Timeouts, Tiebreakers, and The Lead

[Raijinken]

Hello everyone, Raijinken here with a new ruleset discussion.

This has been mentioned frequently in the Suicide Clause thread as of late, so I'd like to start an independent discussion.

Currently, in the event of a time-out, winner is determined by stock (a ruling the game applies itself), and then, for dislike of Sudden Death, by percent (before going to the potential tiebreaker round). However, as any Smash player is likely aware, percent is not a reliable indicator of who is in the lead. This is especially relevant in Smash4 with the presence of Rage, and the occasional odd mechanic like Little Mac's KO Punch.

Currently, a lot of people settle for the Percent Rule because it beats entrusting your tiebreaker to a torrent of bob-ombs. The purpose of this thread is to discuss potential options for an alternate tiebreaker.

I, personally, am interested in considering some of the Results Screen stats as potential sources of this tiebreaker.
The stats provided are as follows, in order:

• KOs (integer by victim)
• Falls (integer by source)
• Self-Destructs (integer)
• Damage Given (integer percent)
• Damage Taken (integer percent)
• Damage Recovered (integer percent)
• Peak Damage (integer percent)
• Launch Distance (integer feet)
• Ground Time (integer seconds)
• Air Time (integer seconds)
• Swim Time (integer seconds)
• Hit Percentage (integer percent)
• Ground Attacks (integer)
• Air Attacks (integer)
• Smash Attacks (integer)
• Grabs (integer)
• Throws (integer)
• Edge Grabs (integer)
• Projectiles (integer)
• Items Grabbed (integer)
• Max Launch Speed (integer mph)
• Max launcher speed (integer mph)
• Longest Drought (integer seconds)
• Transformation Time (integer seconds)
• Final Smashes (integer)

Highlighted in Red are stats that are irrelevant or inconsistent (i.e. matches exist during which these are impossible) with regards to competitive play.
Highlighted in Orange are stats that I, personally, do not reflect directly on the skill of the player.
Highlighted in Yellow are stats that are directly used when determining the winner of a match, and thus don't make sense (to me) as tiebreakers.
Highlighted in Green are stats I believe are directly skill-related and could be useful in breaking ties.
Self Destructs, as defined by the game, depend on the player having taken no damage from other players for the last three minutes, or since they spawned, whichever is shorter. This means you have to literally goof up really hard before you get hit. It also means that unless you spawn and hit the enemy with a suicide move before they can knick you for 1%, your "suicide" was actually their kill, as far as the game knows.

And now for the breakdown. All of these essentially assume (as is necessary when discussing time-outs) that the game is being deliberately drawn to time by one player or the other.
[collapse=Damage Given/Taken]
Damage Given and Taken don't differ very much from the current percent ruling. The key distinction is that this stat remembers all stocks, and Damage Taken includes things like bubble damage or stage hazards (when relevant).
Pro

• A dominant first stock makes it FAR harder for the opponent to stall and win the tiebreak thus, especially for characters like Villager or Sonic who can struggle to actually approach. They have to not only take the lead in the present stock, but also make up for the previous stock.

Cons

• Early kills via gimps contribute little to your tiebreaking value.
• Heavy/survivable characters with high kill power (Lucario, for instance) may take more damage, kill their opponent early via a strong hit, then finally lose their stock, and have little damage output to fall back on for the tiebreaker. Similar to gimp issues.
• Can result in different types of runaway play. A Fox who takes the stock lead may well run and shoot, racking damage in case he loses his stock, instead of attempting to close the stock out.
• Still relies on damage dealt, which again, isn't an accurate measure of lead.

[/collapse]

[collapse=Peak Damage]
Peak Damage directly corresponds to survival, and is essentially the inverse of your opponent's ability to end the stock. These pros and cons assume higher damage is the "winning" side in this tiebreaker.
Pros

• Direct measure of survival.
• Rewards early kill secures or gimps with tiebreaking power (enemy has lower max-damage).

Cons

• Not necessarily clean-cut on what's "better." To have survived longer, or to have taken less maximum damage to begin with? Could deserve some discussion.
• Games going to time can involve high damage on both players, overriding on any given stock any amount of "good performance" on another stock.
• Still relies, to an extent, on percent.
• Can be unfairly distributed due to character properties. Good recoveries/high weight/low kill power/etc influence this directly, without being related to player skill.

