Reign of Gaming

Reinboom on Critical Strike

Originally Posted by Riot (View Original Source)

There are a number of interactions that Critical strike is creating that need their value explored first.

Personally, I dislike Critical chance in league. At its worst, which is randomly realized, a single critical strike can define a game - especially on feast or famine characters - in either direction. Individual attacks have high meaning in League and having that action not be controlled when most of the rest of the game is feels a bit wrong.
And to be further clear it's not the RNG element that's wrong here; it's the RNG element compounded by the reaction window afterwards. RNG that guides strategy and decision making but doesn't define individual events are amazing (Bard meeps <3). Short response window RNG is the icky one.

... However...
Critical strike also does a lot of good things.
To highlight these, I would like to call out and contrast some of the effects of Critical Strike to the most common suggestion in changing it: Flattening the output. (Flattening the output is where Critical Chance becomes a basic attack damage multiplier. 1% Crit -> 1% Basic attack damage amp).

In the broadest of theory, assuming all actions are the same, just changing one for the other would (given that there's lots of games being played) would statistically be the same.
However, that wouldn't actually be true. Not all actions would be the same. Here are some of the key differences (in no particular order):

1. Critical chance proposes a degree of possible threat that people do respond to. An ahead Marksman will be more reluctant to engage on a Marksman that's behind and has a bit of crit because of the chance of a crit still sits in the realms of possibility. This allows a behind Marksman to still project some control whereas if all damage was flattened this would be less true.
   
   

2. There is an interaction with critical strike that I will call "crit fishing". Crit fishing is where someone will poke the enemy and then hard engage if they see a critical strike. If they don't, they back off (with the enemy sustaining back up a bit). Even though what's being tried is luck (ish, thanks to pseudo-rng it will actually be guaranteed eventually), there's a skill and a mini-game wrapped around it on capitalizing on the event. This entire mini game doesn't occur with a flattened damage situation.
   
   

3. Thanks to pseudo-RNG, when there is a decent amount of crit (but not max) on a character in a full team fight where Marksman get to hang around for awhile then the fight will have a mix of crits and noncrits. In mid to late game, given the high damage outputs that Marksman have, this can be pretty defining. The marksman will crit someone into being into burst range, the question is "who and can it be acted on?" This is an element of unpredictability and interaction with other classes of characters (burst mages) that is very similar to point #1, but stretched even further and with a proper response window for teams to react and decide if and how they hold their engagement or disengage. This flowing and shifting long fight dynamic would be lost with the flattened version.
   
   

4. One of the hardest ones to capture: the feeling of crit. When a crit occurs, for a lot of people it's fun. The number is larger, there's an awesome fiery icon next to the damage, the damage number moves differently, your champion does different things (animation and sometimes sound), and the enemy health bar takes a heavier chunk. Each of these aspects feed off of each other. There are an ensemble of elements executing this feeling and the shift itself between adagio to allegro is what matters. Having just one or the other can be quite boring for many.

I'm sure there are more things crit is doing as well. These are just some major ones that have been partially understood.

Now, what the value of each of these aspects are is much more difficult to define and emulating them all becomes rather complicated.
Unlike, say, the dodge mechanism or the phage random slow mechanism of many days past, critical is also much more ubiquitous which makes trying to understand these values much more important.
All of that said, it might turn out that the right answer is just flattening it - or any number of other ideas. The bad of Crit might just be straight up worse than all of these elements combined. However, I can't say with confidence that that's true.
Tricky area. Would love to read more thoughts here on this.

---

The input and the output are different. The output of each is the "good" aspect. The fight that it's RNG based isn't. It's just so happening that RNG is creating these outputs.

This distinction matters because a lot of the effects are harder to capture. I'll go into examples for each:

Why is it a good thing that you have the possibility to win fights that you should't win based on luck? Why is it good that a player that earned an advantage through skill can't press that advantage cause he's afraid of rng?
Also, why are adc and the trinity bruisers allowed to turn the tides of a fight by 'devine intervention' while APs and Tanks need to purly rely on skill for combacks, isn't that a bit double standarts here?

The main thing a range of projected threat (that includes greater than the average) does is deter snowballing and enables people towards wanting to play even when behind. Each player gets a very different sense of this point, and a lot of that choice tends to occur when people aren't aware of it. It's important to note that different people respond different here as well.
Overall however, snowball effects tend to be a good thing. Until it reaches a certain point (extreme rubberbanding), then it becomes a bad thing.

The next issue is what is the value of this particular antisnowball effect? We're always trying to figure that out since it's a tricky point to reach. The more antisnowball effects, the more comebacks and the more long struggles. In the most extreme worlds, too little comeback potential (everything is equal) means that games are decided the moment one line goes slightly behind on CS. Too much means that games never end and you can never seal your advantage.
The question is, is how much is this mechanism specifically contributing (to either direction)? Also, how does it reflect perception of antisnowball on personal levels? Removing it might increase the number of "This game is over AFK" first blood cases, for example.

Crit fishing is one of the mainreasons i hate current crit. It's not about proper use of skills and wavecontroll, it's about getting a random crit in the beginning of a trade and capitalizeing on it.
I honestly fail to see any positiv aspects on that.if i see a caitline useing her passive on a minion and then missing her Q and therfor dash into her face as lucian, and i loose the trade cause she crits and i don't while we both have IE, i fail to see how that 'minigame' has anything to do with skill....

