Boards›Gameplay

A Deeper Dive into Critical Hits: Input/Discussion/Alternatives Appreciated.

ModThe Djinn··36 votes·3,639 views

Hey there, Gameplay!

My previous post suggesting a tweak to the mechanics behind critical hits got, frankly, more attention than I was expecting, so I figured it people might be interested in a more detailed breakdown of the analysis I did. Presented here for discussion are several of the options I've explored up to this point, along with a positive/negative analysis of each and some of the gameplay ramifications I personally think might accompany each potential system.

Discussions and thoughts of your own are appreciated (these are presented without my opinions on which are good/bad specifically as a springboard for discussion around the topic and community suggestions), although I'd appreciate it if we stayed within the statement of intent, which is discussing the mechanics of crit themselves, not the general balance or need to entirely remove it in favor of a different sort of multiplier (which, while very valid discussions, are best addressed in another topic).

TL:DR This is a huge topic. I'm afraid I can't summarize it. :P

#STATEMENT OF INTENT Critical hit mechanics are a tricky thing in League. Riot has stated several times (here and here, for a few examples) that critical hits are a bit of a recurring subject with positives and negatives on both sides, and that it's something the team looks at not infrequently. As such, love them or hate them, it's something worth discussing and analyzing.

An important thing about this post -- and I know this will frustrate some of you -- is that this analysis is not intended to be a deep dive into carry balance and/or large itemization shifts to support carry balance. This isn't a thread devoted to analyzing or attempting to solve anything other than a highly specific issue: the possibility of improving critical hit mechanics in terms of play patterns and player perception. If there's interest I'd love to consider a big discussion thread on carry design/balance/playstyles in the future and how you could redesign crit to reposition its role as a damage multiplier that’s efficient against all targets, but that is not the intent behind this analysis.

The end goal is to identify methods of adjusting critical hits that keep a strong feedback loop while also reducing the perceived impact of randomization. Note that this is not necessarily equivalent to eliminating randomization entirely, although we'll explore some options for that as well.

>Section TL:DR -- Don't expect major balance suggestions, because this analysis is based on the play patterns and satisfaction/clarity output of the core mechanic of critical hits, rather than an analysis of carry balance in general. The goal is primarily to improve the critical hit mechanics, not shift the balance of marksmen.

CURRENT CRITICAL HITS

First off, an important note: critical hits are psuedo-random, in that an attack that is not a critical hit increases your critical hit chance on the next attack, resetting when you land a critical hit. This helps prevent a drought of critical hits, while keeping your expected output at roughly the expected value (2 crits per 10 hits for 20% crit, for example).

Secondly, a point of uncertainty that I'd love to know if anyone knows it: does this psuedo-randomization tick different for champions vs. non-champions? I.e. is champion crit chance on a separate fail-safe tracker than minion crit chance?

Positive features of Critical Hits

  • Provides a strong positive feedback loop: purchasing crit leads to big visible damage numbers, cool animations, and chunky bits of enemy healthbars vanishing as you attack. This feels great to a player in a way a constant 25% damage increase doesn't.
  • Provides some anti-snowball potential, as even champions who are ahead need to be wary of critical strikes. Note: This may not be something the average player is considering. Does it offer enough value to highly skilled players to make it worth retaining? Uncertain. Maybe this can be emphasized?
  • Creates "crit-fishing" gameplay, where a harassing champion can fire off a shot and follow it up with an immediate engage if the initial hit is a critical hit, or where a champion can attempt to predict the psuedo-random RNG to "cook" a crit for use against an enemy.
  • Creates varied and dynamic play through psuedo-randomized damage output variance and the need to react to the same either offensively or defensively.
  • Randomization at lower values creates interesting teamfight opportunities -- if you crit one target and are unlikely to have another crit up soon, do you switch to a priority target or focus down the victim of the initial hit?
  • A portion of the player base really likes it.
  • Creates high points and low points in gameplay (usually a positive thing for player experience).

Negative features of Critical Hits

  • Randomized damage output, resulting in unpredictable gameplay.
  • Largely out of player control.
  • Can potentially swing the results of the laning phase or skirmishes.
  • Even low values can have a powerful effect if they land properly.
  • Variance leads to situations where crit is very strong (you get lucky crits on every attack against an enemy you're harassing) or very weak (you get unlucky and land no crits on attacks against an enemy you're harassing).
  • Can feel very bad to get hit with during a "lucky streak."
  • Can make you feel like you lost due entirely to luck, rather than due to skill, even in situations where the other play may have also outplayed you due to skill. As such, can be disempowering to both players.
  • A portion of the player base has a very strong negative bias against it.
  • Creates lows in gameplay (a negative thing if it occurs too often).

#SOME TAKEAWAYS FOR ANALYSIS Some takeaway questions from the above breakdown, that we can use to judge possible crit models:

  • Does it reduce the perception or impact of randomization, or create better gameplay around it?
  • Does it create a strong feedback loop?
  • Does it create highs and lows in gameplay? If so, does it create too many lows?
  • Does it retain interaction points or introduce new interaction points?

Some other things to consider:

  • What are the ripple effects? Is it small-scale, does it impact other roles, does it require new itemization?
  • How does it impact the laning phase?
  • How does it impact objective control or teamfighting?
  • How might it affect the metagame?
  • How does it impact snowballing?

#ANALYSIS I'll be making several posts in this thread that detail some of the solutions I've explored, and permalinks to the individual comments will be added below. If I see posts with suggestions and breakdowns that I consider to be well-analyzed, I will likely include them in the below list.

These suggestions are explorations into possible models, and not all of them are things I consider particularly compelling directions. Most of them have benefits and downsides, which I have tried to detail.

  1. Precision (Flat increase)
  2. Focus (Crit-stacking)
  3. Overcrit (Layered Crit)
  4. Hybrid Precision (Flat increase with critical hits) 5. Scaling Crits (Static % chance, scaling damage)
  5. Hybrid Scaling Crits (Static % change, scaling damage, flat increase)

87 Comments

ModThe Djinn10 votes

#SUGGESTION 1: PRECISION We'll start with the most commonly suggested model, which I call Precision.

This model is super simple: all critical hit chance is instead converted to a percentage increase on all auto-attack damage on a 1%-to-1% ratio. For example: 50% critical strike chance (a 50% chance to get 100% bonus damage, for an average increase of 50%) changes to a flat 50% increase in damage across all attacks.

Positive Impacts

  • DPS becomes entirely predictable.
  • Low crit values no longer result in lucky spikes.
  • No mechanics outside player control.
  • No bad feelings that you lost a fight due only to random luck.
  • Increases the consistency of harass and trading, as you know every hit will deal the same damage, no matter when or where you land it.

Negative Impacts

  • Removes the entire feedback loop.
  • Feels no different than purchasing AD, save that you also do not get spell damage increases.
  • Removes all high moments and low moments.
  • Removes all anti-snowball potential.
  • Removes all interactive gameplay elements.
  • Removes all engage elements.

#FURTHER ANALYSIS

Does it reduce the perception or impact of randomization? Yes. This model removes randomization entirely, although it may flatten the gameplay surrounding crit too much.

Does it create a strong feedback loop? No. Removal of all surrounding effects and gameplay feel results in the same feedback loop as a flat damage increase.

Does it create highs and lows in gameplay? No. It removes all highs and lows.

Does it retain interaction points or introduce new interaction points? No. It removes all interaction points in favor of standardized damage increases across the board.

Does it have major ripple effects? Quite possibly. Infinity Edge likely becomes a desirable item for assassins, divers, and bruisers due to the large AD along with what is now a powerful boost to all auto-attack damage , and Phantom Dancer/Shiv may not give bruisers stronger dueling or wave-clearing power. Phantom Dancer and IE are the biggest concerns.

How does it impact the laning phase? The laning phase becomes entirely predictable in terms of damage output, making it relatively safer and more likely to retain a lead you keep.

How does it impact objective control or teamfighting? Dependent on crit values at that time. Fewer squishies will be bursted by lucky crits at low crit values, which definitely favors assassins and mages.

How does it affect the metagame? Uncertain. IE/Phantom Dancer purchasing giving much more damage to divers/bruisers, and spike damage flattening could push more ranged characters into lane against melee where their consistent harass can drive them out more. It's hard to judge the exact impact the change would have.

How does it impact snowballing? Removes the ability for a lucky come-back strike, but also removes the frequency with which a champion who is ahead gets a big nova strike. Pushes the lane slightly in favor of the person currently ahead, as the DPS increase is reliable and consistent, and you're not ever in a situation where you miss two critical hits and your opponent lands two.

InTheory5 votes

Removes all anti-snowball potential.

I don't understand that point.

Also I don't really understand all the negative points about missing interaction and similars. I actually see that as something positive because it would increase the focus on champion's abilities again instead of just basic attacks. All the cool crit plays might feel pretty cool if the player is used to it because he mains the Crit-class aka marksmen, but for the mortal players it's too difficult in terms of micro decisions when you can't really predict the enemies damage output as every other class has.

Removes all anti-snowball potential.

The player who's ahead is most likely going to have more crit chance if both are opting into crit as their build (for equal comparison). That increases the chance to get lucky in trades for the person who is ahead even more, making crit more of a snowball effect than the opposite. Also while you mentioned that there's that sort of "bad luck protection", I don't see if the opposite is the case as well. Like if you get lucky for the first 3 basic attacks, are you punished for the rest of the trade or will you only be protected in the case of not hitting your expected number of crits in the fight giving crit user an rng based advantage over the non-crit-user.

ModThe Djinn4 votes

{quoted}Also I don't really understand all the negative points about missing interaction and similars. I actually see that as something positive because it would increase the focus on champion's abilities again instead of just basic attacks. All the cool crit plays might feel pretty cool if the player is used to it because he mains the Crit-class aka marksmen, but for the mortal players it's too difficult in terms of micro decisions when you can't really predict the enemies damage output as every other class has.

I think this one comes down a lot to individual perspective. Given how much the Marksman role relies on auto-attacks, I think there's a lot to be gained from ensuring that there is gameplay in that mechanic other than "click target." It's totally something people may disagree on though.

The player who's ahead is most likely going to have more crit chance if both are opting into crit as their build (for equal comparison). That increases the chance to get lucky in trades for the person who is ahead even more, making crit more of a snowball effect than the opposite...

