Mobility and CC Creep

Mobility and CC Creep

Gentleman Gustaf here, following up on a post from last week. Last week, I looked at whether newer champions were more likely to be considered high tier. This week, I want to look into something more specific, namely whether or not newer champions have more CC or mobility. It's often alleged that newer champions have more mobility, citing champions like Ahri, and more CC, citing champions like Nautilus. While it is true that some recent champions have had high CC or mobility kits, is this a new phenomenon? And is it as strong as it seems? As a proponent of the idea of Mobility Creep, I was excited to look into this topic to verify my opinions. Here's what I found.

Mobility

First off, what is mobility? Mobility generally has three components; base movement speed, gap closers, and movement speed buffs. I have chosen not to post my findings on base movement speed here; they are highly uninteresting. Gap closers come in two varieties, untargeted and targeted. Movement speed buffs, on the other hand, are either single target or AoE (Nunu's is double target, but is the only one of its kind. For the purposes of my calculations, I considered it single target). So how has mobility changed over time? First, let's look at gap closers. There are a few important things to know about all the graphs to follow. According to these graphs, there would appear to only be 75-80 champions. This is because I have counted champions released at the same time as having been released together; they still count as separate data points, but they are all the same number release. The line on each graph is the best-fit line, with its equation in the top right corner. The more erratic 'curve' that goes above and below that line is the 15 champion average; it averages the 7 champions before and after (as well as the champion itself).

We can see that there does see to be a slight increase of gap closers over time.

But Gentleman Gustaf, how can .0018 more gap closers per release be significant?

It may seem tiny, but consider .0018 as a percentage of the y-intercept. It would take about 130 champions for the amount of untargeted gap closers to have doubled. This would be quite the significant increase. But for targeted gap closers it's even more significant. The average number of target gap closers increases by 100% (on average) about every 40 champion releases. In total, that number is about 75; the local average number of gap closers is on the rise, and appears to have (nearly) doubled. On the other hand, if we look at Move Speed buffs, the numbers are much different:

It take almost 600 champions, with the current trend, for move speed buffs to double. So whatever selective pressures there are for gap closers, they don't exist for movement speed buffs. What exactly are these selective pressures? I posit (following many others before me), that the selective pressures are kite-heavy AD Carries, either those with slows or those with gap closers. Because most AD carries get PD, and as such, are quite fast, and as well, have either an AoE slow or a flash, it can be very hard to track them down. If you've ever heard 'Udyr's weakness is that he's easily kited, this is what people mean. Any AD Carry with a slow or a blink is going to make Udyr chase a long ways to catch them; throw in an item with a slow, like Trinity Force, and all of a sudden the champion may just be too slippery for an old-school bruiser like Udyr to catch. So what happened? Over time, more champions needed gap closers or CC reduction (looking at you, Irelia), just to be able to keep up with the AD Carries. All of a sudden, we have a competitive game. Movement speed buffs  are going to lose out to slows, due to the way that the move speed soft cap works, so you need a gap closer to get onto or near an AD carry (of course, there is also the reverse method, CC). However, as more and more bruisers get gap-closers, AoE slows are no longer as effective, as bruisers are no longer just running full-tilt at you. So AD carries needed to evolve to have more blinks or dashes (or at least, more consistently to have them). If you look at recent AD carry releases, most have a blink or dash, and those who don't (Varus or Draven) have been somewhat lackluster. I'm not sure that their lacklusternss is a result of the complexity of their kits or their lack of mobility; only time will tell.

CC

Now for the other side of the equation: CC


CC has undergone quite the interesting change. There aren't MORE CC ablities, per se (well, slightly), but single target CCs have decreased significantly over time, with a near-equal increase in AoE CCs. As well, single target skill-shot CCs have increased slightly over time. Overall, it would take over 1100 champion releases (at the current trend) for CC to doubled. But the shift from single target to AoE CC is significant. First, it is interesting from a skill perspective. Single target, targeted CCs are much easier to land and have much less counter-play, making them less interesting, while AoE CCs have a lot more potential gradations of skill to reveal. That is, there's not much skill at play in between Taric stunning two different targets. But a good Cho can save his rupture to hit 3 targets, or blow it early on one. But second, it really increases the amount of CC, while decreasing its reliability. This gives us exactly what we want. The overall increase in CC is good; it lets you have counter-play to the (now very gap-closing) bruisers in protecting your carries. However, it is CC that requires more skill to use, allowing for more nuanced gameplay.