If lower peak damage is the "better" end, players are rewarded for taking less damage (perhaps playing more flighty and defensive), but are penalized for killing early (can create "play with your food" scenarios). If higher peak damage is the "better" end, players are rewarded for long-term survival, but this also encourages reckless play. Whether that's a good thing or a bad thing will depend entirely on who you ask.
[/collapse]

[collapse=Hit Percentage]
Attacks made divided by attacks landed. Multi-hitting moves, including multi-hit projectiles like Needle Storm, or regular multi-hitting moves like Mario Tornado or Marth's nair, count as a single hit, and are scored as a hit if any part of them deals damage. Hitting a shield does not count as a hit, even if the shield breaks. This includes killing Jigglypuff with a shield break.
Pros

• Direct measure of player accuracy.

Cons

• Discourages "throwing out hitboxes" and projectile walling.
• Consequently penalizes characters who depend heavily on these strategies, including Villager, Samus, and some playstyles of other characters.

Generally, much of this measure's validity depends on one's opinion. Projectile spam and repeat use of moves like Spindash are generally characteristic of playing to stall until time. This measure directly harms those strategies, which is not necessarily a fair way to evaluate styles of play. In a similar vein, this can put a player who wants to time-out on a peculiar kind of punish-based defense, where they play flighty (but not, per se, spammy) until they can get a punish. Since missing any move directly weakens your tiebreaking power, it can become similarly risky for the "winning" player to finish the stock, as one must risk tiebreaking power while trying to kill the opponent, and that kill is not necessarily guaranteed.
[/collapse]

[collapse=Longest Drought]
This measures the maximum time a player went without attacking. Shielding, jumping, taunting, rolling, and spot-dodging do not reset the timer. Balloon Trip (as well as its exploding cousin) do not reset the timer, but Flip Kick () does even if you don't use the second kick part. Grounded Spring (no hitbox) does not reset the timer, but aerial does.
Pros

• Forces certain types of stalling to approach. Villager-style planking will not increase the drought, while his opponent simply standing on the other end of the stage builds up power, forcing Villager to give up some security in order to build tiebreaking power.

Cons

• Could be disproportionately effective on Sonic, who is already good at stalling to time. Run away, stand still a moment, wait for approach, repeat.
• Standing skill is very hard to call a "skill".

[/collapse]

Given a look at all of these, I don't think any one of them gives us a clear solution, as each has drawbacks no matter how you look at them. If I had to pick one, I would go for accuracy, simply because it directly penalizes the two most-dominant methods of safely stalling to time.

A combination of some of these could work. Or none of them could work. A TO may find doing some weird math combination of these measures to be too troublesome.

Well, those are my thoughts. What are yours?

 

[ぱみゅ]

Proved time and time again, Percentage leads is far from optimal but is the less-worse of the available options.


Once someone told me that "Percentage does not indicate who is closer to die, but it clearly shows who is farther from the starting point (0%)", and to this day is the best comment I've heard about it.

 

[Kaladin]

Ideally, it would be decided on who had the momentum (i. e. actually winning). As an extreme example, remember Angel v. Abadango at EVO? Remember how it timed out as angel got the kill? That would count as angel winning/having momentum. If we could have someone objective to decide if the game was in advantage/disadvantage for one player, that could be the deciding factor. If the game was in neutral, with no player having momentum, eh just go with %.

In theory, if we could get objective arbitrators (three of them?) for the important rounds (top 8 at nationals) to decide who had advantage/momentum, that would be the best system. But there's too much subjectivity and time issues there...

Eh. There's really no good way to do this, but I think looking at the game state is the right direction to go.

 

[TheReflexWonder]

If I had to pick a new win condition on time-out, it would be Longest Drought. At least it would encourage players to be active, which is more interesting to watch and generally reduces the likelihood of time-outs, I think. The other points you have listed have cons that I think are rather unhealthy for competitive play.

With Damage Given/Taken, it skews the effectiveness of low-percent KOs and gimps signficantly. I know that if I SD'd or something similar, I would go out of my way to play to the time-out, feeling like that's likely an easier gap to overcome. Characters like Fox may choose to win via time-out by using their Blaster, since they only need to make sure they have the same amount of stocks as the opponent at the end. That would likely be really detrimental to some matchups.

Peak Damage might encourage self-damaging strategies to inflate that number, and it would be skewed in the direction of fatties or people who can live longer. This would've been awful in Brawl, where Snake could tech off the stage via grenades and C4 to rack up percents, and multiple characters had ways to totally stop momentum, like an early Z-Air ledge grab, G&W's Bucket Brake, etc. Things may not be so drastic in Smash 4, but we can't discount what's there.