Somewhat true. Being ahead does favor the person who is ahead in all circumstances, but crit does give occasional windows of opportunity in a few situations, especially with engage-heavy and burst-heavy supports/carries. It's a fairly minor thing that doesn't happen super often, but it's still there.

Sightless662 votes

Given how much the Marksman role relies on auto-attacks, I think there's a lot to be gained from ensuring that there is gameplay in that mechanic other than "click target."

That gameplay should be kit-based, instead of something very homogenous to all champions. Additionally, this is assuming that there is gameplay with random crits at all, and frankly, I think there's very little. You're almost never changing your choices based on crits, baring a rare reaction to an unexpected string or if you're playing a crit-fisher. It's almost always the right decision to play as if crit is going to give you the expected value.

Villainous3 votes

{quoted}

The player who's ahead is most likely going to have more crit chance if both are opting into crit as their build (for equal comparison). That increases the chance to get lucky in trades for the person who is ahead even more, making crit more of a snowball effect than the opposite. Also while you mentioned that there's that sort of "bad luck protection", I don't see if the opposite is the case as well. Like if you get lucky for the first 3 basic attacks, are you punished for the rest of the trade or will you only be protected in the case of not hitting your expected number of crits in the fight giving crit user an rng based advantage over the non-crit-user.

I’d like to contrast thing against a situation in which both champions build flat AD instead. If one person has a longsword and the other has a pickaxe, the person with the better items is going to 100% win every otherwise equal trade. Brawler’s glove v cloak, on the other hand, might even oit in the long run to the cloak guy winning most trades, but there’ll also be plenty, especially if it’s only like 2-3 autos per trade, where there are no crits on either side, or one crit for both, or sometimes even the glove player coming out on top. Crit items aren’t snowbally. Crit’s most powerful/unfair interactions are where two players are on even footing, both building crit, neither in a position to mitigate such things, where one rolls well and the other doesn’t.

AirKingNeo2 votes

Removes all high moments and low moments.

Wouldn't low moments being removed be a positive, thus making this neutral?

Removes all anti-snowball potential.

And also removes snowball potential. You already listed lucky crits being removed which are positive. You kinda made this weakness.

Removes all interactive gameplay elements.

What? It's an auto attack. It has no 'interactive game elements'.

Removes all engage elements.

There wasn't anything engaging to begin with. It's an auto attack.

You also forget to mention it opens up Crit to more classes, like PD on Aatrox.

ModThe Djinn1 votes

{quoted}

Wouldn't low moments being removed be a positive, thus making this neutral?

Weirdly, not really. Having marked highs and lows in gameplay typically ends up putting more emphasis on the feel-good highs, in practice. As long as the lows are TOO low they actually tend to serve to highlight the high points.

And also removes snowball potential. You already listed lucky crits being removed which are positive. You kinda made this weakness.

While true that it removes lucky snowball trends, not having them means that the person who is ahead stays ahead entirely, while the other way has a chance to get back in. It's a relatively minor impact overall, but I still wanted to include it.

What? It's an auto attack. It has no "interactive game elements".

The patterns that build around something (crit-fishing, for example) count as interactive elements. This does remove some gameplay.

You also forget to mention it opens up Crit to more classes, like PD on Aatrox.

GREAT call. I did definitely forget Phantom Dancer, which is a big deal.

Daxsx2 votes

not having them means that the person who is ahead stays ahead entirely

This is already how it works. You don't see a competent player who is behind actively looking to take unfavorable trades and fights on the off chance that they get lucky and crit more often than expected and that the enemy crits less often. That just seems like more of a hypothetical benefit than a real, practical one. What happens far more often is that either the jungler comes to gank or the support of the ADC that's behind lands CC on the other ADC presenting a situation where the bot lane that's behind will have the advantage and a chance to stop their snowball or even swing it the other way.

There's also the aggravation of unpredictable minion crits which can lose you CS and screw with minion prepping while farming under tower. While in a flat damage modifier there's no unpredictability and farming with or without crit feels the exact same. I'd say this is a situation where are no silver linings and it just feels awful all around since it can happen to a player of any skill level where they get punished not because of a mistake they made, but because of an arbitrary design choice within a game mechanic that is not optional if you want to stay competitive.

Lastly I don't see why there's such a push to label lucky or excessive critting as a "high", as a player you're not always intently checking to see if every one of your autos crit or not (especially on ranged carries where you'll be focusing on kiting efficiently most of the time) so you may not actively think about or even notice how often you crit. And even if you do notice that you just crit way more than you should and it turned a fight in your favor I'd say most people would, at most, only feel a "high" in a momentary: "sucks to be that guy", schadenfreude-esque way rather than getting any meaningful satisfaction out of it.

ModThe Djinn8 votes

#SUGGESTION 2: FOCUS (i.e. Crit-Stacking Mechanic) Another variation on a common suggestion, Focus is a critical strike replacement based in part on Caitlyn's passive Headshot.

This model replaces critical strike chance with focus rate. Focus charges up to 100 at a rate of (focus rate) focus once per attack, to a maximum of 199. When a champion hits 100 Focus, their next auto-attack reduces his/her focus to focus value + (old focus value - 100) and triggers a critical hit, which deals 200% damage.

Example Tristana has a current attack damage of 100, and 20 focus rate. She gains 20 focus per attack, so after 5 attacks she has reached 100 focus, and will score a critical hit for 200 damage on her next attack. Once she has used that attack, his focus resets to 20.

Example 2 Tristana has a current attack damage of 150, and 30 focus rate. She gains 30 focus per attack, so after 4 attacks she reaches 120 focus and will score a critical hit for 300 damage on her next attack. Once she has used that attack, his focus resets to 50 (30 base + 20 over).

This may look slightly complicated, but the base starting value is to ensure that the critical strike itself is counted towards your total attacks (so that 20 focus gets you a crit once every 5 attacks instead of once every 6), and the overflow buffer is so that you are not punished for values that don't evenly fit into 100.

Positive Impacts

  • DPS becomes entirely predictable.
  • Low crit values no longer result in lucky spikes.
  • No mechanics outside player control.
  • No bad feelings that you lost a fight due only to random luck.
  • Increases the consistency of harass and trading, as you know every hit will deal the same damage, no matter when or where you land it.

Negative Impacts

  • Removes the entire feedback loop.
  • Feels no different than purchasing AD, save that you also do not get spell damage increases.
  • Removes all high moments and low moments.
  • Removes all anti-snowball potential.
  • Removes all interactive gameplay elements.
  • Removes all engage elements.

#FURTHER ANALYSIS

Does it reduce the perception or impact of randomization? Yes. Focus reduces randomization by putting critical hits on a predictable timer.

Does it create a strong feedback loop? Yes. Players get a new resource to manage, and still retain the big chunky hits, big numbers, and animations of current crit.

Does it create highs and lows in gameplay? Yes. You have predictable high points and predictable low points.

Does it retain interaction points or introduce new interaction points? Yes. There is now a new, visible mechanic that lets players know that they (or their opponents) have a critical hit ready to go.

Does it have major ripple effects? Likely. IE (or any item like it) becomes a super powerful engage tool, specifically on assassins. This can lead to a big boost in the single-target damage coming from divers, AD assassins, and Marksman nova strikes, and likely forces IE as a first rush on any slightly burst damage-focused AD character trying to engage on an enemy.

How does it impact the laning phase? Harass becomes much stronger, as you can routinely predict how to land the highest damage on your opponents. Carries with longer ranges or abilities that synergize well with this become extremely potent. Specific concerns exist for Caitlyn, who already has a similar mechanic and now can double down on it.

How does it impact objective control or teamfighting? Squishy vulnerability increases, as carries can now precision-target them with critical damage.

How does it affect the metagame? Potential for huge impact: increased harass damage can force a lot of carries out of the meta, and could also lead to carries finding significant success in other lanes where their currently less consistent early harass can lead to them struggling. Marksman rushing IE appearing mid or top for the reliable 250% damage nuke every 5th attack seems very possible, and potentially infuriating. Assassins or divers doing the same with low crit values and building them on monsters/minions prior to ganks or engages is also a major risk factor, especially with burst options like Dark Harvest, Duskblade, and Trinity Force existing.

How does it impact snowballing? Likely enables it further, as characters who are ahead are given distinct windows of when they'll be at maximum power, and can use these windows to threaten off other players from farm, or simply to reliably whittle them down before murdering them outright.

ModThe Djinn8 votes

#SUGGESTION 4: Hybrid Precision A hybrid of precision with traditional crit mechanics, this model retains some randomness while backing in predictable outputs.

The model reduces critical hit's damage from 100% bonus damage to 50% bonus damage, but takes 50% of your critical strike chance and adds it to your damage as a flat modifier. This flat modifier does not critically strike.

Example Tristana has a current attack damage of 100, and a 20% critical strike chance. She has a 20% chance to strike for 50 bonus damage, and deals additional damage on auto-attacks equal to 10% of her base attack damage.

The expected output for this is equivalent to current critical strike: Over 5 attacks Tristana would deal 10 bonus damage per attack (for 50 bonus damage), and average one critical hit for 50 bonus damage. This equals the normal 100 bonus damage she's expect from the current crit system.

Numbers in this system can be heavily tuned to favor crits or non-crits.

Positive Impacts

  • DPS becomes more predictable.
  • Retains all the existing gameplay and positive features around crits.
  • Makes the impact of each individual crit swing fights 50% less.
  • Gives some consistent damage to back up crits and reduce reliance on them.

Negative Impacts

  • While it retains all existing gameplay and positive features around crits, it reduces the actual mechanical impact by 50%, which could feel less exciting or make the gameplay matter less. Testing would be required.
  • Still retains randomization that some players strongly dislike.
  • Tones down the existing crit gameplay, which can be bad depending on the player's enjoyment of it.
  • Harder to explain than the current mechanic.

#FURTHER ANALYSIS

Does it reduce the perception or impact of randomization? Yes. It reduces the impact of critical randomness by 50% in terms of actual DPS output.

Does it create a strong feedback loop? Mostly. It retains crit's current feedback loop, but exchanges chunky hits for more consistent damage.

