Red Post Collection: The Harrowing Passes Soon, 4.19 deploying later this week, Improving the Cursor, and more!

With the Shurima event, we spent a lot of time refining the lore surrounding the key players(see our promo page). In this expansion, we reveal that Cassiopeia, driven by unrelenting ambition, betrayed Sivir and opened an ancient, cursed vault. The guardian of the vault, a giant serpentine statue, came alive and sunk it's fangs into Cassiopeia, 'rewarding' her ambition with the serpent's curse. With this passive, I tried to capture the following sentiment (from the Shurima lore), on top of the gamplay goals: "The serpent’s curse is the terrible price of Cassiopeia’s ambition, but it also promises a power she never imagined." Her current passive seeks to capture her ascension into this 'promised power' through use of her new serpentine body-- that is, as she embraces the curse, she unlocks more of its power. 

Hope that explains why the passive might seem a little disconnected from the traditional poison thematic. There's more to Cassiopeia than just "she poisons people," and I should have been more explicit with how I hoped to tie this passive in with her characters identity/thematic. Sorry about that! Maybe the passive should be called The Serpent’s Curse instead. Anyway, i hope it seem a bit more fitting now :p"

"Hey Grand Eleven! I deeply appreciate your tone here, you express disappointment, but in a way which shows you really care about the character and are open to discussion-- thanks for that!

To your initial points, 1) yes this passive gives us powerful levers on when Cassiopeia hits power spikes, but its less about that and more about creating discrete moments where the Cassiopeia players feels like they've evolved. A gradual ramp pattern accomplishes the former, but not the latter so much (of course, 2-shotting a tower with Nasus does some of this, but that's pretty different).

As for 2), yeah the range and readability profile for Twin Fang is a much friendlier profile for DoTs than than the long-ranged, spammable AoE Q. That said, there's a lot to love about that Q pattern, so we wanted to preserve that while bringing it's best case to a more appropriate position.

The poor ability to switch targets a a really good point. The poison => twin fang reset loop is core on Cassiopeia, which is a gate no ADC has to jump through.

That said, I don't think your proposed changes address this. Target swapping is not out of volition but often necessity (the adc hits the tank because she has to, or else she will die). Likewise, Cassiopeia will have to hit the target shes presented with, and thus be unable to elongate poison duration on other targets (I suspect). Further, increasing poison duration undermines the core pattern of Q, EEEE, Q, EEEE, so on, replacing it with Q EEEEEEEEEEE so on. Fundamentally, I think having to weave in skillshots between Twin Fang sets is both fun to Cass players and more fair to their opponents.

You've identified an important point however, so let's talk more about that. You say Cassiopeia has poor target selection, and I think that is totally valid. I wonder if it's possible to give her enough surrounding tools where shes comfortable having the disadvantage of even poorer target selection than an ADC. I suspect it is. For example, it's true that its more difficult for Cass to swap targets, but when she does, she gets a huge ms boost, which is not true of the additional ADC. If this is not strong/ interesting enough, maybe it can be bolstered, or maybe it can be rethought. Definitely an interesting debate wotth having. What do you think?"

Jump abilities and when they deal damage

"(Sorry, this got a little long. TL;DR: I agree, but it's a lot more complicated than you think.)

I have been campaigning internally for this for about a year.

The story went like this: I was assigned a simple enough bug to fix: Jax's Leap Strike would damage enemies even if he got knocked out of his jump. So I started doing some research, and the state of things here is pretty bad.

The short story is, we're all over the place. Some jumps, like Jax's Leap Strike and Irelia's Blade Strike will do damage no matter what (including if the champion is killed mid-jump!), some will not do damage if the champion is knocked out of the move, and a select few (particularly Jayce's To the Skies! is coded very well) do what you'd expect: they only damage the target if the target is still near the attacking champion once they land. There's two different categories that generally are okay:

1) AOE only moves, like Fizz's Playful/Trickster, are generally okay, as they only do damage in an AOE around where they land (To the Skies is a hybrid ability, that both damages the main target and other targets in an AOE; we're only concerned about the main target for the purposes of this exploration). You correctly identified Urchin Strike as one of the problematic abilities here that doesn't check for distance on move end. To Xypherous's credit, it does check whether or not Fizz was knocked out of the jump, to my recollection.

