A MASSIVE design exploration of Ezreal and how to rework him

PotatoMine·8/8/2018, 9:18:34 AM·2 votes·1,480 views

Hi Riot and League of Legends community,

I'm a passionate Ezreal player and game designer that is thrilled Ezreal is FINALLY getting an updated W. As someone who has put a lot of thought into this, I wanted to share the design I’ve come up with over the years in hopes I will be able to add some ideas to the conversation. To this end, I’m going talk about what I would change(or at least test, theory only gets you so far, playtesting reveals things you can never foresee, but theory is all I’ve got to work with).

I’ve spent the past two entire days on this post. Over a dozen hours. To any Rioters reading this, please, it would mean the world to me just to know you read it. And please, forward it to the team actually working on Ezreal.

Prologue: This will be my one chance to speak my mind after years of thought, so I’m going all out. In a normal design environment I would get to have a collaborative and iterative back and forth design discussion with designers at Riot, but since I don’t work at Riot I don’t have that luxury, so I’m front loading everything. Lastly, I have a reading disability, so sorry if my spelling or grammar is bad.

Part One: Introduction to the Mark mechanic

New W:

“Ezreal fires a wave of energy in a line, dealing magic damage to enemy champions struck, and mark them for X seconds. Basic attacks and spells detonate the mark, dealing bonus physical damage.“ Skillshot damage : (Flat based on skill rank) + (Flat based on champion level) + (AP Ratio) Mark damage: (Flat based on skill rank) + (AD Ratio)

(It’s a similar effect to Lux’s passive).

My thesis is this clean understandable effect has deep design implications that address basically every problem we are trying to solve. This essay works top down, analysing the problem at a high level and then will discuss how this effect solves the problems explored. Fundamentally, our goal is to make W be a satisfying spell that still feels like it’s the same ability.

Let’s start by restricting our design domain, why keep it the same ability? Most AD players will tell you they don’t really care about the current W, but I argue that is not actually true. The core geometry of the current W is actually a really good fit for Ez. It reinforces his spell slinging fantasy very well, and feels different from Q since it goes through minions and the hit box does not increase with range like Q. The problem is not the spell geometry, it’s that the spell does not feel impactful, this is both an art problem and a mechanics problem. A good way to demonstrate this is to play AP Ez, and suddenly W is a much more satisfying spell because it’s actually relevant. Additionally keeping the same geometry reduces risk, Ezreal is a very popular champion who already works well, we want a low risk, high reward solution(also reduces development time which is nice).

If we are keeping the same geometry, then we need to make sure Q and W feel like very different spell, even though they are both line skillshots. A large portion of this is solved with the geometry difference as I talked about before(over minions, constant hitbox, but also range, cast time and cast animation). It creates a great foundation but we need to actually build on that foundation. So let’s think about about what role each basic spell in Ezreal’s kit plays and how W should fit into that. Q is long range poke, it is his core damage ability and is what evokes the “minion dance” to snag people through minions. Casting Q is his core game loop, his “bread and butter” spell so to speak. It does this well and does not need help from W to create his satisfying game loop. This means W should not be a spell you cast frequently, it’s a spell you think about and consider, it’s a less common high moment that is used to make plays. But it’s not E, which is even more of a rare high moment. So conceptually W is the middle ground, it’s the spell you cast in a skirmish but casting it does not mean you’ve gone all in. This is reinforced with the shorter range, longer cool down, high mana cost, and the fact that it only hits champions.

Our goal is to improve and highlight what W already does well, not make it do something different. BTW at my fellow designers, I’ve skip graphing the multi-layered pacing curves of everything talked about above, instead will invite you to simply visualize it yourself. Now, we get back to my proposed design. Let’s recap:

Starting from the easy pickings, the fact that a large portion of it’s damage is locked up in the mark means it’s not a poke spell like Q. To get the full effect you must at least have enough commitment to a fight that you hit a target with two things(spell/aa), that means at least 2 cast times and two effects on cool down.

If that effect was Q then you have both your main damage spells on cooldown now, you are in 1000 range(more or less), and had a clear shot through minions. This is what I’ll call the mid range case, what’s important is the mid range case feels very different from the long range poke case that is most often seen with Q. It also feels like a satisfying moment, if you pull this off and take no damage you get a large satisfying chunk of damage and “get away with it”. You feel like a badass spellslinger who tempted fate and got out free. If you do take damage then congratulations our spell design has incentivised our players to fight each other, and fighting is an important part of League’s core design loop, and something that can be difficult to enact in uninteractive safe poke champions.