Does it create highs and lows in gameplay? Yes. They are less impactful than current crit, but they are still present.

Does it retain interaction points or introduce new interaction points? Yes. All existing interaction points are preserved, just at reduced individual impact

Does it have major ripple effects? Unlikely. For the most part the changes are largely gameplay and output neutral, and don't meaningfully change any play patterns.

How does it impact the laning phase? Harass becomes more consistent but less bursty, which may slightly favor characters who can land it reliably, and crit-fishing is diminished in value. Exact effect are hard to judge on this one, as the change is fairly minor overall.

How does it impact objective control or teamfighting? Removes some of the "do I pursue the target I hit or switch targets" effect, but impact on the average player should be minimal.

How does it affect the metagame? Likely no real change, save that crit is less frustrating and damage output is more predictable. Lack of incentive to alter builds and lack of real changes in poke/harass means this one should be fairly free of major impacts.

How does it impact snowballing? Slightly pushes lanes in favor of the dominant role due to the reduced damage for lucky come-back strikes.

ModThe Djinn6 votes

#SUGGESTION 3: OVERCRIT A suggest that's something of a hybrid between current critical hit mechanics and the flat percentage increase, Overcrit transitions 100% crit into a flat increase, and stacks a new crit chance on top of it, flattening damage while retaining some level of randomization.

This model reduces critical hit's damage from 100% bonus damage to 50% bonus damage while also doubling the critical hit percentage granted by all items. Zeal, for example, becomes a 40% chance of 50% bonus damage rather than a 20% chance of 100% bonus damage.

The other core change is that critical change can stack up to 200%, and any percentage over 100% is a chance to deal an additional 50% bonus damage. At 120% critical hit chance, for example, you have a static 50% damage buff from your critical hit chance, and a 20% chance for an additional 50% damage.

Example Tristana has a current attack damage of 100, and a Zeal, which now gives 40% critical strike chance. She has a 40% chance to strike for 50 bonus damage.

Example 2 Tristana has a current attack damage of 100, and 140% critical strike chance. She has a 100% chance to strike for 50 bonus damage, and a 50% chance to strike for an additional 50 bonus damage.

Positive Impacts

  • DPS becomes more predictable.
  • Retains all the existing gameplay and positive features around crits (save at 100% critical strike chance, when it is identical to the Precision model).
  • Makes the impact of each individual crit swing fights 50% less.
  • More frequent crits make purchasing crit even more appreciable.

Negative Impacts

  • Removes the feedback loop at 100% crit chance.
  • Has a weird moment where it "resets" the feedback loop, which can be good or bad pending the individual player's perception of how good it feels to buy power.
  • While it retains all existing gameplay and positive features around crits, it reduces the actual mechanical impact by 50%, which could feel less exciting or make the gameplay matter less. Testing would be required.
  • May be difficult for the enemy to "relearn" their opponent's crit chance.
  • Still retains randomization that some players strongly dislike.
  • Tones down the existing crit gameplay, which can be bad depending on the player's enjoyment of it.
  • Harder to explain than the current mechanic.

#FURTHER ANALYSIS

Does it reduce the perception or impact of randomization? Yes. It reduces the impact of critical randomness by 50% in terms of actual DPS output, while also raising their frequency by 100% for any given crit value.

Does it create a strong feedback loop? Mostly. It retains crit's current feedback loop, but exchanges chunky hits for more frequent crits. Exact effect of this on the loop is uncertain at this time.

Does it create highs and lows in gameplay? Yes. They are more frequent and less impactful than current crit, but they are still present.

Does it retain interaction points or introduce new interaction points? Yes. All existing interaction points are preserved, just at increased frequency and reduced individual impact.

Does it have major ripple effects? Unlikely. For the most part the changes are largely gameplay and output neutral, and don't meaningfully change any play patterns.

How does it impact the laning phase? Harass becomes more consistent but less bursty, which may slightly favor characters who can land it reliably, and crit-fishing is diminished in value. Exact effect are hard to judge on this one, as the change is fairly minor overall.

How does it impact objective control or teamfighting? Squishies are slightly safer against nova-strikes, and crits being more frequent removes some of the "do I pursue the target I hit or switch targets" effect, but impact on the average player should be minimal.

How does it affect the metagame? Likely no real change, save that crit is less frustrating and more predictable. Lack of incentive to alter builds and lack of real changes in poke/harass means this one should be fairly free of major impacts.

How does it impact snowballing? Slightly pushes lanes in favor of the dominant role due to the reduced chance for lucky come-back strikes.

ModThe Djinn6 votes

#SUGGESTION 6: Hybrid Scaling Crits The model sets critical hits at a 50% chance, and converts critical strike chance to critical strike damage, scaling up to 100% bonus damage. Additionally, half of your Critical Strike damage is applied as bonus damage to your auto-attacks. Critical strikes begin at 0% damage, and thus would not trigger.

Example Tristana has a current attack damage of 100, and a 20% critical strike damage from Zeal. She has a 50% chance to deal 20 bonus damage, and deals a flat 10% bonus damage with her autoattacks. Over 10 attacks this equals 200 bonus damage on average (100 from crits and 100 from auto-attacks) -- the same output that the equivalent 20% critical strike chance would get from two crits for 100 bonus damage each.

Positive Impacts

  • All characters that build any crit get a good chance to see critical strike animations.
  • Early game crits hit for significantly less than they do currently.
  • Crit occurrences are far more predictable.
  • Crit isn't hugely impactful in the laning phase as a randomization element, since it's so common.
  • Damage on crits is low enough that it does not impact the laning phase significantly.

Negative Impacts

  • Critical strike animations might become too frequent.
  • Might be frequent enough that gameplay doesn't feel meaningfully different from a flat damage increase.
  • Removes much of the existing gameplay around critical hits.

#FURTHER ANALYSIS

Does it reduce the perception or impact of randomization? Almost certainly. Value of a single crit drops significantly, although randomization is actually still greater post 50% crit.

Does it create a strong feedback loop? Likely weakens it somewhat, as crits are frequent and far less impactful, especially early on.

Does it create highs and lows in gameplay? Not especially. Highs and lows are flattened somewhat, and the frequency likely detracts from the impact.

Does it retain interaction points or introduce new interaction points? Not really. Lowered value of crits means that interaction points only really arise when you start to hit higher values.

Does it have major ripple effects? Unlikely. Crits, while frequent, aren't especially bursty, which discourages most off-roles from purchasing itemization.

How does it impact the laning phase? Crits are significantly weakened at all levels, likely to the point of almost feeling like just an AD boost. This slightly favors the dominant laner.

How does it impact objective control or teamfighting? Not a significant change, save that incoming damage becomes very easy to predict.

How does it affect the metagame? Unlikely to have major changes to the metagame, save that frequency of incoming damage could allow Marksman to solo lane better.

How does it impact snowballing? Doesn't affect lane snowballing hugely (slight favor to the dominant laner).

InTheory2 votes

No more comments to make here, just wanted to say that I read everything thus far. Already pretty late here. I think all in all there's still that point of conflict of having the great feel of scoring a crit versus having to deal with the hard to predict damage output of their oponents for everyone else. But it's interesting to see these ideas being mixed up. Thanks for your work!

Teridax686 votes

I very much agree that crit is a subject worth discussing. As a feature that's been present in the game for so long, it's one many people have gotten used to, but which has also nonetheless caused a significant amount of irritation due to its random nature. Seeing how Riot's been playing a little more with randomness, with mixed results, I think it's also a good time to discuss exactly what kind of contribution we expect from randomness, and what we really don't want from it at all.

With that said, while the OP claims to not want to do a deep dive into crits, it tries do so anyway, and goes into a series of bulletpoints I personally consider rather questionable, namely:

[{quoted}](name=The Djinn,realm=NA,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=,timestamp=2018-02-18T21:23:26.541+0000)

  • Provides some anti-snowball potential, as even champions who are ahead need to be wary of critical strikes.

How is this different from any other kind of nuke? Do critical strikes possess some unique property that allows them to bypass gold or level leads?

[{quoted}](name=The Djinn,realm=NA,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=,timestamp=2018-02-18T21:23:26.541+0000)

  • Creates "crit-fishing" gameplay, where a harassing champion can fire off a shot and follow it up with an immediate engage if the initial hit is a critical hit, or where a champion can attempt to predict the psuedo-random RNG to "cook" a crit for use against an enemy.

Does this actually happen in practice? This sounds very clever on paper, but while I may be wrong on this, I have yet to see this translate into any kind of emergent gameplay even among pro players.

[{quoted}](name=The Djinn,realm=NA,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=,timestamp=2018-02-18T21:23:26.541+0000)

  • Creates varied and dynamic play through psuedo-randomized damage output variance and the need to react to the same either offensively or defensively.

Please describe how pseudo-randomized damage output makes play "varied" or "dynamic": what new decisions does crit generate? Does this happen in practice?

[{quoted}](name=The Djinn,realm=NA,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=,timestamp=2018-02-18T21:23:26.541+0000)

  • Randomization at lower values creates interesting teamfight opportunities -- if you crit one target and are unlikely to have another crit up soon, do you switch to a priority target or focus down the victim of the initial hit?

Does this also happen in practice? From personal experience, perhaps Tryndamere may rely on a potential crit to cheese his lane early, but I have also not really seen crits meaningfully affect teamfighting decisions. Is there any sort of video evidence, guide, or other reliable source that demonstrates the gameplay that is claimed to be associated with crit here?

[{quoted}](name=The Djinn,realm=NA,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=,timestamp=2018-02-18T21:23:26.541+0000)

  • Creates high points and low points in gameplay (usually a positive thing for player experience).

This kind of apes the actual design principles around highs and lows, wherein those highs and lows are a reflection of the player's skill expression -- you don't want to make the player feel terrible for no good reason; rather, you only want to make them hit a low if they make a mistake and get punished for it. Conversely, you generally want to make players feel good, but especially if they achieve a particularly difficult play, and as with low points, giving players high moments for a reason that does not correlate to their skill can come across as confusing in a game that heavily emphasizes skilful play.

On a more nit-picky level, the difference in phrasing between this:

[{quoted}](name=The Djinn,realm=NA,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=,timestamp=2018-02-18T21:23:26.541+0000)

  • A portion of the player base really likes it.