2) "MoveToUnit" type abilities that will get to their target no matter what, following through flashes and the like (for instance Maokai's Twisted Advance or Nocturne's Paranoia). Almost universally we make these abilities uninterruptible, so no matter what happens en route, you WILL get to your target.

My original proposal was very simple: 

Standardize unit targeted "move blocks" (internal jargon for abilities that move you) with an effect done to the targeted unit such that the effect only happens if a) the move block concludes successfully (you're not knocked out of it) and b) the target unit is still within 300 units when you land. Notice we're now talking about an "effect" rather than damage. This is very much on purpose--sometimes these move blocks can stun even after you flashed away (looking at you Pantheon).

There were a LOT of very strong arguments on both sides of the debate. The pro side cites clarity and counterplay mostly, both fundamental design principles of LoL. The against side had a lot of concerns about screwing the player taking an action here, which I'll summarize as: 

1) Certain kits will fall apart if we do this. Pantheon once again comes to mind. Imagine trying to gank a LeBlanc as Pantheon with this change: she would be able to Distort away from you every time. Yes, it does feel bullshit to an enemy to burn Flash and still be stunned by Aegis, but imagine being Pantheon and seeing that your only CC has just become useless against a subclass of champions. A number of proposals were made to offset this, most hopeful amongst these "restore half of E's cooldown if it fails to stun its target", but that brings us into even murkier waters. All of a sudden Pantheon gets to double jump. He may want to trigger W for the movement alone against elusive targets to get closer to the actual target he wants. The TL;DR here is we'll need to rework Pantheon's kit at some point, but the change called for in this thread will definitely neuter him in the mean time, if we go forward with it.

2) Reliability of targeted abilities shouldn't be thrown out like that. This is a frustratingly convincing argument. The auto attack example brought up by GhostHerald is at the core of this. There's a class of ability in our game where counterplay isn't in the action the enemy takes after the ability is used but entirely in the positioning play between the two sides. Auto attacks are an example of this class of abilities. Lucian has a max attack range of 500 units. If he attacks you from this range and you flash away out of range, should the auto attack fizzle? (This is an actual question. I don't know.) If the answer is yes, what about tower attacks? Melee basic attacks? Targeted missile skills such as Taric's Dazzle and Ryze's Overload? These abilities all pay a high prize for being targeted in their ranges, cooldowns, and damages. What they buy themselves for that price is reliability. We would take that away.

(To go on a little extra excursion, what would our game look like if you could use mobility skills to evade auto attacks given you'd leave their max range before they hit you? I'm not sure I want to live in that world. The mechanical demands our game places on its players aren't as high as, say, Starcraft, and this is very much by design. Every mental cycle you make a player spend on mechanical skills is a mental cycle they cannot spend competing with the enemy on tactical and strategical levels. Our game would ultimately be poorer in tactical and strategical play if we allowed split second mechanical play especially on low cool down, high move speed dashes to decide encounters. We're balancing on a knife's edge here, and with champions like Azir we've recently erred a little too much on the side of mechanical skill I think.)

I feel the overall consensus has moved toward making this change eventually. It's on Team Clarity's backlog (a list of things the team wants to do, eventually, prioritized in order of greatest value / lowest cost) The problem is that to execute WELL on this, we'd necessarily need to add feedback on when this interaction was triggered. Maybe floating text? Different particles/sound for hitting/not hitting? We originally identified about 16 skills in the games that would require this treatment. I have so far fixed only the most egregious ones, and only for the "your jump was actually interrupted" case (I've applied this fix to both Leap Strike and Bladesurge; particularly noticeable against Azir's ultimate, where previously Irelia could kill you with Bladesurge even if the ability was instantly interrupted by the phalanx). Someone would need to take on this project and do due diligence with regards to clarity, balance, and not completely ruining certain kits (both Pantheon and Diana would pretty much break; imagine wasting an R cooldown because your Q target dashed away)."