Stock : Time Discussion

[Deleted member 245254]

As we all know there is heated discussion in most every corner of the Smash community right now on what the tournament standard for Smash 4 should be given what we've all experienced and for those who haven't, what they've seen.

With a few issues it's easy to hold off on the relegation that we need more information to make an educated decision but it's difficult to apply the same rule to something like this. We're talking about the tournament standard, the standard for which all tournaments bearing Smash 4 as their lead title will run by. The standard for which all players will practice around. Tournaments are imminent nearly immediately, some have already taken place. It wouldn't be an exaggeration to say that time is not quite in extensive amounts when it comes to deciding what would be best in this category so that we can set the right tone for all future tournaments to follow suit.

I see two main arguments.

- 2 stocks, 5 minutes (For Glory standard)

- 3 stocks, 8 minutes (Brawl standard)

Among these two are several modified ideas like 3 stocks 6 minutes and things like that but I find that most tend to go for the above two.

The main reason this discussion has become important is because of the well known (slightly infamous, still) increase in the width of the surviving area of the stages. Without proper gimping or good off-stage play it is quite common to see %s nearing 200.

Players for 2 stock rule sets propose that this directly suggests that 2 stock is superior. Given high % deaths, there is flexibility still for comebacks and downloads. I've seen arguments that propose that aligning our tournament standard with Nintendo's implemented standard could be good for improving the relationship further with Nintendo, which could possibly mean that the Smash tournament scene ends up benefitting (either through publicity or otherwise). Other reasons include lower potential match time.

Players for 3 stock rulesets argue that 2 stocks is simply too little to gauge a proper winner, by leaving little room for comebacks and inadvertently making SD's absolutely devastating to a match. They propose that there is too little time to adapt and that once players begin to "Play Smash 4" they will begin to have a more improved off-stage game and match duration will be a non-issue.

Personally, I believe two stocks is a bit too little, and that three would be perfect but I'm here to gauge all of your thoughts on this heated topic, before I spill mine own.

 

[Overswarm]

This is really a simple problem with a simple answer.

What you're trying to remove is variance. To reduce variance, you want a longer timer and more stocks.

99 stock with infinite time has the least variance possible.

1 stock with 1 minute timer has the most variance possible.

You want to slide as faaaaaar to the right as possible on these things, given the amount of time available. Given our setup, you also want to be consistent across tournaments. You might be able to do 8 stock matches at your local 12 man tournament, but a 70+ can't.

Given that I played for a whopping one day with the japanese 3DS and less than a week with the demo and I have gotten multiple gimps (some at 0% from one throw!), I can safely say that having less than 3 stocks would make it near impossible to have good results.

2 stock 5 minute being the standard on "for glory" doesn't determine our tournament format any more than Sakurai's timed item-fests in Brawl determined our format.

I'm going to be trying 4 stock at my tournament on Oct. 11th and will be recording the time spent for each individual match. In addition to this, I'll be recording the amount of time spent on friendlies as I play the week going up to teh tournament. Once we get some data behind this, we'll be able to see what is viable.

The variance thing is pretty big -- I've taken the first stock off of Mew2King in Brawl before. That would result in a game win. I was not as good as Mew2King.

Keep in mind Little Mac exists. Can you imagine Little Mac killing you with an f-smash at 90% off of one good read and then having his OHKO punch on your second stock? Talk about variance.

Lucario, as well, is unfairly nerfed by having lower stock.


To reiterate:

We want the most stock and most time possible. I'm starting at 4 stock and will be having a 12 minute timer (3 min per stock) as this is above what most would consider viable, and we can record data and work our way down from there.

 

[Neo Zero]

A lot of it can be simply be attributed to how well we (don't) know the game right now.The for glory standard can even seem long right now, but who's to say how it'll feel in 3 months? Besides that, it's also worth remembering things such as SD's, if you SD in the For Glory format, you've basically already lost. Little Mac and Lucario are also good examples of the both extremes. One character excels and becomes more dangerous, the other looses a bit about how he can play and becomes worse.

At the stage where we're at currently, I'd also argue 4 stocks might be to much, but again this is something that needs to be done for data purposes (awesome job btw OS) to get a better understanding if it feasibly works. Right now, to me, the Brawl standard would seem to be the best in terms of how long a game (or even a stock) can last while still not going overboard, but it could be a very short matter of time before we could be seeing 4 stocks with similar time limits. Between the options given in the poll, I'd be inclined to pick the 3s:8m route myself, but how long that would last would entirely depend on how we start seeing games go. Will they continue to last a long while and be fundamentally slow, or will they perhaps be doing faster.

Worth noting as well in the For Glory argument, we already have an infinite with ZSS discovered. It'd take two clean hits to end you alone, and while I do hope it's patched out before anything major comes of it, the fact it exists, as well as examples like Little Mac just makes it seem not really feasible to have as low as 2 stocks.

 

[Hylian]

The infinite only works on Robin.

As far as stocks go, tournaments are going to use a variety of rules so there will be plenty of data to look at and make an informed decision in the future rather than just guessing now.

 

[Zigsta]

I personally believe it'd be best to use 3 stocks, 8 minutes to begin with since those were Brawl's rules. TOs can look at how long their 3DS tournaments run compared to how long their respective Brawl tournaments ran for. In general it seems like matches are either done REALLY fast or take too long since people are still figuring out the ledge and KOing in general.

 

[Dabuz]

I prefer two stock because yes, we have matches which end in like a minute, there are also matches with character such as DHD who will bring it to time. Right now it feels brawl-esque (Personally I thought brawl would of been best with 2 stock) and if it ends up being too quick, I think adding another stock will be easier to do in the long run than removing a stock (and before you say it OS, I don't have empirical date to back this up -.-). But yeah, tournies + give it a while.