Hit percentage implies that hitting the opponent with a higher percent of attacks is somehow more skilled than a player using missed attacks as a wall or misdirection. Again, skewed against certain characters, and I know that this would guarantee that I would lose every match I ever went to time over.

 

[Raijinken]

If I had to pick a new win condition on time-out, it would be Longest Drought. At least it would encourage players to be active, which is more interesting to watch and generally reduces the likelihood of time-outs, I think. The other points you have listed have cons that I think are rather unhealthy for competitive play.

With Damage Given/Taken, it skews the effectiveness of low-percent KOs and gimps signficantly. I know that if I SD'd or something similar, I would go out of my way to play to the time-out, feeling like that's likely an easier gap to overcome. Characters like Fox may choose to win via time-out by using their Blaster, since they only need to make sure they have the same amount of stocks as the opponent at the end. That would likely be really detrimental to some matchups.

Peak Damage might encourage self-damaging strategies to inflate that number, and it would be skewed in the direction of fatties or people who can live longer. This would've been awful in Brawl, where Snake could tech off the stage via grenades and C4 to rack up percents, and multiple characters had ways to totally stop momentum, like an early Z-Air ledge grab, G&W's Bucket Brake, etc. Things may not be so drastic in Smash 4, but we can't discount what's there.

Hit percentage implies that hitting the opponent with a higher percent of attacks is somehow more skilled than a player using missed attacks as a wall or misdirection. Again, skewed against certain characters, and I know that this would guarantee that I would lose every match I ever went to time over.

Would you use Longest Drought being higher or lower as the winning condition? The measure doesn't go fractional (lowest is 0 seconds though), is not cumulative, and is only actually based on the time between attacks (doesn't matter if they land or not). In essence, it's the opposite of accuracy as a measure. If a player goes an entire match without ceasing to attack, they'll have the lowest time, so if a Sonic keeps using his Spin Dashes and whatnot, the opponent would actually be forced to continually use punishable moves to keep their drought short in case they fail the chase.

In the opposite case, a player would be potentially encouraged to escape from as much as possible WITHOUT attacking. This would hinder Sonic's strategy significantly, but interestingly, not even the exploding balloon trip variant counts as using an attack. It also looks like planting the sapling may count but watering air doesn't (need better testing on that). Taunting doesn't reset the counter, but using Flip Kick (Zamus) does, even if you don't follow it with the second kick. Grounded Spring doesn't reset, but aerial does, probably because of the hitbox.

 

[TheReflexWonder]

Sonic is super-duper awful, but I think that Longest Drought being lower is probably healthier for the vast majority of the cast. I'm not sure that there is a fair way to curb Sonic stupidity that doesn't make other matchups much dumber. Characters with great mobility options or rolls could still abuse it about as well.

Still of the mind that percent is the best indicator we could use, since it doesn't really cater to any particular playstyle.

 

[Kaladin]

Ooo, I just had an idea.

Sonic is super-duper awful, but I think that Longest Drought being lower is probably healthier for the vast majority of the cast. I'm not sure that there is a fair way to curb Sonic stupidity that doesn't make other matchups much dumber. Characters with great mobility options or rolls could still abuse it about as well.

Still of the mind that percent is the best indicator we could use, since it doesn't really cater to any particular playstyle.

Yes it does. It buffs sheik/pika/projectile wall significantly.

Anyways...

Momentum really is most important in smash. Both % and rematch fail to recognize this. Hmmm.... The more I think about it, in really important events' top 8, I do think a judge system would be best. 1-3 people simply decide which character has the advantage (game-state wise). If it's neutral, we go to %, because if you're at kill % in neutral, that's a disadvantage in and of itself.

 

[TheReflexWonder]

It buffs people who are capable of hitting their opponent, which is really the objective we're going for, isn't it?

The problem with a judge system is that there will inevitably be biases in the system. Maybe a particular player is well-liked, or the character/playstyle is not one that is as exciting to watch. I don't think it's fair to judge based on subjective preferences.

 

[Raijinken]

It buffs people who are capable of hitting their opponent, which is really the objective we're going for, isn't it

The problem with a judge system is that there will inevitably be biases in the system. Maybe a particular player is well-liked, or the character/playstyle is not one that is as exciting to watch. I don't think it's fair to judge based on subjective preferences.

I would decline to judge a Sonic ditto.

It'd be interesting to hear what the "Whoever was most active" system reads (the one used in Tournament Mode). Could lead to some interesting theory.

