Bo5 Instead of Bo3

[Armada]

If we compare Europe and USA we can see difference between bracket matches.
In Europe we having best of 5 really early in the bracket but in USA you guys having that in Semifinals or maybe just in finals.

I know best 5 take more time but i think that should be possible to make it.
Or maybe not the time is the reason in USA maybe you guys wanna have Bo3 (i am not sure what the reason is).

But what do you guys think?
Bo3 or Bo5?

For me Bo5 is so much better and so much more fare.
In pools and 1st round in Winners/losers it should be Bo3.

Why we should play Bo5
- You have to win more matches so that's a bigger chance the best player wins.
- If you are playing more matches the best "reader" will notice more things. Read your opponent is a big part of the game so that's a big reason to have Bo5

Why we shouldn't have Bo3
- The ruleset: With stages like RC and KJ it is REALLY stupid to have Bo3 (change the stage rules and change it to Bo5 please),
- Bo3 is to few matches

 

[DoH]

I like this idea; it also takes away the "importance" of the first match in determining the set, which is one of the reasons we removed a lot of counterpicks because some people (who play "real characters" like Falcon) thought they were too extreme and you only needed to win the first match and then put your cursor on the "broken" cp and it was gg.

Bo5 also eliminates being punished for suicides so harshly

Armada; do you guys do Bo5 for the whole bracket or after the first couple of rounds? How does your pools play work?

 

[SwiftBass]

disagree with bo5.


Smash already takes too long as a whole in my opinion. I am not saying that making this change will solely change the overall time spent "smashing" drastically but I DO feel as if this could be the beginning of a trend and an overall step in the wrong direction in my opinion. I can compromise with quarter finals(in both losers and winners) and on being bo5 though.


also to counteract a point made about stages...we could always change the stage as opposed to the length of the set. Why not create a list with less stages that appear to be "lame" or whatever you are suggesting with this:

Why we shouldn't have Bo3
- The ruleset: With stages like RC and KJ it is REALLY stupid to have Bo3

 

[KishPrime]

I agree that Smash tournaments take long enough as is. There were legitimate time concerns at every event I ever attended. I don't see how Bo5 is feasible.

Plus, the best people win anyway. Upsets are somewhat more common, but one upset doesn't win you money, and frankly, trying to eliminate all upsets (making sure the best player always wins) is a pretty poor way to plan your rules. The rules are what define what the best player should be, and then the best player is the one that wins.

For me, the rationale for Bo5 has not been to "make sure the best player wins" but to make sure that both players are given a longer chance to shine when the prize deserves someone who shines.

So I'll throw out the controversial idea that there's no such thing as designing a ruleset so that the best player wins. I always hated that phrase. You design a ruleset to test an optimal set of skills in an environment where both players have equal opportunity to make the same choices, i.e. a "fair" ruleset.

 

[pockyD]

any thoughts on 3stock Bo5?

 

[Armada]

D'oH: Bo3 in pools and first round in WB/LB (depends from tournaments to tournaments).

KishPrime: With Bo3 that's a bigger chance the best player not win.
And i know people can start saying 'why will we not playing Bo7 or Bo9".
But Bo3 is too few matches. It's not even close the best player win.
In close sets it is really important to have Bo5

The best player winning on Bo5 instead of Bo3.

One example.
You (player A) are playing with some guy every day and you are winning 60% of the matches (and player B 40%).
Everyone can see it is close BUT player B have so much bigger chances to winning in Bo3.




PockyD: 3 stocks sucks if you asking me.
One gimp/Edgeguard is too much if you only having 3 stocks.
3 Stocks and you gimp your opponent he have lost 33% of his stocks instead of 25%.

 

[SwiftBass]

PockyD: 3 stocks sucks if you asking me.
One gimp/Edgeguard is too much if you only having 3 stocks.
3 Stocks and you gimp your opponent he have lost 33% of his stocks instead of 25%.

what makes 33% so much more significant than 25%?

One example.
You (player A) are playing with some guy every day and you are winning 60% of the matches (and player B 40%).
Everyone can see it is close BUT player B have so much bigger chances to winning in Bo3.

that is like flipping the same coin everyday 5,000 times and expecting the outcome to be 2,500 heads and 2,500 tails. you can never truly control odds or expect them to deliver as advertised. (in this case 50/50).

How do you describe winning? per game or per set? are you also suggesting that events that consist of "winning"(whether its per game or per set) are disjoint? I find that unbelievable.








In another thread I addressed the "middle margin" of the "best of" sepctrum(1 being the first extreme and infinity being the other extreme). who is to say that a bo7 is more prominent in giving results than a bo5? by the standards talked about perhaps a best of 19 or best of *insert high odd number* is the absolute best. This "middle margin" of the spectrum to me is such a gray area. lets take away time constraints for a sec......would a best of 27 be the better option than a best of 13? What is the threshold on how many games is too many? And more importantly what justifies such a threshold besides time(which is another uncontrollable variable)? Maybe we should create a threshold on how many game are too many(relative to time) and adjust the length of sets in ordinance with the amount of time given to complete a major....that sounds pretty easy right?