The example here you addressed something that is not critfishing. I may have not described it very well then. Sorry about that!
I agree completely that in a committed skirmish that a rare and unexpected crit (e.g. the crit firing from a 1% Crit rune or a 10% from Zeal) is outside the realms of skill. That's exactly a primary problem of Critical chance (notably from low crit percentages!). That is also not Critfishing.

Critfishing occurs when the champions are such that an autoattack initiates the skirmish. Champions that initiate via a moveblock or train play pattern tend to not experience this as much (or see it when it occurs on them), because the fight didn't start based on the Crit itself, it played out in spite of any Crits.
Instead, Critfishing is where a champion will poke the enemy champion with intention to pull off immediately. Only upon seeing a crit does the aggressor engage. It's a chaotic fight initiator.
It's also important to note that during the course of that fight, they will have fewer possible Crits as well due to the nature of pseudoRNG. If properly responded to (avoiding their skills for example), a fish can turn negative.

Again however, this is a single examination of game mastery. The value of this occurrence is unknown.

Some how i fail to see the 'proper response window' to crits in teamfights. If a support or mage dies to a lucian E+AA+Q+AA with 3 crits or a sivir AA+Q+W with doublecrit within less than a sec, how is there a response window?
If your adc gets forced out of the fight by a single headshot-crit from cait, how is there a response window?
If you irelia dives in cause she knows she can tank through the ap-carrys burst and a couble off AAs while stunning an important target, but explodes to a doublecrit ontop of usuall burst befor reaching the target, how is there any window for reponse?
If i have the better initiation and are in a better position for the fight, but are forced to retreate and give up objektives cause the enemie carry did crit some trargets into burstrange while ours was out off luck, how does that add anything valuable to an competitve game?

The response window occurs at the team scale not the individual scale with that point.

Debonair Galio doesn’t always go mid lane...

Originally Posted by Riot (View Original Source)

 

His calculations are so precise, Vel’Koz asks him for help with geometry.
His singing in the shower is so breathtaking, it stole Sona’s voice.
He once stepped on a mushroom and Teemo exploded.

He is...

The Most Interesting Gargoyle-like Construct in the World.

Queue up with class as Debonair Galio, available now in the League of Legends store for 750 RP.

More on State of Nasus

Originally Posted by Riot (View Original Source)

Do you consider it fair and balanced that Nasus can Siphon Strike structures, but Rengar's and Sejuani's ability to do so was considered a bug and 'fixed'? [...]

I do feel it's fair that Siphoning Strike affects tower, whilst Rengar Q and Sejuani W don't. Building destruction is one of Nasus' core strengths, in part so that he's always at least a moderate threat even if your team can stay away from him forever. It's something he needs to function. Rengar by contrast's there to kill a dude a bunch (and then either get out or kill more dudes), bringing mobility and stronger (over a short duration) lockdown. Sejuani's got a lot of CC, most of it AOE and she's there to lock down and control team fights. Neither of them needs to be strong at destroying structures to be useful members of a team.

How do you feel about Nasus's gameplay in lane? On one hand, there's a tremendous amount of strategic play revolving around denying him farm early on, whether it be through precise zoning and lane control or heavy ganking, but there are also occasions where his laning pattern feels pretty low on direct interaction, since he's incentivized to actively ignore his opponent and power through with his healing on minions. How would you go about encouraging Nasus to fight his opponents early on?

Agreed, his laning can be a bit passive at times. I don't think it's inherently a problem if the onus to be aggressive is consistently on one champion (not Nasus in this case). Where I think we do see issues is where aggressive action's just not rewarding, in large part because of the Lifesteal from his passive. Could see an argument for nerfing that (and potentially shifting the power elsewhere if appropriate) if we conclude at some point he needs a clearer early weakness. It does currently allow him to function in a range of lanes he'd otherwise really struggle in however, so not personally seeing a need to change it right now.

URF Doombots Game Mode Idea

Originally Posted by Riot (View Original Source)

URF Doombots was something we considered, but ultimately didn't have time for :( I do agree it would have been pretty hectic!

In order to have a good Doombot experience, the AI would have to be tweaked. Just enabling the URF ruleset makes the bots cast their spells somewhat more frequently, but not to the URF degree that players do.

Another consideration that makes it tough is both URF and Doombots are more game server costly, so making sure the servers don't melt would take some work too :)

Gonna need some Super Saiyan Servers then.
The Servers just needs to scream for 20 episodes & they're good to go.

Indeed :D

It needs to be a trifecta of having:

• additional game server capacity
• good monitoring (being able to detect/predict degraded experiences before players experience it)
• reactionary tools (i.e. being able to throttle game starts for specific queues)

We've made big strides over the past year in some of these areas, but obviously we could always be better.

Champion and Skin Sale: 04.17 - 04.20

Originally Posted by Riot (View Original Source)

 

Grab these champions and skins on sale for 50% off for a limited time:

---

Skin Sales

Give your champions a new look with these skins:

Ghost Bride Morgana 975 487 RP
Ironscale Shyvana 750 375 RP
Samurai Yi 520 260 RP

---

Champion Sales

Add these champions to your roster:

Darius 975 487 RP	Wukong 880 440 RP	Singed 585 292 RP

Other Red Posts

• NA Server Roadmap: South Bridge update