If we detonate the mark with E, Ult, or aa then you are in 550 range or just cast E/Ult, all meaning this is a serious fight. In this case the mark mechanic helps makes the Ezreal player feel rewarded for actually getting into this fight and again makes W feel like a very different spell from Q. It also helps to combine with the other high moments from other spells, if I cast Q W E Ult and hit everything I just got multiple overlapping high moments which creates a single massive high moment. This burst of damage also reinforces Ezreal as a spell caster, he’s not an auto attack based champ like Cait or Vayne, he is a spellslinger.

To condense all this, the mark mechanic does not really do anything new, it simply reinforces and polishes Ezreal’s already loved and well established core gameplay loop.

But there is one more case we have not talked about, what if we cast W and don’t apply the mark? If the base damage is trivial we just shoot ourself in the foot, the whole goal was to make W not feel wimpy. It’s easy for a player to think of the skillshot as the spell, and then the mark damage as part of whatever thing applies it. Similar to how if people get hit by a Zoe bubble, then Zoe lines up a massive Q, it’s easy for players to say “WOW that Q did shitloads of damage” rather than realizing that a large portion of that damage was actually from bubble. That means we need the base skillshot to do enough damage to feel worth casting, particularly in lane. At a deeper level, if a player casts W, hits an enemy and does not apply the mark they need to be happy with that. They need to not have the current problem of having to ask themself “was that even worth the mana?”. Simple solution right? Just give both the base damage and mark enough damage to be satisfying? Sadly it’s not so simple, ADC Ezreal is currently balanced around W being a trash spell(and we’ll get to AP). So we will need to talk about how and where to siphon power out of other places in his kit, but that is for later. Additionally if we make the base damage on W too high then it will just become a poke spell that has this extra mark effect, we need players to feel like the skillshot hurts enough to be feel ok with casting it with it’s base damage alone, while having their hunger for damage not be totally filled, which then lets the mark act as that juicy carrot to fill that hunger. This is all just a number game and I’ll talk later about the role the AP and AD ratios play in this, but what is important to the larger discussion is it can be done. If we get the numbers right then this creates a powerful gameplay loop for the spell.

This concludes the backbone of this design exploration. We explored why keeping the same geometry of W is important, the implications of doing so, and then why the Mark design accomplished the goals we discovered. But don’t be fooled, we are far far from done.

Part Two: AP Ezreal As someone who primarily plays AP Ezreal, it is a given to me that AP Ezreal is something that should be very strongly considered in changes made, but I understand most people think of Ezreal as an ADC for bot lane, not a mage for mid lane. In an optimal world we want to create a mechanic that simultaneously addresses the needs of AD and AP seamlessly. Luckily, the elegante beauty of the Mark mechanic is that it does this.

For AD W is a blank slate, but that is very much not the case for AP.

Before we can get into the details, we need to talk a little bit more about how AP Ezreal is actually built, because 99% of people have no idea. There are actually two very, very, different types of AP Ezreal. There is burst AP and poke AP.

Burst AP is the one most people think of when they think of AP, you normally rush a Gunblade, take Thunderlords as a keystone and max W first rather than Q. This build will fuck you up mid game, it can easily full to dead someone even without ult and is basically an assassin. But. It has a critical weakness that prevents it from actually being good at winning games; It cannot push a wave. This is the fun AP build, the one most people mess around with in normal games where they don’t care as much about winning.

Poke AP is the less fun, but much more powerful AP build. The main difference is you max Q, get lots of early CDR and push waves by spamming lots and lots of Qs. It cannot full to dead people almost at all, but can actually push a god damn wave! Fucking hell you have no idea the shit I’ve put up with as a serious AP Ezreal player, and how much of a godsend having decent but still below par waveclear has been! Shout out to Riot for making Ludens apply on Q, thank you man, thank you. If you can’t tell this is what I play, and while we are here let’s throw in the fact that I’m a high elo player for a little ethos (I float around Plat 1 / Dia 5).

Deep breaths. Ok.

Moving on, there are lots of slight variations on burst builds and poke builds, but what defines the difference is do you max Q first or W first? Q gives waveclear, W gives burst.

On poke builds for most of the game W is about as worthless as it is on AD, so let’s focus in on burst builds because it’s their bread and butter spell(we’ll get back to poke don’t worry). The fact that you can poke over minions in lane is a big draw to this style. It makes him feel like a full blown mage and is important to give AP threat even when people are behind minions. You get to have the spell slinging fantasy of Ezreal, without all this crap about having land Q through minions. It’s fun and satisfying, but the catch is it’s not actually very healthy design.