 

[Overswarm]

If you don't have empirical data to back it up, go back to the lab until you get some!

 

[False]

Pardon for the double post but I thought it would be appropriate for my last post in another thread to be posted here:

Hmmm, I think people are just trying to avoid the length of Brawl matches, hah.

It would just do well for general time constraints. Brawl tournaments generally ended late, and since 'we esports now' I suppose trying to cater to the viewership wouldn't be a bad idea either. Audience retention is at its best up until 5 minutes, so cutting down on the time might show us some improvements we haven't seen before.

For those who don't know, an east coast TO named Doom was trying to organize 1-stock events for Brawl during its dying days, in hopes that it would revive the scene. I heavily supported this because it was efficient, engaging, and was supposed to promote less camping. It also rewarded momentum more. I think we can see similar traits transferring over if we were to adopt a 2 stock ruleset. Not as fast as 1 stock, but not as painstakingly long as 3 stock.

(Also, in For Glory its 2 stocks, just saying.)

Some BRAWL 1 stock matches for you guys, to get some sort of idea: https://www.youtube.com/watch?v=bHnNSEsX-is

 

[Overswarm]

I have to reiterate that length of matches is not the end goal. Consistent tournament results where the higher skilled player is. It's why we don't play with items. Turning off items and then lowering the stock count arbitrarily is coming full circle!

If people are worried about tournaments ending late, just do all pools. It is so, so much faster.

 

[NickRiddle]

A couple of the matches I have played almost went to time, but that's because both my opponent and I are terrible at this game, as is everybody who is playing it.
Wait until people are good before you lower the stock count even more IMO.
Going to be running 3-stock/8-minutes, unless 33% of matches go close to time.
If that happens consistently, I would be an advocate of 2s/5m from a logistics point of view.

 

[Overswarm]

I'd say 33% of matches going to time over a period of time is a pretty good benchmark. It's an arbitrary benchmark, but one I think is pretty reasonable.

 

[NickRiddle]

I'd say 33% of matches going to time over a period of time is a pretty good benchmark. It's an arbitrary benchmark, but one I think is pretty reasonable.

Every benchmark is an arbitrary one.

 

[Overswarm]

Every benchmark is an arbitrary one.

Never say that to an analytics junkie. :B

You can use existing non-arbitrary data sets to determine the standard deviation from the mean (also known as σ) and use this as a non-arbitrary benchmark.

If we had data from previous tournament series, we could determine whether they were acceptable (Brawl 3 stock 8 minutes) and find the amount of timeouts in this admittedly acceptable format. We could then see how many timeouts we have in our current smash 4 format. If it's more than twice the standard deviation, then we've got a statistical spotlight on the amount of timeouts we have!

In Brawl, timeouts typically were around 1-2% at the end of its lifecycle in all the tournaments I collected data for. Typically those timeouts were heavily tilted towards certain players (NOT characters or stages !!!), despite public outcry at Norfair and Wario and the like.

33% is a markedly high amount of timeouts; if we got anywhere CLOSE to that at the end of a game's lifecycle across the board, it'd be a giant flashing sign saying "CHANGE YOUR RULES". This giant gap between the standard timeouts (1-2%) and our arbitrary benchmark (33%) gives a LOT of leeway and is essentially and upper level cap.

If we hit 33% and maintain that over the first month or so in various tournaments, it's likely not "we're dumb" and more "change the rules". If we get close and/or see a downward trend in the amount of timeouts, it's something we just need to observe and potentially change down the road. If we hit around 1-2%, we can obviously do nothing.

The more you know: http://www.mathsisfun.com/data/standard-deviation.html


You could do the same thing with characters and win rates.

If we have a list of characters considered as "top tier" (let's say 6) and define the top tier characters by a agreed upon points system, we could look at the standard deviation to determine if a character is stronger than others within the same group. If you have the top 6 characters with win rates (%) of 48, 47, 55, 67, 69, 88, you can easily find the average is 62.333~%. The top 3 characters are "above average" within the group, the bottom 3 are below average. You can then find that the standard deviation is 15.616. This means that if a character is 15.616% higher than the previous character than that character is statistically relevant. If that character is 31.232% higher than the next character in line, he is an absolute nightmare.

You need a big enough data set to draw broad conclusions, but it's definitely non-arbitrary.

 

[Keitaro]

My Co-TO RJ is actually collecting data of matches for 3 stock 8 minutes currently. I'll see what information I can get from him regarding that. He says he already collected time lengths for 60 Smash 4 3DS matches. Obviously the time length of matches may vary over time, but it is still good data to see for now so hopefully I can get it.

So far in the last couple days of playing the game the matches seem to be longer than Brawl by a small margin. However I was always questionable about the length of Brawl's matches in the first place. 2 stock matches will allow the matches to go faster, the tournament to run faster, and it's simply less drawn out. My main issue with keeping 3 stocks is that I don't see good enough positive points for it compared to 2 stock sets.

 

[Overswarm]

Two stock sets grant a higher amount of variance, resulting in a worse tournament experience and a less precise measurement of skill.

To say "we should use two stock matches" is to say "any win that ever occurred in Brawl's 3 stock matches is invalid if at one point the winner had one stock to the opponent's two stocks". This happened all the time.

If you say "these matches are too long", why are you using an 8 minute timer. If 8 minutes is too long, you've set the wrong time.

 

[Sethlon]

We have two JP 3DSes here at our house in Frisco, got to play with and watch a bunch of matches with me/Denti/AeroLink/Bwett/Dakpo/Strong Bad/others. I have seen and played many 3 stock games where we've gone down to 1 minute left on the 8 min clock when the games ends, and thats without either player actively playing to go to time. It might be something that changes down the line when we get used to things, but 2 stocks is definitely my preferred format at the moment.