 

[Kaladin]

It buffs people who are capable of hitting their opponent, which is really the objective we're going for, isn't it?

The problem with a judge system is that there will inevitably be biases in the system. Maybe a particular player is well-liked, or the character/playstyle is not one that is as exciting to watch. I don't think it's fair to judge based on subjective preferences.

No, it rewards characters who are excellent at racking damage, but balanced by their inability to remove stocks. It unbalances them.

Yeah, the judge system would be biased. That's the problem. But would it be biased enough to be unable to decide whether or not the game is in neutral?

 

[deepseadiva]

It should be the formula: Damage Given - Damage Taken - Peak Damage x Hit Percentage.

This would give the most accurate representation of who was succeeding the most in a match when it is clearly demonstrable that the one in a negative position was forced out of neutral for the remainder of the I'm joking this is a joke.

 

[ぱみゅ]

Judges is a terrible idea.

For starters there is a HUGE logistics issue.
Then we'd have the same problem boxing has: camping will still exist, players will potentially know when they have appeal to the judges.
And more importantly, there WILL be polemic decisions favoring a player that according the public/experts, did not deserve it.
So, yeah.

 

[Kaladin]

Judges is a terrible idea.

For starters there is a HUGE logistics issue.
Then we'd have the same problem boxing has: camping will still exist, players will potentially know when they have appeal to the judges.
And more importantly, there WILL be polemic decisions favoring a player that according the public/experts, did not deserve it.
So, yeah.

The 'judges' would only decide if the game is in neutral or not. If Cap. Awesum just hit you offstage with a stray fair, it's not neutral anymore. I don't think the smash community is THAT corrupt...

 

[TheReflexWonder]

It'd be interesting to hear what the "Whoever was most active" system reads (the one used in Tournament Mode). Could lead to some interesting theory.

I've heard people say that, but is that universal in Tourney Mode? This video by Cosmo seems to show that it just does percent lead...

 

[Kaladin]

I've heard people say that, but is that universal in Tourney Mode? This video by Cosmo seems to show that it just does percent lead...

I think it opporates on damage given? That's what Keitaro seems to think, at any rate...

 

[Balgorxz]

I like damage dealt more than current % because it hurts campy characters more.
also not everyone is 100% sure at what % one died, it's even better in 3 stocks.

 

[TheReflexWonder]

The 'judges' would only decide if the game is in neutral or not. If Cap. Awesum just hit you offstage with a stray fair, it's not neutral anymore. I don't think the smash community is THAT corrupt...

What does it mean to be in neutral, though? You can't easily tell who is in control of the match as a whole a lot of the time.

If five minutes and thirty seconds go by and I get my opponent to 100% on their second stock while only taking 30% on my first stock, but get gimped and take 50% in the last 30 seconds, who would you say deserves to win? What if I got to 70%? What if I SD'd instead of getting gimped?

There's no good way to tell who deserves to win based on a subjective judging system because "who was outplayed over the course of six minutes" leaves way too many questions. The only truly fair judge is through the game's mechanics because it cannot be argued against.

 

[Pazzo.]

Don't we hold to percent leads because of traditional fighting games?

Whoever has the most life at the end wins? Because if this is true, we may consider seriously looking at one of our other actions. Really, if we were able to find a nice balance between the current system and 'Most Damage given', we'd both discourage camping and still promote aggression.
But then again, stalling/projectile walling is not a 'wrong' way to play. All our options, are as previously mentioned, slightly less optimal than the current system.

Perhaps an analysis of optimised Smash play may help? Doubtful, as character play style varies more than in, for example, Street Fighter.

Of course, running the clock is a SF strategy, but weaving in and out of safe and dangerous zones, a la Sonic, is not found exactly in SF. Even SF-like characters, such as heavies, will have another whole world to worry about when their opponent is a character like Yoshi or Mario.

And just like that, back to square one, and post #2 off this entire thread...

Proved time and time again, Percentage leads is far from optimal but is the less-worse of the available options.

Once someone told me that "Percentage does not indicate who is closer to die, but it clearly shows who is farther from the starting point (0%)", and to this day is the best comment I've heard about it.

 

[TheReflexWonder]

To be fair, Vega's walk speed is in its own tier of goodness, and he can use his Flying Barcelona Attack to make sure he never gets cornered too hard. Pretty character-dependent, but it's kind of there.

 

[Pyr]

I like damage dealt more than current % because it hurts campy characters more.