This build basically lets the Ezreal player circumvent of lot of the counterplay built into his kit, and while that is not necessarily a bad thing there needs to be different counter play to make up for it. In this case it’s a large net negative. W is longer range than most mage abilities, has a fast cast time and a fast missile speed. It’s very common to tag enemies going up for minions without giving them a fair chance to deal with it. This needs to be dealt with, but we have ways to deal with it without gutting this playstyle.

If we look deep enough, we find the problem is not the ability to poke relatively safely, that’s like Ezreal’s thing man. It’s the fact that a W maxing AP player just does SO much damage on W(and the art does not communicate it’s threat). It’s not like he does not take any risks at all for poking with W, and it costs plenty of mana currently(mana costs are their whole entire conversation that I will not be going into). It’s ok for him to have some slightly abusive poke, everyone gets to have something a little abusive, it’s part of what makes each champion feel special. We just need to tone it down a little. And guess what! We can do that, with the Mark! But before we get into all that, we need to go on a long tangent, but we will get back to this don’t you worry. Enter another actor to this conversation:

Change two : Slightly Increase the AP ratio on Q.

Why? Since the primary way AP Ez pushes is with Q spam, this will increase his wave clear. The goal being to make it easier to 1 shot caster minions and easier to 2 shot melee minions at different stages in the game. I’m thinking .1 or .2 AP ratio, but there is a lot of math that goes into this. It can’t be too much or Lich Bane Q’s later in the game will hurt too much. This is a small change that will make a big difference(There is a massive difference between one shotting a caster minion, and doing 90% of their health). This could also be done through increasing the base damage of Q, but that has implications on AD. Getting this right is a massive amount of math, but it can be done.

Point is we pull some power out of W’s poke and put it in Q, in doing so we kill two birds with one stone, we reduce the toxic case of W poke and give a massive quality of life buff to pushing power.

However there is a problem. The goal of giving Q a slight buff is to reach the damage breakpoints needed to one shot casters and 2 shot melees right? That means this only works if you max Q first. That of course means you are not maxing W, and that means W will be rank 1 for most of the early and mid game(depending on if you max it 2nd or E 2nd), and a rank 1 W is going to feel very wimpy! Well shit! That was the problem we wanted to solve in the first place! And this is not just a problem for AP, AD also maxes Q first. We want to keep it that way too, AD players are used to maxing Q and it is a lot healthier for them to be maxing Q first most of the time.

Fear not! We have a solution, that is why a portion of W’s skill shot damage scales with champion level. This will let us make W feel impactful even at rank 1, while also not letting it be too powerful if you max it first. The Ezreal now has a choice, Q for more poke and waveclear, or W for more burst. Since Q already effectively has per champ level scaling with the AD ratio and applying on hit effects, if the Ezreal player maxs W then Q will still feel relevant. And if they max Q then W can still be made to feel relevant. The AP and AD ratios on W also aid in this. In both cases both spells feel good but which one you max still feels like a real choice.

Wait, didn’t I just say we want people maxing Q first though? Well, it’s a bit more complicated. It’s ok for people to max W, but Q should also be the default option. Both AP and AD use Q to push, and if they skip ranks in Q then they will not reach their breakpoints needed to kill minions. This means most ranked players will want to max Q, but someone who does not care about winning and just wants to blow people up has the fun option of maxing W first. There is no reason to stop them from enjoying that, even if we don’t support it from a win rate standpoint.

But Potato! I want to max W first AND have good wave clear! Me too man, me too. But that’s just simply not how Ezreal works, and it would take some massive changes to make that happen. If we did not have to worry about those pesky AD players it would be a lot easier O.O

Remember, there is only one main group of players who max W first, burst AP players, and the only thing keeping that build in check is the lack of waveclear. While we can’t solve burst AP’s waveclear problem, we can reduce some of the stress points, and in the process also reduce some of the stress points of poke AP.

To to loop all the way back, we have two different AP builds, both need more waveclear. Poke AP needs more kill threat, yet we can’t just slap more kill threat on because burst AP would then have too much. We can use the Mark mechanic combined with the Q buff to solve this situation.

It’s all a numbers game now. We pull poke power out of W max through reducing the base skillshot AP ratio, compensate through locking up damage in the mark, the goal being for the total damage to be similar to what it is now. That poke power gets moved into Q ratio, so Burst AP has similar kill threat, more wave clear while not having quite as much abusive poke.