And this:

[{quoted}](name=The Djinn,realm=NA,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=,timestamp=2018-02-18T21:23:26.541+0000)

  • A portion of the player base has a very strong negative bias against it.

Comes across as snide. Drawing lines in the sand, then declaring one side to "really like" something, and the other to be really biased against it, itself carries bias, and undermines many of the people who have participated in this debate.

Criticism aside, my personal opinion on crit is that it is fundamentally power without gameplay: perhaps any sort of random output happening on the map could lead to some changes in decision-making (and even then, I doubt that, particularly if those random outputs happen more than once a second), but that is not enough in itself for it to justify the random element, in my opinion. If the objective is to create unpredictable opportunities, by introducing new choices that could benefit either side equally, then perhaps that could happen (though even so, the random element would still need to justify its existence over an algorithm based purely on player decisions), but the problem with crit is that it expresses itself primarily through random output, i.e. by awarding success or failure to either side in a manner that does not intrinsically stem from skilful play. In a game that has taken great pains to aim for competitive integrity by basing the outcome of matches as much as possible on the actions of players, I therefore do not think any kind of system that preserves output randomness deserves to stay. As such, I do not think suggestions 3 through 6 would solve the problems with crit, as they preserve output randomness.

This post is getting a bit long already, so I'll make a separate comment for a crit model I had considered, which proposes crit as a three-strike/add-on model.

Lord Desert2 votes

Well to "fix"/"improve" critchance we first need to understand why theres randomness in a competitive game in the first place. It doesnt make sense from a competitive standpoint! The reason why crit is based on a chance and not just a flat dmg amplifier is because it brings unpredictablelity, uncertainty, thus more "interesting" gameplay, especially for adcs and the viewers.

To put it blantely: The adc gameplay experience is flat out boring. You cant "outplay" the rightclick. Compared to skillshot dependent champions/roles, which can completely miss and do nothing at all, a "zero or hero" playstyle mostly based on your and your enemys skill level, thats interesting because anything can happen! This type of exciting gameplay experience (hit or miss), (in general) doesnt exist for adcs. They rightclick and thats it.

And thats the reason why this randomness (to be clear its controlled pseudo randomness) is present in a competative game. As a quick example: You wouldnt watch sport games if the outcome is mostly like the statistics predict. Nobody would watch it because its boring; Its predicatble, its not interesting.

So if you want to fix or improve this system, ether remove the randomness and also somehow manage to introduce something similar to skillshots for the adcs rightclick, OR just try to improve the randomness a bit, so its not as bad as sounds. Otherwise riot will not even try to considere the idea, because it introduces more troubles than it fixes. My personal stance on this is: "Pure chaos is bad and unhealty, but controlled chaos on the contrary can even improve gameplay experience!"

I did alot of research and dove pretty deep into Leagues crit mechanics and to improve the current system I propose a stricter PRD. One which still is based on pseudo randomness, but outliers and extreme cases are elimiated. So its technically "fair" randomness, but it should no longer be able to royally f*** everything up, just because you or the enemy good extremely lucky. Again, it will not messy with normal cases, ~99% of all cases, just the bad extremes!

I can provide a indepth explaination of my stricter PRD, if wanted. And if I claimed something that is wrong, please feel free to point it out! Thats what discussions are for :D

Teridax683 votes

I largely agree with this, and would personally be very interested in a pseudorandom number generator that would eliminate outliers already (if that isn't already the case). However, while I do agree that a certain level of unpredictability is necessary to keep sports fresh and interesting, I'm not sure randomness is needed even in that kind of situation, since League has a sufficiently complex framework that, given enough balancing, it can have such a vast number of strategies and interactions that it would not be easy to predict a game's outcome.

I very much agree with your stance on controlled chaos, and how it can benefit gameplay, which is why, even with my personal distaste for RNG, I do think there is room for some randomness, as long as it presents equal opportunities for both teams. This is why I don't really think the randomness to drakes is inherently bad (how those random drakes interact with swings in team momentum I think might be a problem, though), though it's also why I think stuff like crit or Zoe's W are terrible for good gameplay.

I also am very much on board with you when you say that the basic playstyle of marksmen would be fairly boring without the inconsistency of crit, since many of them would just be very constant damage pumps. However, I think this is already the case, and crit here only acts as a smokescreen to distract from it. Crit I think innately has no real gameplay, so the end result is that the damage output of marksman also lacks interesting gameplay (though marksmen have a lot of potential for interactivity nonetheless). Because of this, I also agree that any replacement to crit should not simply convert the stat into a persistent damage modifier, but should legitimately add gameplay and emphasize components to the playstyle of marksmen that are more interactive (e.g. attacking for a prolonged amount of time, rather than just once or twice).

Lord Desert2 votes

I'm sorry, I would love to further discuss this topic you with right now, but its getting pretty late over here in EU. And I do have a lot to discuss, it is a wall of text so be ready :D I will hopefully get it done by tomorrow, to tired to finish and to roughly spell check this monster right now. As a bonus: How much do you know/want to know about PRD/ my stricter version?

a) I can give you a link to an old post of mine where I explained it decently, might be a bit outdated now but should still be valid for the most part :D b) a somewhat "short" tl;dr: with the basic principals c) the full version

Teridax682 votes

I look very much forward to it! If you'd like, I'd love to see the full version, along with the link to your old post.

Lord Desert2 votes

Change of planes: I finished the wall of text a while ago, but it was just to big and it wasnt that good to read, so I scrapped most of it and I'm trying to get it as resonably small as I can :/ In the mean time heres the link to my old post, so you dont get tooooo bored :P https://boards.na.leagueoflegends.com/en/c/skin-champion-concepts/Oe9PQnbE-summonerslab-part-2-crit-algorithem-improvement

I'm trying to get done, but it could take still quiet a while, I'm a rather slow writer. Have mercy xD

Teridax682 votes

No worries at all, and thank you for having taken the time to draft up that wall of text! Looking at the system you have proposed, I quite like it, and I think that, if the system doesn't do this already, forcing non-crits to happen in order to completely avoid lucky streaks could help ease some frustrations tied to the randomness of the mechanic. If I had one concern, it is that this kind of mechanic should also be unpredictable enough that no player can exploit it for reliable crits and non-crits, which might just happen in practice with the solution you proposed. I also very much liked that you covered unlucky streaks with your solution, which I think ensures fairness and means even fans of crit or marksman players would stand to benefit from it.

Lord Desert2 votes

Sooo, I'm finally done with this monster! I hope you enjoyed my old post :D First off, I will put most of the math at the end of this so it doesnt interfere that much and I can keep it a lot shorter that way :D And this comment is writen in a more general style, so guys who want to join our discussion have it easier to catch up. But its still an answer to your comment :P Let's get started!

#The Good, the Bad, and the Crit Chance...

  • 1. Can Crit chance be removed?
  • 2. Crit Chance not as bad as it seems?
  • 3. The critchance problem
  • 4. The problem with most suggestions/ crit alternatives
  • 5. Math, Math, Math
  • 6. Super short PRD TL;DR:
  • 7. My suggestion, a stricter PRD

1. Can Crit chance be removed?

From a technical standpoint: Yes, definitely! It is possible to remove it. But it might not be that easy! As you already mentioned, the game is already pretty complex. The "boringness" of AAs can mostly get removed by abilities, mechanics and the overall kit/theme of a champion. The removal of (controlled) randomness is also pretty good from a competativ standpoint, as it makes things absolutly fair. So, given a good alternative, crit chance can be removed!

Before I go to the suggestions to replace critchance, I will quickly sum up some positves and its problem, for completions sake.

2. Crit Chance not as bad as it seems?

Like stated before: crit chance is not pure random, this is very important! It uses PRD (pseudo random distribution), which is a very big improvement in terms of "fairness". To sum it up it looks at your previous attacks and modifies your critchance, instead of beeing completely independent, to get an overall more equal and less spikey outcome. I will explain it later with "some" (a ton of) math later. So to make it clear: Its not completely random, it already eliminats a big amount of bad cases, bad outliers.

Crit chance does not only lower the "boringness"/"predictableness" of AAs by a bit, its made in a way that gives very rewarding visual feedback! The big number with the crit symbol, the chunk of the enemy healthbar and a crit animation/projectial, they all feel exciting and rewarding. Not that "rewarding" for the recieving end but it will still get some excitment, a "holy shit, that dmg!", out of you :D

(Low) Crit chance is a double-edged sword, for you and your enemy!

Let's look a bit deeper: crit chance, while beeing random, is still "fair" overall. It can bite you, but it can also bite your enemy! If you invest in crit chance, the first amount (<=30) is roughly wasted combat stats. Why? Because its still to unreliable! You cannot count on getting the amount of crits you "should" get, in short trades. If you always bet on your luck, it will backfire someday. So on the baseline you "waste" gold at the beginning, the risk is high. Obviously you can get lucky, but by the end of this comment you will look at low critchance and critstreaks with completely different eyes! To summarize it, if you invest gold into crit chance you have the right to get power for your money, but with low crit chance its more a risk than anything. You cant make reliable decisions on low critchance.

3. The critchance problem

We now know that low critchance is rather bad, its very unreliable, thus its the "worst" part of the critchance randomness, right? Yes and No. In terms of randomness, yes; it is the worst part. But in terms of "impact", no! In the early laningphase 1 or 2 crits dont decide the hole game or the hole laningphase. The problem is if there are 2-3 lucky crits in one trade!

If someone buys crit chance its expected that he/she crits you sometimes. A crit is "only" another free AA. Thats something that you should be able to roughly calculate/expect. But 2 or 3 crits in a 5 AA trade with only 10% crit? Thats a bit much! So its not the amount of crits that makes things nessecarily bad, its the amount of crits in a short time period that derail everything!

And it goes the other way around with high amounts of crit chance too! Even with high critchance you can crit less than you would expect, and lategame every single crit can decide the game! The difference between 300 or 750 raw dmg at the end is much more impactful than ~80-100 more in the laningphase.

The problem is not just the random amount/ocurence, its the frequency of crits that makes things bad.

Keep this in mind, its important for later!

4. The problem with most suggestions/ crit alternatives

A lot of people suggest to change crit chance into a flat modifier, or something like "every x AA deals more dmg/bonus effect/becomes a crit" etc., or working like fevor, or something along those lines. The big problem, a very big problem with all of them, is abusability!