The increase in slows, on the other hand, is largely irrelevant with all of the gap closers.

So where does this leave us? Well, the top three AD carries at the moment (Graves/Corki/EZ) all have blinks or jumps or dashes. But champions like Kog'Maw or Ashe are still quite viable; they just need the right compositions around them (isolation/burst comps for Ashe, and protection compositions for Kog'Maw). Both types of compositions require more CC to be available, which, conveniently enough, there is now. So is there more mobility now? A lot. Is there more CC to counter it? Sort of. But what's most interesting is this data in light of my previous post. There has been an increase in mobility, and to a degree, in CC. So how does the game remain balanced around new and old champions? In large part because of counter-play and specific compositions. If one champion on the other team has a lot of mobility, you don't necessarily need a champion with a lot of CC to counter it; you can also just get slightly more CC on every one of your picks. If one champion gets a lot of CC, you get a champion with CC reduction, and so on. And those CC and mobility increases aren't done in a vacuum. Graves may be more mobile than Kog'Maw, but Kog'Maw does more damage late game. Yes, there appears to have been some mobility and CC creep. But since, the data doesn't suggest that there has been power creep, other factors must be at play.

As a slight sidenote: I did note a slight amount of bruiser and AD carry power creep. Nothing intensely serious, but still enough to be notable. For bruisers, I think it is probably just a result of the previously mentioned jungler creep, as many junglers are bruisers or tanks. For AD Carries, given that this power creep is MUCH more notable when you discard Varus and Draven, it is possible that mobility creep has created some power creep with regards to AD carries. As well, the only new (last 6) AP carry who is high tier (Ahri) is made of mobility. I am not saying mobility creep is not a factor in game balance, just that there is no reason to think that there is significant power creep outside of isolated cases, anecdotes, and very specific contexts, in large part because the game has to be balanced around not just champions, but teams and compositions.

19

Comments

  • #15 Vazzu

    Just wondering, who are the champions with 3 gap-closers? Also does a skill which can be activated multiple times count as that many? (Ahri) Also how about the stack system? (Akali) And should Riven have 4 untargeted gap closers then? Great article as always, really enjoying your posts.

  • #16 GentlemanGustaf

    Yes, yes, and I decided no for Riven because the distance she gets is about equal to one gap-closer.

  • #11 Naraxy

    Is Zyra not still considered high tier? She's not very mobile.

  • #6 Dub_Rio

    Are those beta values significantly greater than 0? It could just be variance that you got a beta of .0018. To be honest, coming from someone with a stats background, you seriously may wanna check that cause if some of those beta values aren't significantly greater than 0 your results are null. 

    I mean what's the r^2 of those models? How much of the variance in mobility/gap closers is explained by your independent variable? Where's the ANOVA and what's the p-value for these regression equations? 

  • #7 GentlemanGustaf

    If you're ok with a random ordering of the same numbers as a control, I get a p-value of just under .1942: not exactly scientific, but not necessarily a sign there's nothing, and an R^2 of 37.54%. *shrug* I'm glad to have another math nerd, but as you can see from the below comment, I'm already being pressed to simplify my articles, so I tend to leave the more extensive math out and just give a simple graph that people can look at and draw easy conclusions from.

    But as far as the general tone of the article: I don't believe there is any power creep, and while I believe that there are small amounts of mobility creep (and even smaller amounts of cc creeps), that they are done in ways that don't affect balance. This is why I'm super hesitant to go all out defending my data, because ultimately, I don't think there is a huge correlation between time and power level.

    Of the graphs I presented, only two were even pretend-significant (below 25%), and those are Targeted gap-closers and crowd control single/AoE (and actually, mostly single, which is why those are the only two I commented on.

    Last edited by GentlemanGustaf: 9/19/2012 11:54:06 PM
  • #8 Khoral

    If possible, you could add some sort of "spoiler" tag in the article, that would allow "math nerds" to expand some part of the text with much more in depth math data.