Our design is now pulling double duty, because we gave W champion level scaling, and since we moved some of the power out of the ratio and into the base damage this helps Poke AP have W hurt more since Poke AP builds lots of CDR and mana, not lots of flat AP like Burst, and Poke maxs Q first. Poke AP also gets to reach the minion break points that are so so important for making it a more competitive build from the Q buffs.

Congratulations! We gave both kinds of AP exactly what they need, in total it’s a healthy buff to everyone which is absolutely fine because they both need a buff. And we’ve done it in a way that is also a very healthy change to AD Ezreal as well.

The only concern I have is increasing the AP ratio of Q could make Lich Bane Q’s too powerful late game, the main reason I propose an AP ratio rather than increasing the base damage on Q is because of its implications on AD, an AP ratio is a great way to affect AP without affecting AD. What is important is reaching those breakpoints, be it through base damage or ratio.

It’s worth pointing out that I could just be wrong on the need to not increase Burst AP’s total burst, in which case keeping the same AP ratio .8 and letting the mark be a straight buff to both Burst and Poke is an option. I understand I’m very bias in wanting more power in AP Ezreal and am trying to be very careful about this in my design.

Part Three : Places to pull power from

Remember back in part one when I said “So we will need to talk about how and where to siphon power out of other places in his kit, but that is for later.” We are finally to that later. Here are my proposed places to pull power from:

-Remove the attack speed buff from W -Make Q AD ratio 1.0 -Remove the AD ratio on E -The big one, remove Manamune from the game

Let’s start easy, the attacks speed buff was a given and the Q ratio change is a flavor thing. It’s a freaking auto attack on a stick man! It should have one auto attacks worth of AD ratio, it just feels wrong to have 110% of an auto attack. I understand why it was changed when it was, but this is a chance to fix it. It’s not a ultra big deal if this does not work out but it would be nice. Lastly removing the AD ratio on E seems fair since it was always just kind of a bandaid. This helps reinforce the divide between how AP and AD use E. AP goes in, AD goes out. The ratio was initially there to reward aggressive play, but we have the Mark now to do that. Those changes give us a little bit of power to work with, but we can do more.

Before we dive into this, we could also just stop here, removing manamune is bold, and might be out of scope for this rework, but I said I’m going all out so lets go all out.

Manamune. Let’s first restrict our design domain, luckily Ezreal is the only real user of manamune, so we don’t need to worry about how this affects other champions. And the fact that we have an item in the game that is only used on one champion is a sign that perhaps that item is not actually needed in the game.

Let me ask you a question, do you really like stacking a tear? Is that really what Ezreal is all about, or perhaps is it just a play pattern that he has been forced into? Do you really like having to get a mana item on your first back while your lane opponent gets a BF Sword? Does that really reinforce Ezreal’s spell slinging fantasy? I say no. The current design of manamune is a crutch that makes Ezreal work right now, and it creates a lot of very wonky power fluctuations that just don’t fit on Ezreal. He is weak in lane because he has to get a tear, and to compensate he gets an absolutely absurd mid game power spike when he fills his tear, only to fall off again in the late game(or not, depending on how he builds). It’s not terrible, but we could be a lot better. Instead we can even out his power curve throughout the game to a much more fluid design. Right now his power is full of massive peaks and valleys, and it’s all because of manamune. Don’t get me wrong, peaks and valleys are not inherently bad, but Ezreal is not a champion who needs this level of delta between his low points and his high points. Ezreal is a generalist in many ways. He is meant to be safe and consistent, he is not the late game power house of Vayne or the early game power of Lucian. Besides manamune has its own design problems, it gives you lots of mana, yet you do more damage the higher mana you are? You are incentivized to not use all the mana you paid for, and since there is so much delta in power between full mana and no mana it makes Ezreal’s damage output unpredictable and unclear for both the Ez player and their enemies. And where does muramana give power to Ezreal? His Q and E. And where do we need to pull power out of? Q and E. It’s also important to his auto attacks though, which is a problem we’ll get to later.

We just ripped a massive portion of Ezreal’s power from his kit. We compensate now through giving him a W that actually does things. This has tamed his wild power fluctuations, his lane is now going to be much more satisfying since he has a new toy to play with and can actually deal damage rather than just wait around while his tear stacks up. To pay for this we gave up our bonkers power spike in the mid game. This leaves an auto attack that is weaker than we want, and a wild wild west of itemization options.