2 stocks 5-6 minutes also has the benefit of having generally faster sets, which would certainly be nice as far as finishing tournaments in a timely fashion is concerned. I dunno about you guys, but I'm not a fan of finishing things up at a hotel room/chibo's apartment/random person's house.

 

[Deleted member 245254]

Great thoughts guys.

So for me, I tend to see a lot of arguments for 2s based on the idea that it will encourage shorter matches out of default, I mean the max time is shorter so, it would make sense right?

With the large number of For Glory matches I have under my belt at the moment (relatively to anyone who doesn't have the game yet, of course), I can definitely say I've come across a few games that were very close to being decided but were clipped and taken to sudden death due to the timer, and these were particularly campy players.

Now, I am not an analytical type like Overswarm, but his variance explanation seems to be quite relevant here. The weight of the first stock is most certainly a bit too overbearing under 2s, and many characters have some really fantastic abilities to simply play it abnormally safe if they want to simply build %. I've seen needle spamming Sheiks, Fireball spamming Luigis, DHD in general, roll/Luma spamming Rosalinas, green > yellow stance dancing Shulks and some other variants of play styles that lead to some pretty frustrating experiences for the opponent, and losing a single stock before figuring out a good strategy to counter these types of rare play styles is common. In the For Glory mode, it's basically sealing your fate for a loss.

I definitely see 4s as being a bit too long (IE potential to time out many matches), for reference.

3s simply gives a more optimal amount of time to clearly demonstrate a better player, where as 2s has an exponentially higher rate of creating accidental upsets because something to note is that lowering stocks in Smash Bros can't be done based on the desire for lower match time, because the concept of stocks itself has a vacuum-like effect on the player. You play much differently on your last stock than you do on your first or second. This much is undeniable. Our goal should be to find the most optimal medium for the most skilled player to demonstrate their win and not for the match to end quicker for tournament TOs sake.

The time was set to 8 minutes presumably because that was a timer that was considered the most amount of reasonable time any one match could last before it would be counter-productive for tournament organizing and to a degree, brevity, while still allowing enough flexibility for a true and definitive winner to come out on top. Time is always of the essence in Smash tournaments. I can almost confidently say that I don't see 3s matches going to time under 8 minutes very often if even at all.

The presence of real follow up combos should entail that matches are going quicker than that of Brawl ones at a higher level, or at least I would presume.

 

[Overswarm]

I should also mention that the average time for the early Brawl tournaments had virtually every match going to 6 minutes -- people blamed lack of good combos, followups from grabs, and the fact that no one had any kill moves except for snake and everyone lived forever because their recoveries were "too good".

Most matches by the end of Brawl didn't typically take this long except at the highest level of play, where some common matchups were pretty campy by default. We'll need to be vigilant about how much data we record and not be like "I played some friendlies and..." or "there were SO many timeouts" when it was only 3 in a list of 60 games.

Taking smash down to two stock would be pretty bad, objectively speaking, so if it's going to be a possibility it needs to be done after careful research and planning.

@ Sethlon If you're tournaments are finishing at hotels or at Chibo's house, get better tournament formats! It should literally be impossible for tournaments to last that long unless you're severely lacking setups. Tell 'em to switch to pools only and just be more draconian when you have less setups. In the midwest we can end 60 man tournaments, singles and doubles, with daylight outside! If you have a chance to talk to AlphaZealot about running tournaments he can go into some serious bracket talk with you, he knows his stuff.

 

[Deleted member 245254]

Something pretty notable about the 3DS is that you can never have a shortage of setups, you're pretty much limited to how much space you have available to fit people (all of planet Earth?).

This alone would contribute to shorter tournament durations (temporarily), would it not?

 

[NickRiddle]

Man, you can play in people's cars with the AC on if need be.

 

[Overswarm]

Something pretty notable about the 3DS is that you can never have a shortage of setups, you're pretty much limited to how much space you have available to fit people (all of planet Earth?).

This alone would contribute to shorter tournament durations (temporarily), would it not?

As much as I'd absolutely despise moving to two stock (one stock would actually likely be better!), it's important to note that whatever standard we have for the 3DS will likely carry over to Wii U. If 3 stock works for the 3DS, Wii U will use 3 stock. If 2 stock works for 2DS, Wii U will use 2 stock. If it turns out that we're consistently going to sudden death in matches where this isn't intended, we can't use the 1:1 setup ratio as a crutch.

 

[Zigsta]

@ Overswarm , why do you say one stock would be better than two stock?

 

[Hylian]

@ Sethlon If you're tournaments are finishing at hotels or at Chibo's house, get better tournament formats! It should literally be impossible for tournaments to last that long unless you're severely lacking setups. Tell 'em to switch to pools only and just be more draconian when you have less setups. In the midwest we can end 60 man tournaments, singles and doubles, with daylight outside! If you have a chance to talk to AlphaZealot about running tournaments he can go into some serious bracket talk with you, he knows his stuff.

Sethlon lives very far away from chibo lmao, he was referring to CoT4.

 

[Deleted member 245254]

As much as I'd absolutely despise moving to two stock (one stock would actually likely be better!), it's important to note that whatever standard we have for the 3DS will likely carry over to Wii U. If 3 stock works for the 3DS, Wii U will use 3 stock. If 2 stock works for 2DS, Wii U will use 2 stock. If it turns out that we're consistently going to sudden death in matches where this isn't intended, we can't use the 1:1 setup ratio as a crutch.

I thought that what I said was more a boon to that of 3s than it was 2s (IE less worrying and time spent worrying about tournament duration/setup due to not worrying about systems thus more flexibility for longer matches).

 

[Overswarm]

@ Overswarm , why do you say one stock would be better than two stock?

Full explanation below.