Predictable, guaranteed "crits" are extremely unheatly for the game!!! Especially with low "crit chance".

I cant stress this enough: If something can be abused, it WILL BE abused! Its the same with balance suggestions, always treat the "worst case" as the default one! And if the worst case looks bad than you need to tweak your suggestion.

So any kind of "stacking" or x-hit thingy will not work. The can be abused to easly! Back then Trynda could "scout" if his next AA would crit by looking at his attack animation. If it was the start of his crit animation he cancels the AA and then goes in for a trade with a guaranteed crit! This was very unheatly and was abused a lot. This was luckly fixed, so if you cancel your crit animation you next AA is no longer guaranteed to be a crit.

To make it short, we have 2 choices:

  • improve these suggestions, so they can no longer be abused
  • improve the current system, so these bad frequencies dont occure/ are limited

I personally am ok with the current system, I do have a suggestion to make it even better, but if theres a good alternative I'm fine with that too!

The Math and my suggestion to improve the current situation

5. Math, Math, Math

First off, congraz for making it here and still beeing awake! Not that many people here on boards would read such a massiv wall of text and I'm glad your not one of them! You have my respect.

Since we are approching the limit of text that is endureable in one go, I'll skip the basics of PRD and just refere to an old post of mine. It is still valid, just some minor number tweaks especially in the higher critchance region have changed. It will explain how PRD works and will contain a length description and breakdown of my crit chance improvement suggestion.

So if you want to torture yourself a bit, I recommend you get yourself some snacks and try to enjoy it as much as possible :D Heres the link to my old post, go to point 3. "More Math": SummonersLab, Part 2: Crit Algorithem Improvement And heres the link to a PRD breakdown: PRD Basics

6. Super short PRD TL;DR:

If you have 50% crit chance, you actually only have a 25% chance to crit on your first AA!

If this AA DIDNT crit,_ your next AA gets an additional +25% on top!_ So your 2nd AA has a 50% chance to crit. If your AA still did not crit, it gets another +25%, so it has a 75% chance to crit in total. If it still did not crit, it gets +25% again, so we arrive at 100%, which means this AA will guaranteed crit!

After a crit you will reset back to the initial 25% crit chance.

So this series of Crits: _C,C,C,C _ doesnt not have a 1/16 (~6%) chance to occure ( 1/2x 1/2x 1/2x 1/2), it only has a** 1/64 to 1/256 (~1.6 - 0.4%) chance** to happen (1/4x 1/4x 1/4x 1/4) !

As you can see getting alot of crits in a row is VERY unlikey! Even with decent crit chance. Keep this in mind the next time you see this, that person just got pretty lucky!

7. My suggestion, a stricter PRD

Even with the PRD in place, which already makes the randomness a lot more fair, there are still some outliers present! It is for example possible to get a lot of crit in a row even with super low crit chance(<20%)! It is highly unlikely, but still possible. The same but a bit stricter can happen with high crit chance, since you reset back to the lower crit chance after a crit means that you have in general a lower critchance than you should have, in short periods.

As a quick reminder: Over a longer period prd has the same outcome as normal crit chance, they are equal. But in shorter periods things can get wonky! (PRD is still a million times better than normal randomness, even in these cases) And thats what I want to improve!

In short: We introduce caps to eliminate those outliers that can break "fairness" in shorter periods. We still wand to be random and also dont want to interfere with the normal PRD, we are only touching the troublesome outliers! These caps force us to have NonCritical AAs or guaranteed Crits based on the actual crit chance. Remember the amount of crits IN A SHORT PERIOD was the big problem we identified. Thats what those caps to fix.

The Cap:

  • <=20% CritChance = 2 NonCrits forced after a Crit
  • >20% - <40% CritChance = 1 NonCrit forced after a Crit
  • >=40% - <50% CritChance = max. 2 Crits in a row (3rd AA will not crit)
  • >=50% - <60% CritChance = max 3 Crits in a row AND After 3 NonCritAAs in a row, your next 2 AAs are guarenteed Crits
  • >=60% - <80% CritChance = max. 5 Crits in a row AND After 2 NonCritAAs in a row, your next 2 AAs are guarenteed Crits
  • >=80% CritChance = After a NonCritAA, your next 3 AAs are guarenteed Crits

Yes this is a tbit complicated, but basically these caps ensure that lucky AND unlucky outliers in high and low critchance get removed. If you have questions or want a more indepth explaination of this, just ask :D Reached letter limit again..

Lord Desert2 votes

btw its now 5 am here and I wanted to got to sleep at 11pm... rip me xD Also to stay in the discussion, I feel that crit is fine, could use some improvement and alternatives to it are not ready yet. They still have problems :/ If someone has an idea that would get rid of these problems I would gladly hop on board! And also PRD can be abused, but it not really practical. My suggestion is even less likely to be abused even when I introduced alot of guaranteed crits (which is in general pretty risky/bad).

And I'm sorry that my reply is not as "tailored" to your comment as it should be, but in this cases it makes alot of the explainations and reasonings easier :S If it helps I rewrote this monster 2 times and it was even longer in the earlier iterations, and I still managed to get to the letter limit again xD

Teridax682 votes

Congratulations on such an amazing post! Here are my thoughts on its components:

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e00000000000000000001,timestamp=2018-02-21T03:58:13.836+0000)

As you already mentioned, the game is already pretty complex. The "boringness" of AAs can mostly get removed by abilities, mechanics and the overall kit/theme of a champion.

I very very much agree that crit could potentially be removed from the game, and isn't as essential as many think. Sure, removing crit would require a rebalancing to many marksmen and some melee carries, but stats like attack speed and attack damage have much more room to be increased here. Even if those aren't what get targeted, champion kits are the perfect spot to add in more power, especially more conditional power unique to each champion.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e00000000000000000001,timestamp=2018-02-21T03:58:13.836+0000)

Like stated before: crit chance is not pure random, this is very important! It uses PRD (pseudo random distribution), which is a very big improvement in terms of "fairness". To sum it up it looks at your previous attacks and modifies your critchance, instead of beeing completely independent, to get an overall more equal and less spikey outcome.

This too I think is critically important (/pun). As unfair as crit can feel, its actual power is regulated to limit the impact of extreme luck, so at the end of the day it's basically a less clear/consistent way of adding burst onto every few attacks.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e00000000000000000001,timestamp=2018-02-21T03:58:13.836+0000)

Crit chance does not only lower the "boringness"/"predictableness" of AAs by a bit, its made in a way that gives very rewarding visual feedback!

This is true, though I think this is also somewhat dampened by the fact that crit usually gets stacked to very high amounts, normally 80%. It definitely feels like an awesome burst of damage when it happens less than half the time (when it's at its most unstable), but past a certain point crits just end up being "normal" damage per attack for marksmen, causing crits to either become unimpressive, or become so overly prevalent that crit users end up outputting burst per second.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e00000000000000000001,timestamp=2018-02-21T03:58:13.836+0000)

If you invest in crit chance, the first amount (<=30) is roughly wasted combat stats. Why? Because its still to unreliable! You cannot count on getting the amount of crits you "should" get, in short trades. If you always bet on your luck, it will backfire someday.

This too I agree with, and while I think that is somewhat intentional (Riot doesn't want fighters as a whole class to build crit for some reason), I agree that the current way crit is implemented means there's always this intermediary period where critical strike is really unstable, even with a PRNG controlling its outliers.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e00000000000000000001,timestamp=2018-02-21T03:58:13.836+0000)

In the early laningphase 1 or 2 crits dont decide the hole game or the hole laningphase. The problem is if there are 2-3 lucky crits in one trade!

I actually feel that even one crit in the laning phase really can determine its outcome in certain circumstances. This is why players used to run 1% crit in their old rune setups on the off-chance that they chunked their opponent in a trade. After the first few levels, it's not as bad, but it can really mess people up in the very first few minutes. I do, however, agree that a streak of luck on either side due to crit is guaranteed to frustrate at least one person, which is why random crit distibution needs to obey strict rules if it wants to appear fair at all times.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e00000000000000000001,timestamp=2018-02-21T03:58:13.836+0000)

A lot of people suggest to change crit chance into a flat modifier, or something like "every x AA deals more dmg/bonus effect/becomes a crit" etc., or working like fevor, or something along those lines. The big problem, a very big problem with all of them, is abusability!

I too think that many proposed systems, including that energized crit system, do have a problem of abusability, though I don't think the ability to maximize the potential of a mechanic is necessarily abusive. If I may shamelessly plug my own suggestion here, I proposed an alternative crit model that would basically function like the three-hit passives we see on champions like Vayne or Vi: for sure, the mechanic can be optimized through certain types of play, and would likely require caution/rebalancing towards champions who'd be especially good at applying it, but because the means of applying crits would be innately more interactive (the opponent could play around this model as well), I don't think it could really be qualified as abuse.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e00000000000000000001,timestamp=2018-02-21T03:58:13.836+0000)

If you have 50% crit chance, you actually only have a 25% chance to crit on your first AA!

If this AA DIDNT crit,_ your next AA gets an additional +25% on top!_ So your 2nd AA has a 50% chance to crit. If your AA still did not crit, it gets another +25%, so it has a 75% chance to crit in total. If it still did not crit, it gets +25% again, so we arrive at 100%, which means this AA will guaranteed crit!

After a crit you will reset back to the initial 25% crit chance.

In short: We introduce caps to eliminate those outliers that can break "fairness" in shorter periods. We still wand to be random and also dont want to interfere with the normal PRD, we are only touching the troublesome outliers! These caps force us to have NonCritical AAs or guaranteed Crits based on the actual crit chance. Remember the amount of crits IN A SHORT PERIOD was the big problem we identified. Thats what those caps to fix.

I commented a little on this on another reply, but to reiterate, I really like this system. Introducing hard caps to the current pseudorandom number generator, in a manner that wouldn't lend themselves to predictability, would allow the current crit model to feel more consistently fair, without requiring a total overhaul. While I still wouldn't really be a fan of the remaining randomness (the order in which you crit can be important in some fights), I still think this would be a tremendous step forward in establishing a crit system that is less harmful to the integrity of what is meant to be a competitive, skill-based game. Kudos again on the massive and well-written post!

