Xypherous on Garen's Passive and more on Warmog's; Meddler comments on Sejuani Rework

Continue reading for a collection of red posts concerning Garen's passive, a long winded discussion on Warmog's armor and other itemization points, and a very brief update on the Sejuani rework.

About Garen's Passive
A concerned summoner posted on the forums asking if Riot could do something with Garen's passive. His suggestion was changing Garen's passive to something like "Garen's health regeneration is increased by 0.5% for each 1% of health he is missing." Xypherous responded to this by sharing:

"We actually tested this exact passive on Garen's rework. 

While the passive, on paper, looks amazing and feels like it solves everything - what it actually does is heavily discourage mid-level harass - because if you fail an all-in, he recovers much more health than he would have and, unless you were resourceless as well, you have wasted a ton of resources to amplify his health regeneration. 

That is, unless you manage to kill Garen, you are burning your limited resources to diminish his resources temporarily - but amplify the rate at which he accumulates resources. So essentially, unless you killed Garen straight out (which, well, it's Garen.) - You are screwing yourself really really badly.

Therefore, the end result was that you have to save everything for an all-in - leaving you with even less incentive to interact with Garen."

 When suggested the numbers of the previous idea could just be tweaked a bit, he elaborated:

"As I pointed out - It's not a problem with numbers. It's a problem with the pattern. You waste your resources and cooldown to amplify Garen's resource accumulation.

Therefore, you have zero incentive to interact with Garen at all. That is the core flaw - You can move numbers around until his passive isirrelevant - which is fair, I suppose - or you could force Garen to build anti-synergistic stats to even stay in lane - which is also fair if you want to nerf him - but the overall gameplay pattern was worse overall."

Dropping this debate about an altered passive, Xypherous shared his general thoughts on Garen's current passive, saying:

"Garen's passive is entirely in control of his opponents - and in many ways, that's kind of fine - because Garen has such control of his opponents.

That is, while I agree that the passive is in the hands of an opponent who knows what he's doing - Garen's opponent mostly has to obey the whims and calls of the Garen player, as he's a resourceless champ that has strong zone control (against melee, at the very least.) 

That's essentially the dynamic here - While your analysis is correct - the larger pattern as a whole seems to handle pretty well - Garen is a raging bull that you have to manage carefully to overcome - while Garen, frequently, has to realize that sometimes, charging in *isn't* the best answer and to exert pressure without committing to too much, something that he can do because he's resourceless."

Finally, when asked if this "opponent's being in control" of Garen's passive was a bad design, he responded

"It depends. Skillshots are both in your control and the enemy's control. This makes skillshots more interesting, rather than less so. Your aiming matters as much as the enemy's dodge - and it makes skillshots great despite how basic they are. 

Holistically, you can view this as the champions kit should be in your control and the enemy's control - That what creates the interaction between you and the opponent. While I'm not saying that Garen is the absolute pinnacle of design in this front - what's important to realize is that it should be a direct competition between your strength and how the enemy can counter your strength rather than having two characters that don't care about each others abilities but simply clash to see whose skillset has more numbers.".

Xypherous Talk Warmogs and Itemization

Similar to the above, a summoner posted suggesting Warmog's could be fixed by changing it to
something like ( remember this is a fan suggestion, not a confirmed change or something suggested by a Rioter ):

"Warmog's ArmorStats +640 health
Passive: Gain 20 x Current Level in Maximum Health

Unique Passive: You gain health regeneration equal to 1.5% of your maximum health every 5 seconds."

This attracted Xypherous's attention, to which he replied:

"This particular idea reduces down to: 

Punish players for finishing an item early - by making it so that the earlier you finish the item, the weaker this is and the more gold-inefficient it is.

To be perfectly fair, it is a path that works on certain statistics, particularly late-game statistics like % penetration - in which the disincentive is to gently guide people into realizing when the item is truly good. 

