crappy set of clubs. But without that handicap, it would be even less close. Essentially, by clinging to this mindset, you're forcing yourself to outplay your opponent every game, just to be on even footing. What's important is how you perform vs people of largely equal skill, and on that stage, champion selection is very important. Essentially, sure, you can win despite a handicap through skill, but that doesn't make it not a handicap.

Flaws

Tier Lists change all the time. First, there are nerfs and buffs which can change matchups. Second, the standard style of play can change (or people can become better acclimated to the old one). Either way, old strategies are suddenly less viable. This is not a sign that tier lists don't work, but rather a (good) sign that the game is not static.

Tier Lists are not the whole picture, of course. Tier lists are all about matchups, and of course skill comes into play. So if you're good at a low tier champion, but they would counter a pick on another team (or even be equally as good as a high-tier pick), go for it! So a champion with a 5.2-4.8 matchup with EVERY champion in the game would be a great first pick, but a terrible counter-pick most of the time.

And as well, there's a huge difference between competence with a champion and understanding of a champion. Sure, you can be competent with a champion, and be able to play them effectively. In fact, you can be competent with every champion, and then just pick whoever's Flavor of the Month at the moment. But at all levels of play, you can tell the player who really KNOWS his champion, getting every last advantage out of every situation. As well, there's champion style. Voyboy plays very aggressive, dominance based top champions, while HotshotGG is known for his farmfests. Obviously, the same champions may not suit both of their styles.

This is why we shouldn't be that shocked when low tier champion like Lux show up in tournament play; they may have been picked as a counter, or the player could just be really strong at the champion/style. It doesn't mean the tier system is broken, and it doesn't even necessarily mean the champion should be a higher tier.

So Tier Lists are a really useful tool, both as guidelines and as insight into high level play. But as with any tool, don't let it guide your play entirely: focus on improving, not on winning. In other words, there's nothing wrong with tier lists as a source of information, but don't shape your mindset around them.


 For more of my work:

-- Find old posts @ the RoG forums and new posts every Wednesday (3 PM) and Sunday (9 AM).

-- Feel free to find me in the "A DIFFerent View" chatroom on the NA server.
-- Contact me at [email protected]
-- Follow me on Facebook and Youtube for updates on all of my new articles, videos, and streaming.


 "Song" of the Day

 

37

Comments

  • #32 CatBurglarNami

    Pick Sagat

    Last edited by CatBurglarNami: 1/3/2013 12:38:09 PM
  • #30 Thaddeusmike

    I do have one complaint about tier lists to add:

    DO NOT FORGET ABOUT TEAM COMP

    I don't care how OP your champ is, a team with no damage split is at a big handicap. A team with no tank is at a big handicap. A team with no hard cc is at a big handicap. A team that can't decide if it wants to poke or all-in is pretty risky. All of the above can be done, but only by taking extra risks.

  • #28 RoakOriginal

    Feeeling 10 times worse when losing to karma, then to some sickhead roghtclicker? definitely not... I lose to karma mid as ap kog once, and it wasnt bad feeling... i admired that guy for being able not only to beat me, but he carried his whole team  and we had hard time to win (luckily with my kog full build teamfights become mroe like snipe-fest, so we turned it).. but if i am goin mid/top vs champ like LB/teemo, which is based on 0 skill and u just deal dmg by clickin 2 buttons 3 times without aiming, and those people are zoning me, even when they are unable to lasthit, it is frustrating... they mostly lose anyway, but it doesnt change f***ed up feeling after laning phase...

  • #24 Canvasofgrey

    Considering the tier lists haven't really been updated in forever, I feel the need to understand them is moot until it is.

  • #37 dumbitdownjr

    Yeah that's the biggest problem sometimes i look at a tier list and then i notice the last time it has been updated was 2 months ago. I understand they're really busy though.

  • #20 TreeBurrow
    • "This"
    • "I agree"
    • "LoL"

    The way I see it is that Tier lists are simply one persons opinion of the state of the game/META... Sometimes that person has in-depth knowledge about the game, sometimes that person is highly regarded as being a knowledgable player... Other times it's just some random guy yelling his opinion at people... Basically, sometimes Tier lists can feel as if they're accurate, they can feel as if you're gaining insight by reading them, because the opinion is sought-after and valued... Other times it's just a waste of time, Other times it's just one person telling other people what he thinks about each champion... Tier lists can be biased and "wrong" and terriblly inaccurate... But lets face it, most people have slightly different opinions about each and every champion, when was the last time you looked at Elementz Tier list and decided that you would have put every single champion in the exact same position as he did? Chances are you would have placed your favourite champions, the champions you're most experienced with slgihtly higher up, or the champions that you rarely play would be placed lower...