"Because it hurts campy characters more" is a horrible justification. Eventually people will have to accept the fact that defensive, careful play isn't a bad thing

...but weaving in and out of safe and dangerous zones, a la Sonic, is not found exactly in SF.

Kinda nitpicking a bit, but they do have it. They call it "footsies." That back and forth motion isn't for show.

 

[Wintropy]

I'm reminded of a quote from Churchill in response to the crux of this thread:

"Capitalism is the worst economic system, until you look at the others."

The percentage-based tiebreaker is, as other have duly pointed out, imperfect. It is nevertheless the most efficient, and debatably the closest to workable, system we have. The other important tidbit to recall is that people are used to it, plus it's common practice in most fighting games, so I imagine it will stick around for some time.

Rather than get into the fine details of whether the system itself is ideal or should be reconsidered (everybody else has done a fine job and said it more succinctly than I ever could), I wish to approach this topic from a different angle:

Does it need to change, and would the potential benefits of enforcing a change outweigh the difficulty of convincing the community-at-large to accept it?

I don't know if it needs to change, so I won't comment on that (yet I will say I have no extraneous issue with it as a competitor or viewer). With regards to whether it would yield any net benefit...I don't know if it would. We know too well how difficult it is to agree on anything in the competitive Smash community, and for valid reason (whether it's a fine reason is another matter entirely): Smash, contrary to most fighting games, doesn't have the virtue of its ruleset being predefined by the nature of the game itself. In most fighting games, the rules boil down to "two fellows fight, whoever gets knocked out first is out". As a necessary contingency to this rule, in the event of a time-out, whomever has the most vitality left (or, to put it another way, whomever has accrued the least damage) is declared the winner.

With Smash, the issue becomes muddier. As you and I have discussed in your other thread on the benefits of a For Glory-style ruleset, the competitive Smash community is essentially built on the fundamental principle (whether explicitly stated or merely a de-facto ethic): what the game says =/= how we do it. This, to me, is the core issue at the center of every semantic debate in competitive Smash, from stage selection to customs: we don't have a clear, de jure ruleset to refer to. We have to go against the grain and make our own rules, because we have determined that the rules of the game just don't work with the kind of competition we want to play.

Before I end up going off on a wild tangent here, I'll draw back to my initial thesis: for what we have here, with this current system, we have a game that isn't broken and which I question if we really need to fix. I don't doubt that there is every realistic possibility that, if we were to hypothetically redefine the rule as it currently is, it could yield tangible benefits; hell, for all I know, it could restructure the entire foundation of our competitive ruleset with good results! My concern is that the effort to yield this kind of change would cause turn out to be more detrimental than beneficial. The theory is sound, there's every reason to suggest it would be worth experimenting with - I just worry that the technical demands of this kind of socio-competitive experiment would turn out to be bitter fruit. Upsetting the apple cart is good if it's done with a clear intention to fix the cart itself, but if you're doing it just to tip the vehicle and see what happens, you've got a very sticky mess on your hands and a bushel of apples you need to deal with.

Maybe I'm just being Cassandraic about this. I don't want to come off as a conservative fuddy-duddy with reactionary tendencies, I just wonder if this is something that, right now, when there are other real issues that have yet to be resolved in our community, this kind of pseudo-problem is really worth the time and effort it would take to redefine the system.

VERY IMPORTANT MASSIVE DISCLAIMER:​

I am absolutely and 100% playing devil's advocate here for the sake of debate. There's every possibility that I could have just missed the point of the thread by circumventing the question and undermining the very intent of your thesis (i.e. "would this work / should we consider this?" rather than "this is an issue we need to resolve immediately okay?"). Just posting a few notions to help keep the discussion dynamic. You know I love your meta-critical ruleset-consideration threads~

 

[ParanoidDrone]

I would decline to judge a Sonic ditto.

It'd be interesting to hear what the "Whoever was most active" system reads (the one used in Tournament Mode). Could lead to some interesting theory.

IIRC it's total damage dealt. In the event of tied stocks, whoever dealt more total damage to the other is declared the winner. (IDK what happens if that is tied also.)

 

[ぱみゅ]

Just for the sake of it, I'll say that there MUST be a formula involving damage dealt, total of attacks performed, hit percentage, and peak damage that helps decide who has been more active throughout the game, but coming up with it might be a hassle and if it doesn't convince anyone, it might not even worth the pain.

 

[Ghostbone]

% is easy to keep track of for players though, plus there's no risk of accidentally skipping the results screen.