TL;DR explanation: One stock has less of a negative impact than two stock, as two stock has two distinct "modes". The first is "take the opponent's stock" and the second is "I'm ahead, rack up percnetage". Given Smash 4's properties, it is incredibly difficult for someone in the first mode to compete effectively with those in the second mode. A superior player, over time, will come out on top. Two stock does not allow a player of marginal superiority (think a 65% favorite) to come back from a self-destruct, gimp, or one good read. Reducing it to one stock allows base sets to be best 3 out of 5, which allows players to lose a game via SD (the same likely outcome from 2 stock) and still be able to come back and win the set. Two stock does not.


The goal of a tournament is to find the best possible player. The best way to do this is to do a complete round robin where every player plays each other under the same ruleset. This is obviously not viable.

We typically define "best possible player" as someone who is able to beat his opponent in a 2 out of 3 series, where we have a "starter" and up to two "counterpicks". In later sessions, such as finals, we use best 3 of 5 or 5 of 7. The reason we increase the amount isn't for "more hype" but to reduce variance. The more games there are between any two players, the less likely that an inferior player would win.

This is why you may have heard "winning is not the acme of skill"; the phrase means that anyone can win one individual aspect of anything! It's why sudden death is disdained across the board. If we wanted faster tournaments, why not make them all sudden death? Because of variance!

In addition to doing 2 out of 3, we also use as many stocks as we can within that timeframe. 4 stocks was the norm in melee, 3 in Brawl.

As long as results are consistent and individual sets do not see variance then the overall effect of lowering stock is negligible -- you'll find changes with characters like Lucario and Little Mac due to special properties, but overall it will be minor.

We can pretty easily predict Smash 4's issues with 2 stock by looking at Brawl. It was incredibly common for a player and his opponent to both be at two stock and then for one player to lose his stock and then have a comeback and take both of the opponent's stocks.

It was also incredibly common for Self Destructs and gimps (including 0% kills!) to occur, even in high profile sets.

Despite the relatively common comebacks, self-destructs, and gimps in high profile Brawl sets, variance was overall fairly low. This was because you had a long enough period to make a comeback over time if you played better than your opponent. You do NOT have that luxury with two stock.

You don't have that luxury with one-stock either, but it allows you to increase from best 2/3 to best 3/5, meaning that an SD or gimp that would win the game ANYWAY in two stock now has less of an overall impact.


I'd hate either option.

 

[Krynxe]

@ Overswarm I don't seem to understand everything you say. The following two statements seem to be incredibly contradictory to me:

We can pretty easily predict Smash 4's issues with 2 stock by looking at Brawl. It was incredibly common for a player and his opponent to both be at two stock and then for one player to lose his stock and then have a comeback and take both of the opponent's stocks.

This was because you had a long enough period to make a comeback over time if you played better than your opponent. You do NOT have that luxury with two stock.

The way I see it, playing with two stocks is similar to playing a 2-out-of-3 match, similar to Street Fighter. Losing a stock is like losing a round, you are down but still have the opportunity to turn it around and make a comeback. I feel as though with Brawl, the third stock was often unnecessary, and was only there to serve as a way to prove to your opponent how superior you were when you could keep all three. Extra stocks were a necessity in the previous smash games, 64 and Melee, because of how quickly a stock could be taken from you by an opponent. Multiple stocks were, in those games, more part of a match as a single combo or interaction could be the difference in stocks. However as we're looking at newer versions of Smash, combat is centralized around a lot more small interactions rather than large combos and gimps. This is a general statement, and I know there are exceptions in Brawl, however it's undeniable that - outside of MK and ICs - gimps and huge, sudden leads were very rare to occur and instead matches would more resemble a battle of attrition. With the changed ledge mechanics and - what appears to be - significantly improved game balance, it would seem like the necessity for a high amount of stocks is reduced drastically.

Not to mention that for a series that's making it increasingly more difficult to kill yourself, we're now at a stage where even moves like rollout and illusion no longer put you into freefall. I see no reason why we should be worried about leniency of people killing themselves when it's so difficult to do so to begin with. You're making a mistake and should be punished, and even so, losing a stock in a 2 stock game is better than wasting a game in a 1 stock ruleset. This only isn't true if you believe that losing a stock accidentally means you automatically lose the match, which is far from true.

2s seems like a ruleset that's more fitted towards the gameplay style of Smash 4 as it looks right now. I believe the gameplay is slow and methodical enough to allow plenty of time to adapt and adjust in a set, especially given that there are multiple matches in a set. An attribute of a skilled player is being able to adapt quickly and efficiently, so I think it's a bit silly that so many people are trying to argue that 8 minutes a match is required for players to effectively do so. I've played melee competitively for many years and I am able to pick up on most of the important things I need to understand about my opponent after a single stock. In brawl, from my experience, it takes less than that. Simple interactions in the neutral game after a bit of time will allow you to adjust and understand how your opponent plays. To again compare to Street Fighter, the first stock is much like the first round. Both players are feeling each other out and learning from it, but the player who wins the first round/stock is the player who can adapt and play accordingly most efficiently. Having to focus on learning and playing simultaneously is what makes playing new opponents exciting for all parties, but watching two players who are familiar with one another go through the same motions from the very start only makes a ruleset like Brawl's that much more off-putting. (see: most top level MK dittos)

Not only that, but it allows for more efficient tournament time, which is a big issue we've seen in Brawl's time as a competitive game. Honestly, I believe people are just sticking to 3s 8min ruleset because they don't want to deviate from Brawl and what they are comfortable with. It's important that we try to look at the benefits and loses of each and weigh them accordingly. It's for that reason, that the benefits seem to outweigh the loses by a significant amount, that I support the 2 stock ruleset.

 

[Sethlon]

@ Sethlon If you're tournaments are finishing at hotels or at Chibo's house, get better tournament formats!

You mean, like, changing the stock number?