    This way, people without math knowledge would stay away from it, where people like Dub_Rio would have their answer :)

  • #14 Hahano9

    I don't feel the complex math is the issue. It's fine to get math nerdy about things, but you're diving into statistics that I feel don't really sum up people's feelings entirely. Do people think Darius is OP because he has a <50% win rate? Do people feel the new champions have power creep because of an analysis of the win rates and CC/Mobility trending? Or do they feel that the new champions are stronger because the last few releases are:

    Rengar - Gap closer, cloak, slow, snare, sustain

    Diana - Gap closer, slow, displacement, gap closer reset, high burst

    Zyra - reliable snare, AoE knockup, huge burst

    Jayce - knockback, gap closer, team movespeed boost, poke, tanky, great laning, etc. etc.

    I think taking a mathematical approach to an extremely complex set of interdependant variables is a mistake. You can certainly say "hey, I ran some numbers and this graph shows that there is/isn't a trend here." That can solidify your otherwise baseless opinion on something. Math doesn't really have a place without explanation and relation so copy pasting some Excel sheets does literally nothing for most people, even if they can read the graph.

    You need to take a step in explaning further what that particular graph means to your entire point. You did a much better job with this article, but half of your analysis is mathematically saying it would take XXX champions to get to double CC. What significance does that have at all to anyone? Sure someone that is more familiar with graphs and statistics and logic can see that it's moving at a rate that is/isn't worth yelling about, but to the casual observer you're missing a massive opportunity to tell a story. Use those graphs to solidify your argument, not just to show a trend and explain what the trend is.

    Approaching something like power/CC/mobility/whatever creep can't be done mathematically and Hanshinshin's article on passives also addresses something similar. There is no complaint that CC is on the rise because Sejuani has a bucket of CC and...nothing else. The issue of new champions feeling overpowered is a combination of release balancing, champion kits, people's ability to deal with them and many other things that can't be expressed on a graph. Instead, I would try approaching these topics in a more philosophical and inquisitive nature with some math to show you're not a chump pulling stuff out of your butt.

     

    EDIT: Think risk analysis. The numbers can have extreme variance (like the dollar value of a life between agencies) due to infinite variables. To combat this, there are generalized conclusions and tests that may or may not prove something, and plenty of publications on their findings. The publications, however, are very little about statistics and very much about how, why and what the variables and conclusions say.

    tl;dr I like the articles, but I feel you're trying to express too small a piece of a pie to have a statistical approach. Readers aren't looking for a mathematical dissertation, they want you to grab their attention, massage their sweet eye holes and express your opinion. Math solidifies this approach, but shouldn't dictate it.

    Last edited by Hahano9: 9/20/2012 12:09:11 PM
  • #17 Dub_Rio

    Good point on keeping things simple. Most people probably won't be able to follow higher end maths/stats tests. 

  • #18 GentlemanGustaf

    @Hahano9

    Thank you a lot for your feedback; I will really try to take this into consideration for my math post on Wednesday.

    Last edited by GentlemanGustaf: 9/21/2012 12:30:18 AM
  • #19 GentlemanGustaf

    Quote from Dub_Rio »

    Good point on keeping things simple. Most people probably won't be able to follow higher end maths/stats tests. 

    Yeah, if you want to see some of the random math I end up doing and discarding, I'll consider doing the 'spoiler' thing, although that's a lot of work. I've also plans (at some point) to get the math up somewhere else.

  • #5 supportking

    anyone a tl;dr?

    Last edited by supportking: 9/19/2012 9:02:57 PM
  • #13 l3rowncow

    just look at the graphs with pretty titles and lines and stuff...

  • #1 Jolan

    So what would the best generalized idea be, here? Nerf all AD carries so that they are not as mobile as before thus lowering the effectiveness of kiting? Then nerf all the gap closers and leave the MS steroids as they are?

  • #2 GentlemanGustaf

    They've already sort of done that (reduced the movement speed on PD a little back). To be honest, I think the amount of power creep is minimal enough to really be a negligible amount.

  • #3 Jolan

    I did approve of that nerf, but I was thinking something oriented more-so towards flashes and blinks. Just like how Corki took a CD nerf to his Valkyrie, I was hoping for something like that on Arcane Shift.