    I could come along, lay down a Tier List, have a huge amount of explaination about why each champion is listed in that specific order... Who would read it? Better yet, why would they read it?... I have 0 credibility, but i'm still entitled to my opinion, whether it's accepted or not... That's all a Tier List is, it's one persons opinion, you either agree with it or you don't... It can be used as a rough guide, some abstract point of referrence, but it won't always apply to you, people have different playstyles, people use champions to different effect, some people can use certain champions better than others, LoL is a hugely diverse game... It's great... But basically, no-one should follow Tier Lists to the letter, it should just be used as a rough guide at the most, perhaps it makes a good point of first referrence for new players, I don't know, I haven't been a new player in a long time... Every few months I'll skim a Tier List just to see what other people are thinking at the moment, but i won't change my playstyle to suit what their opinions are...

    I don't believe in following Tier Lists, just as I don't believe in winrates... And in my opinion, you shouldn't believe in them either, they don't know you, they don't know how you play the game, they don't know your skill level, so why should you know them?

  • #21 GentlemanGustaf

    I feel like this problem is solved by only reading tier lists written by really good (high-Elo or pro) players.

    Yes, I would have placed my favorite champions higher, but I'm not even remotely qualified to do a tier list. It's why I've never done one.

    Yes, people have playstyles. Yes, people have favorite champions. But that just is their ability to execute a champion, not the champion itself. Every champion has certain capabilities, and the question is whether the best played X champion beats the best played Y champion. Refering to things like individual preferences misses the point of a tier list. The point of tier lists is to provide abstracted, objective rankings, not tell you who you would play best.

  • #26 Wick3d_Rav3n

    What pro players actually write tier lists apart from Elementz and Stonewall? I have not found anything in my searches :(

  • #36 Mockstar

    Absolutely agree. However, some champions are viewed as a lot weaker than they really are. So many people view AP Nidalee and AP Ezreal as troll picks, and neither of them are top tier for anything, but when I see someone pick them, I get scared, because they're either very confident or trolling, and its usually confidence.

    As an AP Nidalee main, I see her as much more viable than most players I talk to. Even Bischu, commonly said to be the best Midalee in the world, says that she loses every single lane and is pretty much useless in her own lane before team fights and roaming start. I typically wreck my lane with her brutally, then gank when I feel like it. I have found only one lane that I can't win with her, and that's Brand. Niche pick. Rare as hell outside of his free weeks and new skins. So I would put her as a low tier 1 high skill cap champ. Elementz has her at the top of tier 2 I believe, mainly based on AD Nidalee, but he mentioned being impressed with her mid laning as well recently.

    Now, I don't think I should choose not to play my far and away strongest champ in my first few ranked games just because she is viewed as weaker than most champions. I mean, there were still Eve players after her nerf that could stomp with her.

    The tier list is a basic guideline to first picks. The later your pick, the less important it is and the more important counter-picking and a large champion pool becomes.

    I can first pick Nidalee very comfortably, and it usually gives my top lane a chance to counter pick that Jayce or Jax pick to counter the Nidalee they think is going top. Nidalee Tier 1! Thanks for the great article, good read :)

  • #19 sgtcolon

    Whilst this article hits the nail on the head with a square and resounding 'thwack'. I've been struggling recently to keep up with the changes and general volatility that season 3 has brought. Perhaps this is because I mainly have experience in bot lane and have recently moved to mid lane, leaving me with precious little experience with specific match ups. Although, despite looking over websites like ChampionSelect, I've found it difficult to get reliable information on match ups, counters and playstyles to make up for my shortcomings. So, this brings me to two questions:

    1) Are there any reliable and up-to-date resources for top/mid lane picks? Including counters, play-style tips for specific match ups etc. If so please share them!

    2) How hard would it be to create a RoG moderated guide or database for picking champions and lane match ups? Ideally with a commenting or voting system (just in some way restricted to up-to-date and reliable information! Hence tying in with the RoG userbase etc.).