A 60 person singles/doubles bracket happening in one day without going over time is standard. We get that number of people for some of our weeklies for PM here in DFW (which start at around 8pm and generally end at 1-2am). Now, trying doubling the number of entrants and add another 2 events for an additional smash game. Now, double THAT number and add ANOTHER smash game, which is your average national (maybe a bit below average entrants for this age?).

 

[Overswarm]

The way I see it, playing with two stocks is similar to playing a 2-out-of-3 match, similar to Street Fighter.

Street Fighter involves a complete reset. One stock would be similar to street fighter.

Losing a stock is like losing a round, you are down but still have the opportunity to turn it around and make a comeback. I feel as though with Brawl, the third stock was often unnecessary, and was only there to serve as a way to prove to your opponent how superior you were when you could keep all three. Extra stocks were a necessity in the previous smash games, 64 and Melee, because of how quickly a stock could be taken from you by an opponent. Multiple stocks were, in those games, more part of a match as a single combo or interaction could be the difference in stocks. However as we're looking at newer versions of Smash, combat is centralized around a lot more small interactions rather than large combos and gimps. This is a general statement, and I know there are exceptions in Brawl, however it's undeniable that - outside of MK and ICs - gimps and huge, sudden leads were very rare to occur and instead matches would more resemble a battle of attrition. With the changed ledge mechanics and - what appears to be - significantly improved game balance, it would seem like the necessity for a high amount of stocks is reduced drastically.

I played Brawl for a long time. Low % kills were not uncommon. Virtually every chaingrab to death or edgeguard, be it from Dedede, Falco, Pikachu, whatever character were large game changers and characters like MK were so powerful precisely because of their fantastic ability to kill at low %. We're already getting 0% edgeguards on certain matchups. The game hasn't been released in America yet.

Regardless, if you played Brawl at a competitive level you almost certainly saw 3 stock matches go down to the last stock and have someone who was down one stock to two come back and make a return. This was incredibly common at high level play. If Smash 4 is anything like Brawl, two stock will almost certainly increase variance to an unpleasant degree.

Imagine playing Melee at two stock. The implications of a Jiggs rest or a Fox air dodge off the side are massive. Similar things will exist in Smash 4, guaranteed. I've already seen players SD or get gimped, it's not like those things will disappear.

Not to mention that for a series that's making it increasingly more difficult to kill yourself, we're now at a stage where even moves like rollout and illusion no longer put you into freefall. I see no reason why we should be worried about leniency of people killing themselves when it's so difficult to do so to begin with.

The goal of a tournament is not to punish people for mistakes -- it is to make the best players win and to clearly illustrate their success. Allowing the better player to lose simply because we've altered the tournament format makes our tournaments less appropriate competitive venues.

Mistakes can occur, but in a best 2/3 between two highly skilled opponents can often be determined by a mistake such as an SD. Having two opponents in which one is more skilled can make that SD into a game-losing mistake for the better player. This is not the case in higher stock matches, as there is time for that skill to manifest itself in gameplay.

You're making a mistake and should be punished, and even so, losing a stock in a 2 stock game is better than wasting a game in a 1 stock ruleset. This only isn't true if you believe that losing a stock accidentally means you automatically lose the match, which is far from true.

A one stock ruleset would reduce variance only if it increase the amount of games played. To keep it at 2 out of 3 would be bad. I personally would not want a one stock ruleset, but believe it to be preferrable to two stock in that it will actually reduce variance.

And yes, losing a stock typically means you are more likely to lose the match by definition. It's how games are lost. If the skill gap is high enough, the better player will still win. The more skill there is, the less skill gap there needs to be to clearly determine who is the better player. With two stock, the skill gap must be pretty wide.

I cannot emphasize enough how silly it would be to run two stock tournaments.

2s seems like a ruleset that's more fitted towards the gameplay style of Smash 4 as it looks right now. I believe the gameplay is slow and methodical enough to allow plenty of time to adapt and adjust in a set, especially given that there are multiple matches in a set. An attribute of a skilled player is being able to adapt quickly and efficiently, so I think it's a bit silly that so many people are trying to argue that 8 minutes a match is required for players to effectively do so. I've played melee competitively for many years and I am able to pick up on most of the important things I need to understand about my opponent after a single stock. In brawl, from my experience, it takes less than that. Simple interactions in the neutral game after a bit of time will allow you to adjust and understand how your opponent plays. To again compare to Street Fighter, the first stock is much like the first round. Both players are feeling each other out and learning from it, but the player who wins the first round/stock is the player who can adapt and play accordingly most efficiently. Having to focus on learning and playing simultaneously is what makes playing new opponents exciting for all parties, but watching two players who are familiar with one another go through the same motions from the very start only makes a ruleset like Brawl's that much more off-putting. (see: most top level MK dittos)

I'd have to disagree. I don't see how Smash is even remotely close to Street Fighter in this regard and, having played both Melee and Brawl for years, am pretty confused at the thought of "simple interactions in the neutral game after a bit of time will allow you to adjust and understand how your opponent plays" being true against anyone at or above a player's skill level.

Not only that, but it allows for more efficient tournament time, which is a big issue we've seen in Brawl's time as a competitive game. Honestly, I believe people are just sticking to 3s 8min ruleset because they don't want to deviate from Brawl and what they are comfortable with. It's important that we try to look at the benefits and loses of each and weigh them accordingly. It's for that reason, that the benefits seem to outweigh the loses by a significant amount, that I support the 2 stock ruleset.

There are no benefits to using a two stock ruleset other than "games end faster'. None.

For this reason, I would hold off until it becomes an actual issue. People aren't even complaining about timeouts, they're complaining about the match taking up too much of the same timer we've had for over a decade. Tournaments are planned with each set taking the maximum number of time per game in the first place. Unless timeouts are occurring against both player's will, it's kind of a moot point to begin with.

 

[Overswarm]

You mean, like, changing the stock number?