It's more hype for the crowd too, if a game's going to time and nobody knows who's winning then it's a lot harder to see meaning in any of the final hits.

 

[1FC0]

Health based tiebreakers work in other games because you can see how much percent of the damage needed to get KOed someone has. If we could approximate this in Smash then we could have a pretty good tiebreaker.

That is why I think that (current damage) / (Peak damage) is a good one. The peak damage gives an indication of how survivable you are, and dividing that by the current damage gives an indication of how much percent of the damage that you need to get KOed you have.

For example if you play a 2 stock game and you and your opponent are both on your second stock and you have 80% damage but previous stock you survived until 160% damage and your opponent has 70% damage but survived until 100% damage, then it may be probable that you have around 80/160=1/2=50% of the damage you need to get KOed while for your opponent it may be probably that he has around 70/100=7/10=70% of the damage he needs to get KOed. In that case you should win.

A few disadvantages are that it requires calculation and in a match with more than 2 stocks your survivability is best measured by taking the avarage peak damage per stock instead of the peak damage. Also if the match ends without anyone losing a stock then this tie breaker results in a tie itself.

 

[Baby_Sneak]

wouldn't playing 1 stock 3 mins after tieing with %-based leads then added on be effective enough?

 

[TheReflexWonder]

How is that any more effective than the current state? The same win condition is potentially called into question in the current rules ("Is Bowser at 80% really losing to a Fox at 70%?").

 

[Pyr]

It'd give people losing the option to basically avoid gameplay to get a new stock and a better chance to win. I mean, who wouldn't want all their moves unstailed in a new match?

 

[Baby_Sneak]

It'd give people losing the option to basically avoid gameplay to get a new stock and a better chance to win. I mean, who wouldn't want all their moves unstailed in a new match?

if they were losing in the last match, then they're going to lose in the tie breaker unless they adapted.

How is that any more effective than the current state? The same win condition is potentially called into question in the current rules ("Is Bowser at 80% really losing to a Fox at 70%?").

becausein the actual matches, %-based wins aren't taken into account.

 

[Raijinken]

How is that any more effective than the current state? The same win condition is potentially called into question in the current rules ("Is Bowser at 80% really losing to a Fox at 70%?").

If only there was some objective measure of survivability based on weight, jump height/count, floatiness, recovery specials, etc. Then we could use that as a modifier on percentage to actually determine things.

But that doesn't exist, to my knowledge.

 

[Pyr]

if they were losing in the last match, then they're going to lose in the tie breaker unless they adapted.

It isn't that black and white, though. I mean, Evo, grand finals for Street fighter, you could just feel the momentum get killed after the pause. Having to do another 3 minute match fresh? It's a momentum killer. Also do keep in mind that, in addition to a time out, this adds, up to, 3 minutes to a single game. Huge increase in the worst case scenario time for TOs.

Really, what we have currently does work and isn't broken, at least it isn't compared to all the other possible ideas. (Hit percentage? So I get punished for walling with B-Air, spacing, and playing footsies? Come on, people...) No need to fix a system that isn't broken. We don't have to try and be unique special snowflakes with this game.

 

[Raijinken]

It isn't that black and white, though. I mean, Evo, grand finals for Street fighter, you could just feel the momentum get killed after the pause. Having to do another 3 minute match fresh? It's a momentum killer. Also do keep in mind that, in addition to a time out, this adds, up to, 3 minutes to a single game. Huge increase in the worst case scenario time for TOs.

Really, what we have currently does work and isn't broken, at least it isn't compared to all the other possible ideas. (Hit percentage? So I get punished for walling with B-Air, spacing, and playing footsies? Come on, people...) No need to fix a system that isn't broken. We don't have to try and be unique special snowflakes with this game.

It's less a matter of trying to fix what isn't broken, and more of searching for gold that might not be there.

Your mileage may vary, but I'd take any amount of convolution if it meant having an objective way to rule by who had a real lead. That's why I'm interested in searching (plus I don't have that much better to do in Smash while Schrodinger's Customs are still in the box).

 

[Ghostbone]

Your mileage may vary, but I'd take any amount of convolution if it meant having an objective way to rule by who had a real lead. That's why I'm interested in searching (plus I don't have that much better to do in Smash while Schrodinger's Customs are still in the box).

Some extremely convoluted "objective" rule is still worse than the % rule. The % rule is simple to understand, easy to for players and spectators to see who's winning at any time, more hype as the time runs out as the lead goes back and forth, no risk of players accidentally skipping the results screen, more intuitive than any other rule, etc.