Lord Desert2 votes

Kudos again on the massive and well-written post!

I very much appreciate it! At first it was just 1 massiv text without much that structure and when I looked at it again, even I wouldnt want to read till the end xD So I had to do some big restructuring, hopefully it was bearable :D

Sure, removing crit would require a rebalancing to many marksmen and some melee carries ...

Yes, it definitely would! Which is why riot wouldnt do it in the first place. The amount of work put into it wouldnt be justified by the outcome :/ But some "small" fixes should definitely be fine. Maybe when they look at adcs in general again, then there might be a chance to make some bigger changes, if they would want to do some.

I think this is also somewhat dampened by the fact that crit usually gets stacked to very high amounts, normally 80%.

At this point, aka the late game, it IS wanted that crits arent "luck based" anymore! At that point it servers as a flat dmg modifier like most suggestions suggest. You dont want to have luck involed if every AA, every Crit can decide the game ;D

Crit chances "anti-boringness" phase is especially designed for the mid game. Where most people have everything availabe, but adcs still lack ether the AS or the dmg to create "interesting" situations. They can use show off all their tools, but the outcome is pretty low at that point. (Remember when it was tank meta where adcs were oblivous because they werent even close to fight tank in the midgame on equal grounds? Thats when botrk and the crit items where buffed!) Crit is mainly there to make this (adc-)midgame "interesting" and later serve as a flat dmg modifier.

I havnt seen your suggestion but I surely will check it out! To adress the "abuse"/maximize part: If a champion is balanced around that, that is complettely fine. But it has severe results for the champion, if "normal" people cant reliabley use it, only skilled players. This will create problems like Ryze and Azir! Bad in normal skill range and good in high skill range, or worst in normal and balanced in high. Its easier to remove "abuse" chases than to balance everything around them.

So if the champ is balanced around this, its fine, but if this abuse is not the adressed by the overall balance, things can and will get bad! Really bad!

For my suggestion: It doenst work wonders, its only designed to remove the worst cases, you still can get pretty unlucky! So it makes things better, but is still 100% (pseudo-) random, just so you dont get your hopes up to high :P And like I said most would never even notice a change. You can still boost your chances but not even close to beeing "abusable".

Its pretty nice to actually have a discussion once in a while here on boards xD They are (sadly) pretty rare :/

Teridax681 votes

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e0000000000000000000100010000,timestamp=2018-02-24T19:45:25.884+0000)

I very much appreciate it! At first it was just 1 massiv text without much that structure and when I looked at it again, even I wouldnt want to read till the end xD So I had to do some big restructuring, hopefully it was bearable :D

Very much so! I very much enjoy reading your posts, and your structuring is excellent.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e0000000000000000000100010000,timestamp=2018-02-24T19:45:25.884+0000)

Yes, it definitely would! Which is why riot wouldnt do it in the first place. The amount of work put into it wouldnt be justified by the outcome :/ But some "small" fixes should definitely be fine. Maybe when they look at adcs in general again, then there might be a chance to make some bigger changes, if they would want to do some.

I wouldn't put it entirely past Riot to rebalance several champions in view of a stat rework, since we've had massive stat passes in the past for less (at one point every champion got armor buffs after a single rune got reworked, for example), but I do agree that such a change would likely require mid- or preseason level work, and wouldn't be done on an average patch. I personally wish Riot could do another pass on marksmen as well, since I think the Marksman Update didn't really improve the class as much as it could've, and missed a huge opportunity in not reworking crit then.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e0000000000000000000100010000,timestamp=2018-02-24T19:45:25.884+0000)

At this point, aka the late game, it IS wanted that crits arent "luck based" anymore! At that point it servers as a flat dmg modifier like most suggestions suggest. You dont want to have luck involed if every AA, every Crit can decide the game ;D

Crit chances "anti-boringness" phase is especially designed for the mid game. Where most people have everything availabe, but adcs still lack ether the AS or the dmg to create "interesting" situations. They can use show off all their tools, but the outcome is pretty low at that point.

This is true, though I also think it highlights some of the issues with crit in its current form: in lower amounts, the stat is inherently unstable, so a crit in the mid game can feel amazing for one player, but terrible for the other. Past a certain point, though, the stat may as well be a persistent damage amp, and as such it categorically offers no additional gameplay relative to other effects. I agree that the bursty output of crit is important to making marksmen feel more interesting to play, but because of the way that output is currently implemented, it's forced on this sliding scale of unfairness versus boringness: the less consistent crits are, the more exciting each one is, but the less fair it feels as well to get hit by one; the more consistent crits are, the less exciting each one becomes. Crits as a random damage amp I feel are therefore doomed to never truly enhance gameplay without introducing some severe tradeoffs. This is also, however, why I like your proposed improvements to the current crit PRNG, because your suggestions make the very inconsistency of crits more consistent, via enforced crit downtimes or uptimes, and therefore could make it feel a lot fairer.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e0000000000000000000100010000,timestamp=2018-02-24T19:45:25.884+0000)

To adress the "abuse"/maximize part: If a champion is balanced around that, that is complettely fine. But it has severe results for the champion, if "normal" people cant reliabley use it, only skilled players. This will create problems like Ryze and Azir! Bad in normal skill range and good in high skill range, or worst in normal and balanced in high. Its easier to remove "abuse" chases than to balance everything around them.

I agree with this 100%, and I think it is important for all champions to have a baseline amount of reliability, so that they don't become completely useless if things don't work their way, or becoming impossible to play against if they're in the hands of someone like Faker. However, I think one big problem with marksmen right now is that they're too reliable, and lack interactivity to their damage output especially. In this respect, introducing gameplay to their crits that both they and their opponents could play around could alleviate those issues, and hopefully open more opportunities to make marksmen easier to pick up otherwise. I believe marksmen are a class that are heavily balanced around artificial difficulty, i.e. difficulty that isn't dependent upon interaction with other champions, and that they'd become significantly easier to play with some basic improvements to the move and attack-move commands. Doing so while also making marksman less reliable against their opponents could therefore keep their power level at around the same, but also make them feel both a lot smoother and much more interactive.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e0000000000000000000100010000,timestamp=2018-02-24T19:45:25.884+0000)

For my suggestion: It doenst work wonders, its only designed to remove the worst cases, you still can get pretty unlucky! So it makes things better, but is still 100% (pseudo-) random, just so you dont get your hopes up to high :P And like I said most would never even notice a change. You can still boost your chances but not even close to beeing "abusable".

I don't think it's really the job of any single mechanic to work miracles, particularly since I feel many of the problems with marksmen are due to their fundamental class design, and not just a single stat, but I do think your proposal could introduce some major improvements by completely eliminating certain outliers that should never happen. If nothing else, your proposal would guarantee that there would never be even a possibility for exceptionally lucky or unlucky streaks, so even if there would still be complaints about the randomness of crits, there wouldn't really be any question of their fairness, since marksmen would get almost exactly the number of crits that they paid for, at all times in every fight.

[{quoted}](name=Lord Desert,realm=EUW,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=000e0000000000000000000100010000,timestamp=2018-02-24T19:45:25.884+0000)

Its pretty nice to actually have a discussion once in a while here on boards xD They are (sadly) pretty rare :/

Agree completely. I think there are plenty more people who genuinely want to also have a productive discussion, healthy debate, etc., but the structure of the Boards pushes people to farm upvotes and compete in popularity contests. More generally, people often find it easier to yell at each other than to try to understand each other, and that's been limiting the range of discussion that can realistically happen in this space.

HarrowR2 votes

Well its funny you generalize all adcs as boring right clickers where there is no zero or hero mechanic,you seem to forget theres a lot of skill heavy adcs,Jhin,Ezreal,Xayah,Varus,Lucian,even kog has 3 skillshots,Draven has two skillshots and a mechanic that requires him to move and catch the axes,Kalista has a similiar mechanic where the skill of the player has great influence on the success of playing her. Most of the time the outplay of adcs isnt about whos gonna crit or not,its about whos gonna kite better,which is a skill in its own cause it takes hella of a lot more APM to kite a melee with an adc than with lux where you just throw your spells out and keep walking

Lord Desert1 votes

You cleary missunderstood my point xD I'm not talking about the complexity or abilities or mechanics or skill of adcs! I'm just saying that adcs gameplay mostly resolves around their autoattacks and that you cannot dogde them with skill (except for some spezial cases like jax e, or twitch r aso).

And if you cannot dogde/avoid something with skill, it becomes pretty predictable and rather "boring" compared to skillshots which always can miss completely. My point stands: You cannot outplay the rightclick, aka AAs, which makes AAs in general less exciting. Technically point and click abilities are also on the "boring" side, as a sidenode.

My point is that crit chance wants and does make AAs alot more unpredictable, thus more "interesting". I'm not saying that crit is good or bad, it just fullfilles this spezial purpose and also not that all adcs are just boring rightclickers. The AAs are boring, not the Adcs themselfs! Thats my point. Nothing more and nothing less.

Hope that clearifies things a little :D

HarrowR1 votes

"The AAs are boring, not the Adcs themselfs!" " point and click abilities are also on the "boring" side" Well i guess we just have to agree to disagree,since i find most champions outside of adcs and some aa reliant melees like Yi kinda boring,or at least not as fun as crit/aa champs,also i prefer champions with more point and click,less things to miss and mess up mechanic,for instance i dont play a lot of mages but i love playing Malzha,who has basically one skillshot. I just dont see champions and gameplay as fun or not from a "does it involve skillshots or not" point of view. its more about the whole dynamic,the whole kit and gameplay of the champion.Is it skillshot or not affects my satisfaction of one champion much less than what the actual abilities are and how the champ plays in general.

jadelink1 votes

Yes, people really DO play around crits, though probably not so much in low Elo. Trynd will often 'crit fish' and then all in if he gets one.

I remember a clip from boxer pete (a tynd main streamer) who says, just before spinning in on a teemo with blind down, that if he got a crit he won, otherwise he lost.

I've seen (as a jungler) a trist stop running and leap INTO me in a gank, as she got a double crit and could now safely go in, get resets and murder us all.

"How is this different from any other kind of nuke? Do critical strikes possess some unique property that allows them to bypass gold or level leads?"