  • #22 GentlemanGustaf
    1.  I tend to go into LoLPro guides on champions and look into their 'counters' tab.
    2. I wouldn't think this would be useful. poll based guides like that are pretty terrible, imo.
  • #25 Waaargh

    There is a page suggesting what champions to use for counterpicking, Championselect, based on general opinion of the visitors there. Lolkings also have added it to their guides. Thing is you don't know if it is a competent guide maker who did the work and/or how much experience the person has against niche counterpicks.

  • #27 sgtcolon

    @GG: Thanks, I use lolpro all the time for guides, so not sure why I never noticed the 'counters' tab!

    @GG and Waaargh: Yer, I guess you could argue that's mostly why things like championselect are unreliable. It depends on who is commenting / upvoting etc. I guess I just thought I could get behind a consensus from the RoG community a bit more.

  • #18 Benegesserit

    Exactly what "Song" of the day am I looking at?

  • #17 FleurDeLiz

    The problem of tier lists in team composition games is exceptionally simple: they promote the generalist over the specialist, almost regardless of how they fill a team composition. The safest picks, as you said, tend to be rated highest irrespective of how well they would go together with each other.

    You used Lux as an example that shouldn't really shock people, but I really want to break that down a bit. Lux has exaggerated pros and cons due to her dual specialty.

    Lux Pros: Long potential range. Great base damage within autoattack range due to passive. absolutely ridiculous ult cooldown at level 16.

    Lux Cons: All of her spells are under-par in terms of scaling from AP and the safety of popping her passive goes down as the game wears on. Also exceptionally weak to anyone that can jump directly to her after she has used Light Binding.

    She is very strong in poke scenarios and can fight well from the edges of a fight while being able to do surprising damage early to anyone caught within her attack range that can't retaliate effectively. She is very weak when the opposing team has one or more heroes that can drop on her and end her before she accomplishes much.

    And I'm pretty sure almost everyone that has played the game for a decent length of time has seen one or more incredible Laser snipes, of champs, dragon, or even Baron.

    These factors tend to pull champions down tier lists, yet there is a space in which Lux does excel that other champs do not. Very few champs can actually fight competently at extreme range with their entire kit, yet that is exactly what Lux is able to do.

    I do find it really interesting that the vast majority of champions in this game float around a 48-52% win rate at varying pick rates at the platinum ELO level and above. The exceptions above are generally easy to use champs that are very strong in the current meta and the exceptions below generally either have glaring weaknesses with strengths that are hard to leverage or have simply been overnerfed deliberately. (the only champ below Rengar/Diana in win rate is Karma, who doesn't even have a statistically significant sample to base a win rate off of, this is how unpopular Karma is)

  • #23 GentlemanGustaf

    in solo queue, perhaps, but in tournaments, we consistently see champions with much higher win percentages.

     

  • #29 blackraven1425

    Yes, although you have to consider that WE is going to boost TF's pick/ban and win rate at IPL because of Misaya's play with him. Typically, tournaments are skewed by the winners' champion preferences like that for both pick/ban % and win rate. On top of having those preferences skew the numbers in their favor, the entirety of most tournaments isn't going to be all that significant statistically, so I'd take any extrapolation of those numbers with a massive grain of salt.

  • #31 Thaddeusmike

    I think hashinshin had a post about this. Results from any single tournament aren't significant. To use my favorite example, Gragas wasn't OP season 1, Shushei was just that good. On the other hand, Shen has been a solid pick in tournaments since his buffs. The consistent record over time (with no significant changes) is solid evidence that Shen is really good.

     

  • #33 FleurDeLiz

    Tournament play brings its own bugbears. Unless a champ is ban/pick in nearly every match, the odds of even having a statistically significant sample on a per tournament basis are very low. I mean, let's take a theoretical situation where, say, Rumble gets 3 total matches throughout an entire tourney. For a niche champ, this level of being picked isn't bad. But one thing you can't really do is draw very many conclusions from a winrate at this pick level because they're simply not being picked enough to isolate them as the number one factor determining their wins. The team aspect of the game adds confounding factors that hinder this.

    Now if you were to aggregate tournament data over a period of time (let's say 3-4 months), you would have a better chance of having those good sample sizes that would let you draw conclusions on trends and such. It kind of interests me that RoG doesn't really have someone doing this even though the blog does have someone that does some amount of tournament coverage.

  • #34 GentlemanGustaf

    The problem with that sort of data is that the evenness of tournaments varies wildly depending on the participants. Once the Circuit starts up, I'll definitely do something like this.

  • To post a comment, please or register a new account.
Posts Quoted:
Reply
Clear All Quotes