Rulesets should favour simplicity as much as possible.

 

[Raijinken]

Some extremely convoluted "objective" rule is still worse than the % rule. The % rule is simple to understand, easy to for players and spectators to see who's winning at any time, more hype as the time runs out as the lead goes back and forth, no risk of players accidentally skipping the results screen, more intuitive than any other rule, etc.

Rulesets should favour simplicity as much as possible.

To the first part, the entire point of this discussion was that you can't really judge who is winning by percent.

To the second, the very fact that Smash's competitive playerbase spits on For Glory and the game's actual tiebreaking system (both very simple rules) suggests that even if the statement is true, the community doesn't really believe it in their hearts.

I agree there has to be a level of practicality to a rule, but I still feel like the rule measuring what it's intended to measure should take priority. Of course, as we've generally discussed, none of the results screen stats really seem to offer that.

 

[Ghostbone]

To the first part, the entire point of this discussion was that you can't really judge who is winning by percent.

You can if the % rule is the one being used.
We're talking about timeouts specifically here, with the % rule it's easy to know who's in the lead.

To the second, the very fact that Smash's competitive playerbase spits on For Glory and the game's actual tiebreaking system (both very simple rules) suggests that even if the statement is true, the community doesn't really believe it in their hearts.

I agree there has to be a level of practicality to a rule, but I still feel like the rule measuring what it's intended to measure should take priority. Of course, as we've generally discussed, none of the results screen stats really seem to offer that.

I don't think you understood.
We don't play sudden death because it's uncompetitive and encourages the losing player to go for a time out. Sure ignoring sudden death is more complicated then just going with it, but we do have to make the ruleset more complicated to cover uncompetitive aspects of the game.
But we shouldn't needlessly make the ruleset convoluted instead of using something simple like a % rule.

I wasn't saying we should have as simple of a ruleset as possible (that would just be turning on the game and going to a random stage with 2 min item matches), I'm saying that we should generally strive for the simplest competitive ruleset.

 

[Raijinken]

You can if the % rule is the one being used.
We're talking about timeouts specifically here, with the % rule it's easy to know who's in the lead.

I don't think you understood.
We don't play sudden death because it's uncompetitive and encourages the losing player to go for a time out. Sure ignoring sudden death is more complicated then just going with it, but we do have to make the ruleset more complicated to cover uncompetitive aspects of the game.
But we shouldn't needlessly make the ruleset convoluted instead of using something simple like a % rule.

I wasn't saying we should have as simple of a ruleset as possible (that would just be turning on the game and going to a random stage with 2 min item matches), I'm saying that we should generally strive for the simplest competitive ruleset.

The rule is arbitrary, though. It's used for the sake of giving a rough idea who has hit their opponent with higher-value attacks a few times, or low-value attacks a lot, without killing them. Percent is only even close to an objective measure of "the lead" in Stamina Mode.

It basically amounts to "Percentage doesn't indicate who is going to win, until it becomes obvious that at least one player is running the clock. Then, and only then, does your percent lead (or deficit) matter."

Edit/Addition to avoid Double Posting:
My general assumption/knowledge is that the only thing seen as "uncompetitive" about Sudden Death is that it's raining randomly-spawned explosives past a few seconds. Someone correct me if I'm wrong on that point, because the rest of this suggestion depends heavily on that being the only problem with running Sudden Death.

Smash has an option that readily enables us to simulate the high-volatility of Sudden Death without bob-ombs. This can be done via Handicap. We can literally run a 1 stock handicap match (doesn't even take setup besides clicking the stock down and the handicap up) with the same or a reduced timer, and avoid everything that makes Sudden Death uncompetitive. And still gain match closure through stock-out.

To encourage players to still engage when a match is being taken to time, we could alternately start the Sudden Death at 125-150% (kill threshold for most characters) and then add damage approximating what the player ended the match with. Unfortunately, Smash4's handicap levels aren't very fine-grained, so we're limited to 10, 20, 30, 40, 50, 60, 80, 100, 125, 150, 200, and 300% intervals. We could round up, round down, whichever. This lets people keep their percent advantage, but doesn't default us so high that heavies lose their weight advantage unless they'd already taken most of it.

Thoughts on that?

 

[Touchebag]

My general assumption/knowledge is that the only thing seen as "uncompetitive" about Sudden Death is that it's raining randomly-spawned explosives past a few seconds. Someone correct me if I'm wrong on that point, because the rest of this suggestion depends heavily on that being the only problem with running Sudden Death.