I KNOW how much damage an annie does with those items/lvls. I don't know if that adcs might get a crit, or 2, I thus have to 'roll dice' to calculate if I can all in them as a juggernaut, with predictable damage/engage patterns. RNGeesus can get you out of a gold/lvl lead disparity with crit, due to its unpredictable damage, is what he is saying.

ModThe Djinn5 votes

#SUGGESTION 5: Scaling Crits A rather unusual suggestion that keeps crits more consistent but weaker at early levels, and keeps them bursty at later levels.

The model sets critical hits at a 50% chance, converts critical strike chance to critical strike damage, and doubles the value of critical strike damage (so Zeal would give 40% critical strike damage), scaling up to 200% bonus damage. Critical strikes begin at 0% damage, and thus would not trigger.

Example Tristana has a current attack damage of 100, and a 40% critical strike damage. She has a 50% chance to deal 40 bonus damage. Over 10 attacks this equals 200 bonus damage on average -- the same output that the equivalent 20% critical strike chance would get from two crits for 100 bonus damage each.

Positive Impacts

  • All characters that build any crit get a good chance to see critical strike animations.
  • Early game crits hit for significantly less than they do currently.
  • Crit occurrences are far more predictable.
  • Late-game crits feel REALLY chunky.
  • Crit isn't hugely impactful in the laning phase as a randomization element, since it's so common.
  • Crits still feel good in the late-game.

Negative Impacts

  • Critical strike animations might become too frequent.
  • Crits can end up really bursty -- since values over 50% crit chance don't exist, damage has to increase by 200% bonus damage instead of 100% max ranks.
  • Removes much of the existing gameplay around critical hits, and especially skews later fights if you can engage right after a crit has landed, since hits are bursty.

#FURTHER ANALYSIS

Does it reduce the perception or impact of randomization? In theory, but might actually increase feelings of unfairness due to the huge crits that are possible.

Does it create a strong feedback loop? Yes. Crits are, if anything, bigger and better as the game goes on.

Does it create highs and lows in gameplay? Absolutely. Highs become really big, lows about as flat. The frequency of those highs may be an issue though.

Does it retain interaction points or introduce new interaction points? Yes. All existing interaction points are preserved, just at reduced individual impact. Frequency of these crits may weaken this though.

Does it have major ripple effects? Possibly. Hard to predict how reliable crit access would affect off-role itemization.

How does it impact the laning phase? Harass becomes more consistent but less bursty, which may slightly favor characters who can land it reliably, and crit-fishing is diminished in value. Heavy crit rushes (over 100%) can lead to very burst situations.

How does it impact objective control or teamfighting? Really changes where squishies have to be placed in the mid-to-late game, as those bursts become huge.

How does it affect the metagame? Likely pushes Marksmen into being squishy-assassins further, which is problematic.

How does it impact snowballing? Doesn't affect lane snowballing hugely (slight favor to the dominant laner), but when out of lane early crit purchases can make you a nightmare for squishy opponents.

InTheory3 votes

Out of all the systems I dislike this the most. The stronger a burst crit becomes, the more likely a fights is gonna have a different outcome depending on if you crit or not. What was formerly an 80% critical strike chance would now be +160% critical strike damage. Landing such a crit at the start of a fight against a squishy support for example could instantly remove that from the fight and change the outcome at the very beginning because of that "lucky 50%".

It is obviously the most extreme system so far and increases certain negative aspects of crit even more.

ModThe Djinn2 votes

{quoted}

Out of all the systems I dislike this the most.

Total agreement here, and the reason the next one is a hybridized version of this one that tackles the extreme burst edge casts. :)

Sire Hippington4 votes

First off, there is some other negative aspect off the RNG: It allows a class that is ment to be about sustained dps to have upfront burst. I also have issues with crit in it's funktion as aditional multiplyer for AAs, as it moves all power to non interactive righclicks and awqay from the actuall kits of a champ while forceing a lategame hyperscaleing powercurve, but let's focus on getting rid of RNG for now.

As for anti snowball, i allways found that to be a stupid argument, i don#t think crit ever influenced decisionmakeing alot and if RNG contributes to a comeback, it doesn't feel like a deserved comeback at all. For me, that's more a negative point, and definitly something that hardly anyone would ever notice if it's gone. I also experience the critfishing not as positive or exiting gameplay at all, so that one can argueably be put on negative as much as on positive, since crits don't create interesting scenarios you can react to, they just create a onesided numerical advantage and that's it.

So about the existing models:

  • Precision I feel like Precisiion would be the easist way to solve it, it basically wouldn't need any big balance or kit followup while solveing the RNG related issues. However, it would be certainly the most boring option, though since most adc reach 80-100% crit at fullbuilt, the 'boring' consistency doesn't seem to bother them...

  • Focus A bit complicated which causes it to be hard to keept track of. In Theory, it's 100% predictable, but in practice most players will not be able to keep track of their opponents focus and thus crit spikes will still come supprising and create moments of frustration, just how few players actually can keep track id shive procs ans such. Ranoness is not really and actuall thing, it's a precived thing, In theory, if you have all the data at hand, you can calculate the outcome of a dice role by applying physics. But in practice, we call it random because we simply do not have the needed data(and most also lack the needed equations) to actually calculate or atleast ppredict the outcome. A similar issue comes with focus, in theory you can calculate it, but in game you most likely can't and thus the procs are precived somewhat random. . It also creats play patterns similar to caits headshot where you built up the crit and then zone your opponent, which means you can get really high value from even small crit numbers duering the laneing phase. It also still creats upfront burst for a sustained dps class. While this is gameplay, i don't htink it's healthy gameplay. . P.S. seems like you just did copy-paste the positive ans negative points from Precision. You might want to fix that, cause focus for sure creates high moments, gameplay and feedback.

  • Overcrit It retains RNG but smalers the impact. While i would prefer it to the current crit, i think if we were to adress crit and it's RNG, we should fully remove the RNG and not go for something half assed that retains all the issues just on a smaller scale.

  • Hybrid Precision Same as with overcrit, it doesn't fix the issues but lowers the impact. Not a fan.

  • Scaleing Crits and hybrid scaleing crits Again not getting rid of RNG and while it's an improvment by lowering the impact of crits, it's not really solveing the issues.

So, what i purpose is something that goes into the direction of Focus and Scaleing Crits(maybe call it 'Critical Strike'

Every 4th hit against the same target is a crit and applies special effects. Baseline it does the normal 100% damage with no further effects. Items could either simply increase the damage of that hit, or add a speciall effect to it similar to hiw energized works, IE could change it from every 4th to every 3ed hit

The positive aspects would be:

  • it's predictable both in theory and since it's a fixed number of hits also in game(unlike focus)
  • Since it's a fixed rate, one can be creative with on-crit effects which you can't do in a system where crits can proc on nearly every attack, allowing to evt add
  • some utitllity rather than just raw damage to the mechanic, allowing to adress the issues of crit as hyperscaleing multiplyer on point n click damage ontop of the RNG.
  • It creates gameplay and emphasises the singltarget dps focus of adc with the per target stacking while backloading the damage to avoid upfront burst and the 'caitlin situations'(holding onto the stacks for zoneing)
  • it Creates big moments with statisfying feedback that gets boosted by the 'finishing-move' character of the backloaded burst.

The negative aspects would be:

- can't be translated 1 to 1 from the old system, so it's alot of work to implement that requires major work on items and champions

  • evt creates an overload on the crit, makeing it so that getting that 4th hit is all adc are about, meaning champ like lucian that can do so faster and more reliable would be dominant
  • 4hits on the same target might be a to harsh condition to it, so maybe it needs to either be reduced to every 3ed hit and with no items that change that, or to a not targetz bound stacking, but therefor with different stacks for heroes and minions/monsters to still avoid the caitlyn-zoneing and upfront burst.
Teridax683 votes

#Suggestion X: Jinxylord Crits/Add-On Crits

  • The model:
  • For every champion, every third strike on a target, i.e. basic attack or offensive ability, is counted as a critical strike, using the critical strike animation if the triggering effect is a basic attack, and a third basic attack animation is lacking.
  • Crit items no longer give a random chance for a crit, but instead add on a bonus effect to each crit. For simplicity's sake, let's assume current crit items grant bonus physical damage on critical strikes equal to triple their critical strike chance.
  • Current three-strike effects are now crit modifiers, adding their own bonus effect onto the champion's critical strikes.
  • Critical strike modifiers need not only be damage, and may add utility or crowd control to every strike, depending on the modifier.
  • Examples:
  • Vayne gets Zeal. On every third strike, she deals her bonus true damage, plus 60% AD bonus physical damage, for especially meaty crits.
  • Shaco backstabs an opponent. This triggers a guaranteed critical strike, activating his own bonus damage plus any other effect, damage or otherwise, he may have picked up from his items.

##Pros:

  • No random output.
  • Stabilizes crits by making them happen at a fixed rate per attacks, instead of the current curve where crits start out infrequent, then very frequent.
  • Maintains the bursty nature of crit with "normal" attacks leading into damage spikes.
  • Creates an extremely solid feedback loop, as demonstrated by existing three-strike effects.
  • Relies on gameplay that has already proven to work on several kits, reducing the level of uncertainty in implementing such a model, in contrast to more theoretical propositions.
  • Provides meaningful decisions for both the crit user and their opponent, due to a mode of application that would require commitment, and therefore telegraphing.
  • Fully emphasizes skill by exclusively rewarding more interesting plays.
  • Creates highs and lows that fully depend on the interaction between two players.
  • Fits much better with the thematic of critical strikes, which are meant to be about squarely hitting an enemy with your full power.
  • Creates a framework in which non-marksmen can also have their own crit-based gameplay, via their own crit modifiers.
  • Allows a much more diverse and more balanceable set of gameplay effects besides just bonus damage, as the power of any CC or utility effect would be tempered by the pacing of these new crits, as well as made more stable.
  • Would allow us all to call three-strike passives "crits", thereby better integrating the mechanic into the game.