  • #4 GentlemanGustaf

    With Ezreal's recent dominance, I'd be very surprised not to see at least that, especially given the CDR on Ezreal's Q

  • #9 Zulkyr

    It's worthwhile to note, GentlemanGustaf, that while you have been accruing the number o releases to math it out, you may note that not every release gets integrated into "tier-wise" gameplay. Meaning: doesn't matter that we didn't have a significant enough mobility creep in AD Carries, when the only "viable-ish" ones have those dashes/jumps.

    What I'm trying to say is: while the effect may not be generalized in regards to mobility and CC creep, you have to deal with the fact that these are dominant champions these days, which severly hinders the viability of non-mobile, non cc heavy champions. And while you praise the pick-counterpick aspect of "Ashe and Kog'maw can still fit the right comps", the critique we can make is that: why are there champions that have been reduced to niche or non-use while others are just able to fit anywhere and still be all over the place?

    This is both sad and stupid. It creates the trinity force of AD Carries and puts Vayne, (my favorite) Draven, Sivir, (tristana may yet be rediscovered), Varus and pretty much every non-jumping AD carry out of order. It also creates situations that are extremely unfun. Take the case of high level top and mid lane. It has shifted towards boring/safe matchups of champions that are just impossible to kill or drive out because you just have to protect yourself against jump-happy junglers, which in turn makes the junglers turn away from ganking and into fast-clearing, control and counter-jungling. Which is fine by the junglers, but very annoying for people who like aggressive top and mid lanes.

    This is turning into article size, so wrap-up: while there may not be a notable amount of gap-closers and cc in new releases or over time, the relative strength that these mechanics posess is giving them an abnormal "weight" in the game, making any release that contains one of these that much more viable AND putting a lot of fun champions and gameplay idea to the sidelines. Draven was a double-fail? Hell no, it's one of the most engaging and fun champions to play, but the way people jump in your face today makes your lack of a jump AND your need to run out of your way to catch an axe or lose on DPS an impossible mission in lategame teamfights. This is bad for the game, and new releases that don't find a way around this will make it worse.

    P.S.: Just saw Kha'Zix and guess what, just like Rengar, he has both JUMPS AND STEALTH.

  • #10 GentlemanGustaf

    This is true if you look at the slice of meta that is today, but just a few months ago, Kog'Maw, Urgot, and Graves were the strongest AD champions, and Ashe has always been a strong pick despite her lack of a steroid. Without any dash, and without any steroid, she is still considered tier 1 by Elementz. Vayne was stupidly OP on release, and the reason she isn't played today isn't really her lack of a dash, but the weakness of her laning, and when Caitlyn was being picked all the time, it was because of the strength of her laning.

    Right now, yes, the meta appears to be prioritizing mobility. Why? We're seeing a lot of compositions against which a blink or dash is crucial. But just a few months ago, it was prioritizing CC and hard engages, and before that, it has prioritized poke, and M5 even briefly showed us the power of just having a tanky composition.

    Its easy to look at a snapshot and say there's more X or more Y. But trends matter, and shifts matter, and it's only recently that EZ/Corki/Graves have dominated.

  • #12 Zulkyr

    I'd take a longer look at that. Ashe is still tier 1 because while she doesn't have that dash, she is probably the strongest kiting champion of the game, which, if you look at it, is related to mobility. People look sideways at caitlyn but she is both strong at kiting and has a (limited) dash, her lategame is by no means supbar, but her damage cannot compare.

    Graves and Corki have ALWAYS been strong, the only ezreal crept up lately is because both got nerfed heavily, which hit their lategame potential.

    I do believe meta shifts can bring things that were under the water, but I also believe that mobility and crowd control are two factors that require very delicate tuning. Right now, I think this is something that requires balancing on both sides: AD mobility nerfs would make other options available and CC/Mobility adjusts on bruisers and mages would also help that "in your face" feeling on non-jumping ADs.

    As usual, time will tell. It's just a little bit unnerving to be playing the same three champions over and over again when you see that Olaf/Jax/Irelia/Ahri/Gragas pop on your screen.

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