Part 4 : Hybrid

What if I told you that AP and AD Ezreal do not have to be two different things? What happens when Ezreal does not have to rush a tear? What does he build instead? I give it a high chance he gets a sheen first back, but then what? As a player it makes my mouth water at all the possibilities to theorycraft, and what is important is even in bot lane his options are not limited to AD items. Gunblade rush? Nashor's Tooth? Duskblade? Ludens? There will always be the standard Triforce, but since we are not locked down by the tear our options for what to build before and after Triforce are opened up massively. If we remove manamune Ezreal becoming a hybrid champion is not just a cool idea, it’s simply unavoidable. Even right now it’s very common to see hybrid builds on him. This is not a bad thing, we have been trying to open up bot lane to different play styles for awhile, and this helps address a historical problem of ADCs feeling too similar to each other. We just need to be sure the traditional style is viable too.

When faced with an item domain this large it is simply impossible to design around particular items. Instead we must look at what each stat does on his kit. AP is spell damage and burst, AD is spell damage and auto attack power. Let’s finally talk about the ratios on the new W. Another recap:

It’s not ultra complicated, AP gives you skillshot damage, AD gives you mark damage. The fact that it’s simple and understandable is an intended part of the design, but we are designers, we want to understand why it is this way. AP Ezreal already has a well established relationship with W, you fire it and deal damage. None of this mark jazz, so it’s just the logical choice to put the AP ratio there. For AP, the mark is a way to shave off some poke power, it’s not designed to remove that poke power. The need to shave off poke power is lessened as the game goes on for AP, and this is reflected in the placement of the AP ratio. By late game it is ok for AP players to not care about the mark much at all since that fits the playstyle they are used to.

AD on the other hand has to work harder for their burst damage, they have to actually get that mark. This makes sense, if you build AD you should not get to burst people as easily and safely. Having an AD ratio also helps with the “rank 1” problem. Early game the skillshot section does fine damage for AD, but if they don’t put points in they start to feel it once the mid game rolls around, as they should. As the game goes on the divide between AP and AD becomes larger, AP cares less and less about the mark while AD cares more and more. Also, let's think about what effects can actually detonate the mark, auto attacks, Q, E and Ult. Most of the time it will be detonated by auto attacks or a Q, and guess what? Auto attacks and Q are both the AD themed portion of Ezreal’s kit, thus putting the AD ratio on the mark helps even further to reinforce the AD playpaterns.

Also notice how the skillshot does magic damage and the mark does physical? Not only does this just fit, it also helps with penetrations.

We are finally on the last stretch, but have one more major problem left to solve. The removal of the muramana I expect will leave Ezreal’s auto attacks weaker in the mid and late game than desired. In order for the AD play style to work his auto attacks need to be powerful enough to justify the risk he’s taking by getting close. It’s hard to say if that will actually be a problem without playtesting, but if it is the solution is simple, just buff his passive, most likely make it more powerful at higher champ levels.

Part 5 : Conclusion

Let’s tally up all the changes made.

Passive: -Increased power at higher champion levels

Q: -AD ratio reduced to 1.0 -Slightly increased AP ratio and or base damage

W: “Ezreal fires a wave of energy in a line, dealing magic damage to enemy champions struck, and mark them for X seconds. Basic attacks and spells detonate the mark, dealing bonus physical damage.“ Skillshot damage : (Flat based on skill rank) + (Flat based on champion level) + AP Ratio Mark damage: (Flat based on skill rank) + (AD Ratio)

E: -Removed AD ratio

R: -No change

Other: -Remove Manamune

When you actually tally everything up, these are not a lot of changes. When it is all said and done all these changes are easy to communicate to players and most importantly Ezreal still feels like Ezreal. And that is perhaps the single most important reason I stand by the Mark mechanic. It’s extremely low risk and extremely high reward. I’m glad Riot agrees with this given their road map post:

“We didn’t need a map to know that dramatically changing his abilities would be a disservice to one of League’s most iconic marksmen. “

And ya know what they also said in that road map post?

“Ezreal players aren’t in love with his current W, Essence Flux, so we will be updating that ability to be more in-line with his theme of an ace spellslinger—lining up shots then knocking ’em down.”

Lining up shots then knocking ‘em down you say? Perhaps you hit someone with a skill shot then follow up...somehow…... don’t know…....perhaps by making the skill shot apply a mark? HMMMMMMMMMMMMMMMMMMMMMMMMMM HOW INTERESTING AN IDEA.

If I am right, then I FUCKING CALLED IT BITCHES!

-PotatoMine

A MASSIVE design exploration of Ezreal and how to rework him

0 Comments