I would say the main problem is that it heavily favours characters with fast/ranged attacks such as Link and his bow. Since basically every attack will kill, slow, strong characters like Ganondorf have basically lost before the SD even begins.

Smash has an option that readily enables us to simulate the high-volatility of Sudden Death without bob-ombs. This can be done via Handicap. We can literally run a 1 stock handicap match (doesn't even take setup besides clicking the stock down and the handicap up) with the same or a reduced timer, and avoid everything that makes Sudden Death uncompetitive. And still gain match closure through stock-out.

To encourage players to still engage when a match is being taken to time, we could alternately start the Sudden Death at 125-150% (kill threshold for most characters) and then add damage approximating what the player ended the match with. Unfortunately, Smash4's handicap levels aren't very fine-grained, so we're limited to 10, 20, 30, 40, 50, 60, 80, 100, 125, 150, 200, and 300% intervals. We could round up, round down, whichever. This lets people keep their percent advantage, but doesn't default us so high that heavies lose their weight advantage unless they'd already taken most of it.

If we ignore the fact that this takes more time (TOs hate everything adding to game time) I really like this idea. It's basically like overtime in many sports.

However, this begs the question of why we have the time in the first place. If we're just gonna continue a timed out game where it left off why not just turn off the time completely? The time limit is mostly there to ensure tournament matches take too long. This would essentially make time meaningless.

 

[GhostUrsa]

Personally, I don't like the % victory condition because it's skewed heavily in favor of light characters. A light character could be in kill % to a heavy by as low as 50%, but that heavy would need 150% to be in the same league. These rules favor the light, are decent for the middle weights and are slanted against the heavy weights. This has always rubbed me the wrong way, even before I started playing a heavier character.

I've always clamored that Smash should get it's idea for tiebreakers more from physical sports than other eSports due to it's unique structure. (I've described it as the MMA of fighting games!) I personally have been playing with the idea of a rematch first, which if that also comes up tied we call a draw here. In single elimination tournaments, both get the second place prize and be done with it. In double elimination, they would fight the loser bracket winner as the determining factor of 1st/2nd/3rd. (A loses to B, but C beats B so B is 1st, then C, then A) if in this part the two from the winner's bracket tie again then it's a draw and the single bracket rule applies. TO's could call for a double tiebreaker (think double overtime in normal sports) but that would be something for an organization to determine.

Yeah, it would be a little more complicated than going with the % ruling, but the current % ruling throws away fairness to all weight classes for the simplicity that other knockout based fighting games have.

 

[TheReflexWonder]

Heavier characters generally KO opponents at earlier percents, though, so the primary win condition of "KO the opponent" is, in a way, skewed to benefit them, especially with Rage.

 

[GhostUrsa]

@ TheReflexWonder Which actually makes my point that % isn't the best option more on point. It adds another wrench into the works that determine how the match would have went had there been another 15-30 seconds. (Which is what the TO would be trying to boil down in the event of a tie.)

Just for spitball math with what we can easily track, if 50% at for KO against a light is roughly 150% for the KO against the Heavy we could say there is a ratio of % quality of roughly 1/3 health % value for the heavy. A heavy at 60% would roughly be a similar value in how easy to be to KO as a 20% light using that spitball. Using only % as the measurement, how can we say the light at 21 is the victor compared to a heavy at 30% when that heavy would be at 10% of the light's equal health? (Which makes it so that the heavy would need to be juggled for another 5 hit combo to be in the equivalent bad shape.)

Using a single metric that is only one piece of many variables that actually determine a winner seems like a poor way to determine a victor. It would be like determining a victor of a boxing match that went to Decision by looking at each fighter's BMI rating. It's a boggling way to determine a solution. (I'll admit I'm a little frazzled about this because I study data generation for a living, and being able to accurately determine a Pass/Fail state is something I do all the time. Credibility comes from accurately looking at everything to make such a decision, and doing so quickly.)

Despite what we argue here, I'm about 90% certain this won't change in our career's life time. This was argued back when Melee competition was still a grass roots movement, and will still be argued when Smash 6 comes out. It will most likely stay that way until we have a National/Global League to participate in (And thus a standardized set of rules.), like most organized sports do. Currently we just have 3 big tournaments feeling around with some legacy rules' tweaks to see what works for themselves and smaller tournaments following suite just for simplicity's sake and not everyone coming together under one banner.