However, this is a case where you are taking an item that is already not cost-efficient for raw health - punishing players that finish it early by making it even less cost-efficient thereby merely making anyone who finishes Warmog's armor before the health-break even point into an ineffective build. The problem is compounded by that fact that it is Health that you are gating so the most desirable early game statistic is not available by finishing the item - thus what is most likely going to happen is that players will merely stack Giant's Belts to get around your attempted fix. 

While this will nerf the item, most definitely - it does so merely by making it a foolish option to upgrade the item until the numbers pan out. Did you buy Warmogs before level 15 to 18? Congratulations, you made a mistake. You should have bought another Giant's Belt because you have tricked players into buying an item with negative gold efficiency until you deem it to be "correct." 

Players should be rewarded for finishing items that they want to buy - That is intrinsically the point of items - This change effectively "solves" rushed Warmogs by giving rushed Warmogs negative reward - punishing players who are doing well rather than solve the gameplay that a rushed Warmogs presents. Don't get me wrong, this tactic is certainly valid in some games - if you want to be really heavy handed about it - but at a certain point, you should just say "Warmog's Armor - Level Requirement: 15" and be done with it. It's a much cleaner way of accomplishing what you want innately. It's no real surprise that most RPGs do use the idea of level restrictions for this very purpose.

However, level restricting an item doesn't really fit into what we want for itemization."

 When asked why they removed the minion kills give bonus HP mechanic from Warmog's in the first place, Xypherous responded:

"It's mostly due to the fact that this particular mechanic only opens up Warmog's as a good defensive option if you are already winning. 

Let's say Warmog's was the best option to win games - by putting a farm and stack mechanic on it - you effectively negate the ability of the opponent to stabilize the game by buying the same item. Intuitively, if you have the same build as your opponent - you should be able to fight them to a stalemate."

Another summoner commented that the original "gain health per level" mechanic was similar to Rod of Ages. Xypherous replied:

"The key difference is: Rod of Ages rewards you for rushing it, because you get the stats earlier the earlier you rush it. 

This version punishes you for rushing it, because rushing it only gives you negative stat valuation. 

Recall, the two items still have to be balanced by the end-game somewhat - RoA self-balances in the end-game, because it always has a time requirement - thus we can make RoA's base statistics crazy and still have it balanceable for end-game - because there's always an additional cost."

As the discussion continued, a different poster brought up Bloodthirster and questioned the logic behind it's minion kills = stats component if in light of the old Warmog's stacking mechanic being removed. Xypherous explained:

"Bloodthirster is interesting because it only indirectly helps you out in terms of preserving the stacks. While the person winning is far more likely to have full stacks - they are quite frequently killable with only a Bloodthirster. 

Seeker's and Bloodthirster also mitigate their snowball advantage by having a relatively small max stack cap - 30 minions is roughly 5 waves, which can be accumulated in 1 or 2 minutes, rather than 100 of the old. So, while the patterns are flawed in the same respect - the small stack caps make it a small flaw rather than a large one.

Finally, Seeker's has the release-valve of, the upgrade to the item essentially gives you maximum stacks - thus giving an alternate to needing to farm it out."

As a counter point for the original community idea of making Warmog's stronger per level instead of up front, someone pointed out that Locket of the Iron Solari and Shen/Trinity Force/IBG ( which all scale with base AD ) are items that currently scale directly with level. Xypherous responds:

"As I said in my original post - "To be perfectly fair, it is a path that works on certain statistics, particularly late-game statistics like % penetration - in which the disincentive is there gently guide people into realizing when the item is truly good." That is - it would be a trap to make these items appear to be attractive to rush - because they are bad when you rush them - not in terms of overall effectiveness - but because it is just a bad item if you rush it. 

Contrast, Warmog's - Early Health is what you want to rush. That's where it has the *most* relevance not the least. An item that's +3% Attack Speed per level wouldn't really be that bad - because Attack Speed is terrible to rush anyway and doesn't need to be gated. It self-gates. Here, it's preventing people from making a mistake until the item is good - whereas the Warmog's example is simply punishing players from wanting to make a good decision. 