##Cons:

  • Would require readjustments to kits that innately rely on random crit, namely Gangplank, Tryndamere and Yasuo.
  • Would require readjustments or even full reworks to certain items and runes that rely on random crit, namely Fleet Footwork, Infinity Edge and Statikk Shiv.
  • Would alter the gameplay of marksmen by back-loading more of their damage (not necessarily a con).
  • Would complicate the balancing of kits with spammier and/or long-ranged abilities, as the model includes abilities as components to critical strikes.
  • Would introduce complications to AD fighters and assassins, who may wind up abusing the non-random nature of this crit model (especially if they have spammy abilities).
  • Quite a few players are tired of seeing a repetition of three-strike effects.
InTheory2 votes

One last thought here before I go to bed (getting late in europe):

I do somewhat like the hybrid versions because they at least treat the RNG output of critical strikes by flattening the best and worst cases out of constantly landing or not landing critical strikes. Mathematically they'd be an improvement to the current system regarding my own interest.

HOWEVER: The major fact that completely disqualifies these hybrid systems is, that it's really hard for the average to track down what they get power wise when buying crit. They'd now have to do the math upon buying crit in terms of crit chance + crit damage or alternatively crit chance + basic attack amplifier. Riot has always tried to keep things as simple as possible and that'd be something I consider to be too difficult.

The best point you made is that while marksmen are tied so heavily to basic attacking, they should have something to play around that makes the single thing they constantly do more attractive. Yet I think this is based on the class to be not the way I'd like them to be. I think these crit interactions should be tied more towards a champion's kit like caitlyn or ashe to make the interaction with crit a) more intersting overall, b) more unqiue for every champion and c) makes it easier for players to keep track of critical strike as a state because they are part of champions/marksmen's kits.

I guess you're not gonna get too much attention on this post - it's too long and gameplay boards members don't like long posts, complicated themes and questions without populist answers. Easy answers to complex questions it the to go for this sections (I made a post that was so long that basically nobody wanted even to read it - even if they were interested in the matter).

Keep up your ideas and maybe you come up later with ideas about how to improve not only crit itself but also its environment in league.

Good Night!

Lil Homie busta2 votes

I think crit is missing the anti-tank thing. Maybe critical strikes could ignore a percentage of the target's bonus armor (non-base armor) and the bonus damage scaling could be decreased as consequence? That would make crit less threatening to non-tank classes and more consistent around the "shred" theme.

MagicFlyingLlama1 votes

I was coming here to suggest the exact same mechanic as Focus.

It would keep the same feel and roughly the same DPS as current crits, not be a balance nightmare, but also eliminate randomness and give people something to think about.

ModThe Djinn1 votes

{quoted}

I was coming here to suggest the exact same mechanic as Focus.

It would keep the same feel and roughly the same DPS as current crits, not be a balance nightmare, but also eliminate randomness and give people something to think about.

That's interesting, as I personally feel that Focus was, while possibly the most interesting, the biggest balance nightmare of the bunch due to how it allows assassins or divers to build up critical hits in preparation for a super-damaging engage, even at low crit values. A single Brawler's Glove, for example, lets you load 100% of your AD into a burst.

A possible solution would be to put a maximum uptime on how long you can retain a focus "proc," but that might feel punishing to Marksman by back-loading their teamfight damage heavily.

only zhule2 votes

[{quoted}](name=The Djinn,realm=NA,application-id=3ErqAdtq,discussion-id=voeyqoFA,comment-id=00060000,timestamp=2018-02-18T21:36:53.017+0000)

That's interesting, as I personally feel that Focus was, while possibly the most interesting, the biggest balance nightmare of the bunch due to how it allows assassins or divers to build up critical hits in preparation for a super-damaging engage, even at low crit values. A single Brawler's Glove, for example, lets you load 100% of your AD into a burst.

A possible solution would be to put a maximum uptime on how long you can retain a focus "proc," but that might feel punishing to Marksman by back-loading their teamfight damage heavily.

If I may make a suggestion. Change the focus of WHO builds up the crit. Instead of the champion building a crit, instead the target builds a debuff that reaches 100% and the next attack (from the one building the debuff) does the bonus damage to them.

This keeps champions from building a crit on minions then getting a free orbital nuke on a squishy.

Tzuyu Is My Bias2 votes

This would then be too punishing for switching targets.

only zhule1 votes

{quoted}

This would then be too punishing for switching targets.

That's kind of the point to be honest. High constant damage when you focus a target, but gives new targets a brief window to react to being targeted before you are landing 3 critical hits for every 4 attacks taken.

jadelink1 votes

This would require across the board heavy buffs and nerfs. YOu cant reasonably 3 hit an ulting vayne, or an akali with shroud, for example, giving them massive new combat bonus's. On the other hand juggernauts would be far easier to reliably 'crit to death'.

Zhonyas/hourglass would also relieve crit pressure, and may well be worth taking on a tank (hourglass at least) to 'juggle aggro' in 2 man frontlines.

Then the list of targets who can make themselves 'undesirable' targets are also buffed, but to differing degrees (yi say, as compared to Elise).

Champion Skin1 votes

You could just have it fall off outside of combat (would remove the assassin problem) and or have it build up slower on minions (would reduce laning power and this could be tweaked up and down making it less or more useful during laning phase)

A possible solution would be to put a maximum uptime on how long you can retain a focus "proc," but that might feel punishing to Marksman by back-loading their teamfight damage heavily.

And ADC usually builds at least 50%+ Crit meaning that they will crit on their second AA anyways which doesn't really back-load their damage that much

Lord Desert1 votes

I commented on your first post already, so I will make it short:

The main, probally even the only, reason why Crit works with random chances and not just a flat dmg amplification is to make the adc gameplay experience interesting/diverse!

Every other role is more or less focused on skill shot abilitys which can miss! Aka "interesting" gameplay. So it can fail and do nothing! Adcs main power come from their AAs, which always do dmg. They cant miss completely and deal zero dmg (in general).

You want uncertain, (normally skill dependent) outcomes! You will not watch a match if you already now the outcome from the get go, based on statistics. You watch it because of the uncertainties!

So if you want to talk about potential crit chances, always keep this in mind: Removing the "random" (its not completely random, its decently controlled pseudo random) aspect from the adc playstyle, you will make it a lot more boring! Thats why riot will not "exchange" crit chance for anykind of "flat"/"totaly predictable" dmg amps, if you dont fix this "boring adcs playstyle" in some way too.

I'm personally pro critchance aka. pro randomness, IF we are talking about controlled randomness! Pure randomness is bad for a competativ game. Controlled randomness on the otherhand isnt a problem in certain amounts, rather the opposide! Its good for it! It makes things more interesting.

Secondly, a point of uncertainty that I'd love to know if anyone knows it: does this psuedo-randomization tick different for champions vs. non-champions? I.e. is champion crit chance on a separate fail-safe tracker than minion crit chance?

I cant guarentee it with 100% certanty, but I'm 99,9% sure its not split into Non champs and Champs. There is no real point in splitting it, it doesnt make alot of sense. You can "stack" crit on minions/monster to get a better chance vs champs. But its not easy and only makes sense on some levels of critchance. I personally tested crit chance alot and I never saw anything like a split. So its pretty safe to assume that its definitely not split into to crit chances.

jadelink1 votes

I'd love a reply on this from RIOT. I remember the (horrible) crit-stack trynd rort, and how annoying it was to lane against someone had a guaranteed crit 'stacked' up their sleeve.

Lord Desert1 votes

While I would love to get a Rioter to reply to this, but I dont think it will happen :/ I do have some very technical questions regarding Crit left unanswered, which are pretty much impossible to figure out/test by normal means. It would take serveral hundreds of hours and I still couldnt guarantee certanty of the results...

I'm sorry to say but this "crit stacking" on trynda still works! Its a bit harder and its not really "guaranted", but you can still cheat the system to bring the odds a decent bit in your favor :/ Another trick for trynda was to cancel his AA when he saw the start of his crit animation. So hes know that he will crit with his next attack. This one luckly got patched :D

jadelink1 votes

Interesting to know. Yes, the old animation cancel, so he could 'store' a guaranteed crit was annoying, glad that ones truly gone.

Lord Desert1 votes

Normally the only way to "abuse" the PRD system is if you would have like 20-40% crit chance while still beeing in the laning phase. There you could try to "stack" your critchance on minions a bit and then have a bit better odds vs the enemy champ. But the "abusability" is pretty low, only yasou and tryn should be able to pull it off and even than the "advantage" is not that big! Its miles away from "guaranteed" crits or so :D

Disulfiram1 votes

Why shouldn't other options be considered besides damage? For example: considering that "critical hits" are essentially the game's way of displaying lucky hits on exposed areas, why couldn't they for example:

  • ignore a portion of enemy armor (as if you hit an exposed part, thus strengthening ADCs vs tanks at the cost of damage vs squishies)

or

  • cripple the enemy (a lucky hit to the eyes would blind anyone, one to the legs would slow, one to the hands would make it difficult to wield weapons, etc... one of these could be chosen as the default crippling effect of critting, and this would make the ADCs more self-reliant at the cost of some damage).
ModThe Djinn2 votes

{quoted}

Why shouldn't other options be considered besides damage?

I'm not saying they shouldn't be -- just that it was beyond the scope of this particular exercise, which was relative low-scope changes in the perception of the mechanic. :)

More from Gameplay

3 Refunds per season, upvote if you want 3 refunds per season.
Gameplay764 comments
7,167
votes
3 Refunds per season, upvote if you want 3 refunds per season.
Gameplay156 comments
3,510
votes
UPVOTE FOR URF
Gameplay606 comments
2,440
votes
Yes plz.
Gameplay456 comments
1,721
votes
3 Refunds per season, up vote if you want 3 refunds per season.
Gameplay114 comments
1,716
votes
When is ANYONE at riot going to address the awful new player experience?
Gameplay634 comments
1,383
votes

More by The Djinn

@Riot: In-Depth Feedback on Teamfight Tactics
Teamfight Tactics95 comments
114
votes
Veigar's Particles: Please Sell the Ult More
Gameplay57 comments
110
votes
Improving Report Feedback
Player Behavior74 comments
100
votes
Qiyana: Awesome visuals, but one big missed opportunity.
Story, Art, & Sound146 comments
95
votes
Fixing Critical Strike: A Proposal.
Gameplay148 comments
85
votes
Shotgun Knees for Fun & Profit [Urgot Suggestion]
Gameplay140 comments
77
votes