No. I mean tournament formats, not tournament rulesets. Bracket is inefficient and is primarily used because it is simple.

A 60 person singles/doubles bracket happening in one day without going over time is standard. We get that number of people for some of our weeklies for PM here in DFW (which start at around 8pm and generally end at 1-2am). Now, trying doubling the number of entrants and add another 2 events for an additional smash game. Now, double THAT number and add ANOTHER smash game, which is your average national (maybe a bit below average entrants for this age?).

"I've decided to do all my exercises in five minutes to fit it into my schedule. One pushup, then one situp, then one pullup, then I run for 5 feet, then...."

Cramming multiple smash games together without enough time to properly run them all is its own problem. Why TOs run multiple games with eight minute timers and don't compensate for every game lasting 8 minutes is beyond me. It's like Tournament Organizing 101. Going over late is a natural progression of poor organization.

If Smash 4 has an eight minute timer, and Melee and Brawl had an eight minute timer, there is no way Smash 4's stock count of ANY variety could actually make a tournament last longer than planned for any game in the series.

I humbly request that people not try to alter Smash 4's ruleset to make room for older games in the same venue or emulate their speed.

 

[Keitaro]

I don't see the issue being time for the event. 2 stocks for Smash 4 will help the event run faster. The issue is simply that the game atm on average has sets lasting longer than Brawl. Brawl was already heavily questioned for stock count for years. Allowing Smash 4 to continue this trend from the start is nearly making the game look like a continuation of Brawl, which barely anyone is for. I still believe Brawl should have had 2 stocks by the way.

Just about everyone at a tournament I went to yesterday was annoyed at how long the matches were. They like the game a lot, but most of them are vouching for 2 stock matches instead. They have a lot of fun playing For Glory but become restless with 3 stock matches. Simply saying "change it to 2 to be less bored" may sound silly, but that factor alone can help with viewership, sponsors, and simply people staying to play the game. As much as I didn't like 1 stock, the viewers loved it. If initial viewers of this game like watching it, they may want to play. If the general consensus is "screw this game" off the bat, we may see a shorter life span than Brawl.

3 stock on the other hand? Check out the comments of this video. It's pretty bad. They are mainly complaining about the game looking slow, but making a game that looks "slow" play out with 3 stocks makes it look even slower. https://www.youtube.com/watch?v=cq2fkkQnVGI&feature=youtu.be

 

[Deleted member 245254]

I don't see the issue being time for the event. 2 stocks for Smash 4 will help the event run faster. The issue is simply that the game atm on average has sets lasting longer than Brawl. Brawl was already heavily questioned for stock count for years. Allowing Smash 4 to continue this trend from the start is nearly making the game look like a continuation of Brawl, which barely anyone is for. I still believe Brawl should have had 2 stocks by the way.

Just about everyone at a tournament I went to yesterday was annoyed at how long the matches were. They like the game a lot, but most of them are vouching for 2 stock matches instead. They have a lot of fun playing For Glory but become restless with 3 stock matches. Simply saying "change it to 2 to be less bored" may sound silly, but that factor alone can help with viewership, sponsors, and simply people staying to play the game. As much as I didn't like 1 stock, the viewers loved it. If initial viewers of this game like watching it, they may want to play. If the general consensus is "screw this game" off the bat, we may see a shorter life span than Brawl.

3 stock on the other hand? Check out the comments of this video. It's pretty bad. They are mainly complaining about the game looking slow, but making a game that looks "slow" play out with 3 stocks makes it look even slower. https://www.youtube.com/watch?v=cq2fkkQnVGI&feature=youtu.be

Maybe it's because I play Sonic...lol, but I do not even one bit see how it's reasonable to conclude that this game is slower than Brawl. What is the conclusive reasoning that this could even be possible in the first place? They literally increased ground speed. They didn't change anything that would make things slower.

The only slow I see right now are players. Players not having full mastery of their characters.

And to be fair, the comments on that video referring to the game being slow are just trollish, probably haven't even played the game.

At what amount of sacrifice do we give up the proper environment to demonstrate a clear winner in skill to its excitement to watch?

 

[Hylian]

No. I mean tournament formats, not tournament rulesets. Bracket is inefficient and is primarily used because it is simple.



"I've decided to do all my exercises in five minutes to fit it into my schedule. One pushup, then one situp, then one pullup, then I run for 5 feet, then...."

Cramming multiple smash games together without enough time to properly run them all is its own problem. Why TOs run multiple games with eight minute timers and don't compensate for every game lasting 8 minutes is beyond me. It's like Tournament Organizing 101. Going over late is a natural progression of poor organization.

If Smash 4 has an eight minute timer, and Melee and Brawl had an eight minute timer, there is no way Smash 4's stock count of ANY variety could actually make a tournament last longer than planned for any game in the series.

I humbly request that people not try to alter Smash 4's ruleset to make room for older games in the same venue or emulate their speed.

I would say your pretty disconnected from the scene if this is what you are requesting. The majority of the smash community play multiple smash games and having several at a tournament is a huge appeal for the scene. Tournaments for some may be about finding the absolute best player etc etc but for most people it's not. If that is something you are not happy with I'm sorry but that is how the smash scene currently is in most places.

 

[Shaya]

Maybe it's because I play Sonic...lol, but I do not even one bit see how it's reasonable to conclude that this game is slower than Brawl. What is the conclusive reasoning that this could even be possible in the first place? They literally increased ground speed. They didn't change anything that would make things slower.

The only slow I see right now are players. Players not having full mastery of their characters.

And to be fair, the comments on that video referring to the game being slow are just trollish, probably haven't even played the game.

At what amount of sacrifice do we give up the proper environment to demonstrate a clear winner in skill to its excitement to watch?