Secondly - it's also a question of scale - The patterns of Locket's shield is a problem (it makes it pretty awful on supports, but it's a support item - and supports tend to be lowered level. :c ), to be fair - but it's a small enough such that it might be acceptable. However, the question here is literally, you go from no stats on upgrade to another 1200 gold. That *is* a problem when the solution is magnified to that extent.

As for the Sheen/Tri-force/IBG line - ask yourself - is a late game Sheen worth it? No. The power of Sheen is incredibly front-loaded. Whereas the power of this Warmog's suggestion is incredibly back-loaded. While perfectly true in that these scale with level - rushing Sheen is good - and rushing this item simply loses to getting another Giant's Belt. However, a late-game Sheen does very little for your overall damage - which makes it the inverse pattern - Sheen tries to not punish you if you get it too late - but it is essentially irrelevant late game."

When questioned about the goal of a such a heavy HP item, he responded:

"The ideal goal that we're tweaking towards is to make items more fluid. Frozen Heart or Randuin's Omen, for example, is an excellent counter to the version of BoRK that is going up now. Which are in turn destroyed by Black Cleaver and Last Whisper. Which are in turn are destroyed by Warmog's Armor - which in turn is destroyed by Torment/BoRK. 

The problem is that currently, the cycle is "stuck" on Warmog's Armor - so clearly the wheel needs further grouping so that it flows more continuously. You should feel fine sticking to your preferred item build of choice defensively - but your 2nd or 3rd item will prove much more effective if you react to your opponent's build and if, in turn, they react to yours.

Obviously, it's not quite there yet - but we're going to keep grinding until it is."

One forum poster  responded "Why not make the regen out of combat only?", Xypherous commented on this by saying:

"I tried this before - What I discovered was that it makes matchups incredibly swingy. For a default item to cause wildly varying swings in matchups was rather odd. 

That is, champions who could farm from afar and stay safe benefits from the item the most - while characters that need to be active in lane are punished. You worsen some of the worse cases (Cho'Gath, for example) while weakening some of the ideal cases - A heavy trade champion like Lee-Sin. 

Unfortunate - Kind of hoped it would work."

Xypherous also shared his thoughts on wishing he could produce a late game regen item that was appealing to only tanks instead of bruisers.

"The thing about the late-game regen item is that ideally, I would like it to exclusively be a tank late-game regen item, as I think tanks without sustain are the class of characters that a late-game sustain item should support. 

For example, Leona, Amumu, Nautilus etc. - These are the characters that need a heavy late-game regeneration option - not Xin Zhao, Irelia, Jax.

It may be that it might be impossible to create one without it being supplanted by bruisers - unsure - but I know that, if we created a high regen item, we'd want it to be more tailored towards true tanks like Leona / Mummy rather than Fighters."

When asked if cookie cutter builds where really where he wanted the game to be, Xypherous responded:

"I suppose the answer to this is very champion specific. Some champions are going to have cookie-cutter builds. They have certain requirements that they *need* to have in order to function - and for the most part, that's kind of fine. 

However, some champions should be able to build with far more flexibility - like supports or mages - because their jobs tend to be far more flexible. The carry should also be able to build with far more flexibility to his needs - because his needs tend to shift the most depending on the enemy compositions.

However - for fighters, their job/role doesn't change much. They are heavy divers that try to get to the backline. They don't have as much conditionals - and I don't really expect their items to diverge too much - as opposed to every other class who actually have shifting priorities. This may be a problem with the Fighter class as a whole, perhaps - but that's something we'll also have to solve on a more champion specific basis as well."

He then moved to talk a little bit about why some bruisers are so dominant after the S3 item changes.