We've lost dynamic buffering. Apparently only being single moves now (I don't know if that means we can't buffer dashes either or not), but that slows the game down at a practical level.
Furthermore the game is being played on 3DS controls without a C-Stick. The c-stick gives us up to 7 inputs in a single frame (in previous games). Not having a god-stick slows us down. And not having the fluidity of gamecube controllers / the top players not being used to the 3DS slows us down.
Slow also refers to game length. Icreasing stage blast zones and removing most characters free early percent combos (something that was a significant imbalancing aspect of Brawl), definitely sped up matches when every 0% meant an easy 0-50% for most viable characters. We'll come to see, but when we have Dedede get a good 40% from a down throw at 0% and others are yet to be discovered, those will shape viability as well as the speed at which matches play out. The large blast zones means something common to high level Smash (good DI) allows excessively long survival and we've yet to adjust to this shift, we're all talking about going off stage to secure the kill, but we're not really sure how capable we can be at this yet, once we learn more that will also speed up.

 

[Ishiey]

Speed of the game is one thing, but reducing time doesn't necessarily change gameplay. 2s/5m is 2:30 per stock and 3s/8m is 2:40 per stock, I doubt that 10 second difference per stock (6.7% increase) will do much to fundamentally affect how we play or change the amount of time-outs. There are also some arguments like people having more incentive to time out, from a stamina perspective (easier to hit 3min with one stock than 6min with 2?). Not sure how those hold, but something to consider regarding lower stock count that might work against pace issues.

@ Overswarm , you've brought up before that the timer is balancing the line between how long we have to run the tournament, and how valid we want time-outs to be as a win condition. For example, take the standard deviation for stock length, and extrapolate for match length with a varying number of stocks. If we set the timer at one sigma over the average, we're basically saying that time-outs are... what's that, like 15-20% of all matches, and thus a reasonably valid way to win? But if we go with two sigma we're down to 2.5% or so, and by making it that rare it's essentially saying that we don't want people to win via time-out, labeling it as a "less valid" win condition.

I think a logical approach would be to first find out how much time you can afford to have each game take for a tournament assuming they all went to time and last game, then figure out how many sigmas above the average match length you want that to be, then pick the amount of stocks that most closely resembles that sigma. All that being said, something like 2-stock 6-minute likely hits a logical sweetspot (6min being a bit above the average length of some top-level Brawl matches), but this ignores the other issues that you mentioned. Ex. the effect of a stock lead/defecit, character-specific effects, matchup-specific playstyles, etc.

Also, variance can be measured with standard deviations, but even then you're arbitrarily picking how many standard deviations is acceptable. Why don't we run Six Sigma practices?


I'm like... almost 100% positive someone came out with data from Brawl MLG matches or something, saying that 75%+ of matches where a player took the second stock first resulted in them winning the set. Was that you, OS? I think the thread concluded with logical people saying that stock decreases wouldn't really introduce much variance, but idk, just wanted to bring that up in case anyone else knows where the thread is and all.

I think it's important to be flexible for when the WiiU version comes out, it may be best to simply let rulesets run wild for the 3DS release and then put out a ruleset guide 2 months after when the game is out for WiiU and we've had some time to observe community reaction.

Sorry if this post was bad and disorganized, haven't had much free time as of late :x

 

[Overswarm]

Speed of the game is one thing, but reducing time doesn't necessarily change gameplay. 2s/5m is 2:30 per stock and 3s/8m is 2:40 per stock, I doubt that 10 second difference per stock (6.7% increase) will do much to fundamentally affect how we play or change the amount of time-outs. There are also some arguments like people having more incentive to time out, from a stamina perspective (easier to hit 3min with one stock than 6min with 2?). Not sure how those hold, but something to consider regarding lower stock count that might work against pace issues.

We've tested lower time count in the midwest. The shorter the timer is, the more likely a timeout will occur. There was a direct correlation between time getting to around 2 minutes at all and time outs. Viewing of those matchups showed that two minutes is what most players see as "time running out" and that's when they simply try to camp, rather than a "the game forced me to time out" thing.

We tried running with no timer and the games went much faster as there was no incentive to stall for time in the first place. This is unfortunately not feasible in the long run due to things like venue closing times.

@ Overswarm , you've brought up before that the timer is balancing the line between how long we have to run the tournament, and how valid we want time-outs to be as a win condition. For example, take the standard deviation for stock length, and extrapolate for match length with a varying number of stocks. If we set the timer at one sigma over the average, we're basically saying that time-outs are... what's that, like 15-20% of all matches, and thus a reasonably valid way to win? But if we go with two sigma we're down to 2.5% or so, and by making it that rare it's essentially saying that we don't want people to win via time-out, labeling it as a "less valid" win condition.

It's not that we don't want people to win via time-out -- it's that we already have an approved upon standard we can use as a non-arbitrary reference point.

I think a logical approach would be to first find out how much time you can afford to have each game take for a tournament assuming they all went to time and last game, then figure out how many sigmas above the average match length you want that to be, then pick the amount of stocks that most closely resembles that sigma. All that being said, something like 2-stock 6-minute likely hits a logical sweetspot (6min being a bit above the average length of some top-level Brawl matches), but this ignores the other issues that you mentioned. Ex. the effect of a stock lead/defecit, character-specific effects, matchup-specific playstyles, etc.

Also, variance can be measured with standard deviations, but even then you're arbitrarily picking how many standard deviations is acceptable. Why don't we run Six Sigma practices?

Two sigma is a universally agreed upon standard for "statistically significant"; we could do six sigma, but I see no reason to.

I'm like... almost 100% positive someone came out with data from Brawl MLG matches or something, saying that 75%+ of matches where a player took the second stock first resulted in them winning the set. Was that you, OS? I think the thread concluded with logical people saying that stock decreases wouldn't really introduce much variance, but idk, just wanted to bring that up in case anyone else knows where the thread is and all.