"So, I'm not going to lie - the vast majority of champions that are crazy now are bruisers with free resistances or resistance shreds of their own. I kind of knew this going in - that shifting baseline itemization was going to cause a lot of the things we did in the past to bite us back. 

Like for example, Kayle, Xin Zhao, Jarvan. Notice a trend? Shreds. A lot of free damage. A lot of free resistances. Why did they need it? Because they couldn't actually deal damage in S2 or live in S2 without a bunch of free statistics - but now that the classes are more balanced around each other - those free things that made them function are the same things that are making them crazy. 

In the end - I think we'll get there slowly, by either taking away offensive power and keeping the resistances, or taking away the resistances and keeping their power - but in many ways taking away a lot of the unnecessary tools they have now that they desperately needed in S2."

In response to these last two comments, one poster asked if Xypherous's was afraid of making the game too bland. He responded:

"The statistics of a champion needs to synergize with the playstyle of the champion. In some cases, the character really needs the shred because that's who they are as a character. In some cases, a character doesn't need the damage because that's not who they are as a character. 

The analogy would be something like this: You have a unique character - then you have to glue a random ball of stats onto him to make him work. The random ball of stats obscures the uniqueness of that character somewhat. 

Nautilus, in many ways, is defined by his lack of mobility - Yi, in many ways, is defined by his incredible free mobility from stats. Both of these are completely core to how the champion feels and the stats help to make the characters more unique. 

However, Kayle randomly shredding your armor/mr doesn't really make Kayle more Kayle. Her old AP/AD passive made Kayle feel even more like Kayle.

Free stats are a powerful force for uniqueness - when used correctly. However, in S2 - these characters needed so many other free stats that every fighter started looking like 'Okay, here's their free resistance steroid, here's their free damage amp / item efficiency / armor penetration steroid - here's their burst mitigation stat - and now we can actually make the character.'"

He continued on, discussing what item's should bring to the table:

" I just want the champion's kit to have exactly what it needs to be unique and stand out on his own. Every single tool that we have (damage / cc / statistic) is something we use to make a champion feel more unique and his own. 

Items should let players customize the champion to whatever they want - but what's important is that we don't feel forced to put certain things on certain champions because they need it because that's the only thing making them work. You know? Like, the statistics shouldn't be holding them up and viable - the statistics are yet another tool we have to make the champion feel awesome.

You don't know the number of times we had to tack on a random AD scaling thing just to make a champion relevant late game - in a way that even our players don't really like. "Why does Darius have 25% Armor Penetration for free?" - Because he really needed it, strangely."

Someone also asked for Xypherous's opinion on AD carries being able to build Warmogs. His response:

"I don't consider Warmog's on AD carries a problem - I feel like there's a real and pretty awesome healthy debate on which is the best defensive item for AD carries. RO? Locket? FroMallet? Warmogs? GA? 

While this might settle back into 'you buy X on AD carries, it is the best defensive option' - that's much more of a concern. 

I kind of want to push that even more - if an AD carry feels like he can really choose between 5 or 6 defensive items for his one defensive item and really choose between 5 or 6 really good items for his attack speed item - then mission accomplished. 

However, we generally frown upon tank mages that deal a ton of damage - mostly because what they really are at that point is AP Fighters that weren't ever really designed to be AP Fighters - thus when that happens, we try to retune them close back to the original view.

(It should make sense for Graves to want a Bruiser item as his defensive item. He's a Ranged Carry that is 30% Fighter. It makes sense for Caitlyn to want a Frozen Mallet as her defensive item - she's a ranged carry that wants to kite or train. Etc.)"

Brief Update on Sejuani Rework
Lastly, Meddler dropped off a small update on what's up with the previously discussed Sejuani rework.

"Hi all, just a short update this time. We're currently investigating some options on the visual side of things and looking at how those tie in with gameplay. Stuff's still going on and we haven't forgotten this thread, nothing new and concrete to discuss yet though, will update when there is or again after a while."