It was me! But that data didn't show that to my knowledge and, if it did, I'd say a full quarter of the matches being invalidated is a prety huge change.

I think it's important to be flexible for when the WiiU version comes out, it may be best to simply let rulesets run wild for the 3DS release and then put out a ruleset guide 2 months after when the game is out for WiiU and we've had some time to observe community reaction.

Sorry if this post was bad and disorganized, haven't had much free time as of late :x

Possibly. I'm going to do a lot of experimentation during this time frame, for sure.

I would say your pretty disconnected from the scene if this is what you are requesting. The majority of the smash community play multiple smash games and having several at a tournament is a huge appeal for the scene. Tournaments for some may be about finding the absolute best player etc etc but for most people it's not. If that is something you are not happy with I'm sorry but that is how the smash scene currently is in most places.

I don't care if "the scene" as a whole grows to like playing 5 smash games or 20 fighters or loves to have a mariachi band play a song whenever someone gets a shine spike in tournament.

The ruleset and format for Smash 4 should in no way be altered at its core simply to make room for other games which, conveniently, aren't lowering their stock count.

You want to edit games to make room? Make Melee 3 stock and PM 2 stock. Their tournaments will run faster and you'll have time for Smash 4.

 

[Shaya]

I think expecting multiple smash events to have similar time lengths (varied by attendance / equipment) is fair. Rule sets reflect tournament dynamics, otherwise we wouldn't have a timer nor a 300% stalling rule, why not all sets are best of five or seven, etc. While it may be a dangerous "creep" that we previously did not focus so much on, it's definitely relevant for us as a backroom (or TOs/otherwise) at this stage to ensure we're maintaining and expanding the strongest the smash community as a whole has ever been.

There are tons of concessions we have to make for the new dynamics that we're dealt with, some amazing TOs have been juggling melee, brawl and PM and every event/day is almost like living on the wire in terms of managing everything involved. Live streaming is huge and requires a lot of scheduling and expertise to manage multiple games for. TOs/Livestreams are going to give people what they want and that's inevitable. The environment we're in has to be considered.

 

[Overswarm]

That mentality leads directly to a schism in communities. The midwest-east is moving almost entirely to Smash 4; Melee might have a few events here and there but PM has just been a placeholder for most of us until Smash 4 came out. Our tournament rulesets shouldn't be affected by Dallas wanting to have Smash 4 3DS/Melee/Smash 4 Wii U/PM/Brawl/64/Street Fighter/Dance Dance Revolution/Iron Man/Crews/Doubles with triple elimination/whatever combination tournament.

There's a hard limit on how long a tournament can last and the solution is not cramming as many games as possible into one venue. At one point or another, something will have to give. The proposed solution now it to hamstring the new game to allow for the extended life of the old. I vehemently disagree with that premise.

 

[Overswarm]

If you guys want to run multiple events for multiple games, by all means do. I'm running Smash 4 tournaments.

If including Smash 4 into your current tournament schedule makes it last too long, this is an issue with your tournament organization -- not Smash 4. The timer has been at 8 minutes for over a decade now.

Feel free to lower the stock count for all your games, remove pools and go straight to bracket, do best of 1 instead of best of 3, or all sorts of other things that lower the competitive integrity of your event so you can cram it all in one place. If that's your prerogative, go for it!

But at one point you'll have to pick and choose which games you actually want to play. This may very well be that point and, if it isn't, the Wii U release of Smash 4 that will actually take up setups and require different equipment almost certainly will be. Smash 4 is going to be my choice, just like Brawl and Melee were before it, and my ruleset won't be standardized by poor tournament organization.

 

[Shaya]

I don't think that's the premise at all though. How is expecting a 50 man Melee tournament to be within a fair variance of a 50 man smash 4 tournament lengthwise hamstringing a game? What schism in communities would be worse than Melee vs "us" again when we're giving them less resources whilst their contributions are the same?

In a game where hopefully zero-response scenarios (i.e. infinites/debilitating chain grabs) are minimal, stock count does end up impacting less. Anecdotal, but I've given Brawl a chance at 1 stock for a fair amount of time, but Ice Climbers (and I guess ZSS) ruined it; characters which multiple stock variance were required to not have inane match ups. I advocate/killed PM with 4 stocks in my region and we're comfortable with 3, perhaps recovery nerfs may change our minds, but events are smoother now with the change.

When it comes to "us" and rulesets, there are no direct givens. Is impartiality a goal? Probably. If your events are Smash 4 focused, you can go for small variations that impact meta little without complaint. I think it'll be a different story for the console version.

If resources budge, then they budge, and less events are run. Right now that has less to do with the rulesets and more to do with how many people attend and how many set ups they have. I think there's a likelihood we'll need Smash 4 only events to deal with the numbers, but we don't know yet. I personally feel the need to drop PM from my events the moment smash 4 gets underway though, as I don't see it being possible to run events for three games with all events having the same/similar amounts of people and at least around here there's a general melee preference.

[I think it'll be a different story for the console version].. transition to LCD set ups invalidating the ability to share resources between games in a way we haven't dealt with before, forcing melee out of the rotation depending on the scale of new people/new technology.

 

[Overswarm]

I don't think that's the premise at all though. How is expecting a 50 man Melee tournament to be within a fair variance of a 50 man smash 4 tournament lengthwise hamstringing a game? What schism in communities would be worse than Melee vs "us" again when we're giving them less resources whilst their contributions are the same?

Who says all Smash tournaments need to have the same average time per game? Melee isn't the gold standard for set length for League of Legends or Killer Instinct, why would it be the case for Smash 4?

Expecting Smash 4 to fit into the mold set for Melee is the definition of hamstringing the game. If Smash 4 is most comfortable with longer sets, then longer sets it should have. It's important to note that the max time amount has yet to change, so Smash 4's games taking longer would have no actual impact on tournament organization.