TL DR; the argument brought up by armada is counterintuitive

 

[Armada]

If someone gimp you you still have a really good chance to coming back with 4 stocks but not with 3.
But having 4 instead of 3/5 is hard to give good arguments why we should have so i guess i have no better arguments at that point.

I understand your point but for every move you are going (from Bo1 to Bo3 or from Bo3 to Bo5) the chances for Player B are so much worst.
And a Bo5 taking not so much time.

But a good reason why we should have Bo5 and not Bo13 is because peoples get tiered and too many matches is not a good thing.

Draw a "perfect line" is not easy and giving arguments why Bo5 are better than Bo X is not easy but the most people say Bo 5 is so much better. We have to draw a line somewhere and i guess the most people agree we have to draw a line between Bo3 and Bo5 (maybe Bo7 in finals but for me Bo5 is really good here to). And if we only have this 2 choices (Bo 3 and Bo5) we should have Bo5 because the reasons i just gave you.

We have to take regard to time and the competitive part and Bo 5 is so much better in a competitive view than Bo 3 and it is possible to have tournaments with Bo 5.

And i describe winning like winning a whole set.

 

[KishPrime]

You're missing my point. My point is that the rules set the criteria for who the best player is. The rules do not determine who the best player is. If you have it in the wrong order, then you are designing a ruleset because you think that Player A is losing under these rules most of the time, but he should really be winning because he's "good."

The good players are the ones that win, given the ruleset. Best of 3 is perfectly fine. It worked for years and years, and the best players won tournaments. There will always be upsets.

In fact, most upsets don't occur because of the number of rounds but because he SDs or gets "gimped" (the player deserves to lose because of his own bad play) or because of bad matchups (the player deserves to lose because his knowledge of the game is not thorough enough).

 

[Armada]

"The good players are the ones that win, given the ruleset".

So if we starting to play Bo1 with 1 stock is that fare?

That argument is really bad and i know my example was dramatic but that shows how bad your argument is.

And i will tell you why the best player in USA always win.
All the best players in USA (Ken,M2k,Mango) have been so much better so it dosen't matter for the rest in USA.

Bo3 or Bo5 dosen't matter in sets when Player A is so much better but if two players are close Bo 5 Is so much better.

 

[KishPrime]

Yes, by definition, Best of 1 is totally fair. Both players have equal opportunity. The reason we don't do that is because it does not provide enough opportunity for the players to show the variety and depth of their skills, which has been deemed valuable by the community.

If you already know who the best player is and want to design a ruleset where he wins every time, why would you even play the matches? The only was the "worse" player can win a match is if you have predetermined that player to be "worse" according to your own arbitrary judgment. If they're close in skill, then they should both occasionally win sets, regardless of if it is Bo5 or Bo3 or Bo13.

 

[Marc]

I think the idea is that a difference in skill will show more clearly when there is more time. You can basically treat a set as an aggregate of match results and more data means less room for noise. There really is no reason not to have more matches other than time restraints. At European tournaments it's often feasible to start with Bo5 rounds earlier in the bracket than the US ever does and I'm not entirely sure why that is as it holds true for bigger events as well.

 

[Armada]

Equal opportunities is not enough.

Equal and fare opportunities will it be.

In Bo 3 you haven't enough "opportunity for the players to show the variety and depth of their skills".

 

[unknown522]

even still, the issue is the time constraints for tourneys. We will run out of time if the sets are Bo5.

 

[Skler]

Yeah, we run out of time with Bo3 here all the time.

Everyone would love to play Bo5 otherwise. Depending on tourny size I've seen people make semi and quarter finals Bo5.

 

[KishPrime]

I personally feel that Bo5 doesn't add nearly as much as it's being given credit for. 95% of matches will end the same, you'll waste 33% more time, and of the remainder of the 5%, most matches no one will care about anyway.

But I also believe everyone can run their tournament however you want, and there's certainly nothing wrong with Bo5. I just think it's incredibly inefficient.

 

[TheZhuKeeper]

I would hate Bo5 for anything more than semi's. As a spacy, I'd be tired as **** LOOOL.

 

[Divinokage]

I would hate Bo5 for anything more than semi's. As a spacy, I'd be tired as **** LOOOL.

I personally wouldn't mind Bo5 and I think it would determine who the better player is. Again personally I think I perform better in Bo3 due to having a choice to choose a BS stage for myself in order to almost for sure win the 2nd round.

A Bo5 early on would definitely have more variety and show more who has more experience with different stages with different match-ups. It would also help the tier list a little bit more even if it's quite clear already. Wouldn't it be better statistic wise?

Also, I think it would add more hype to some matches, you would probably need a much stronger mind in order to regain or maintain your momentum during sets. I think it would be awesome.. but ya time restraints sucks. It would work more smaller tournaments but bigger tournaments I'm not so sure. I say we give it a try.

 

[Nihonjin]

When you play for a money spot, best of 5 is the only way to go.

 

[KishPrime]

It's not going to add any hype. It's going to make it drag. Do you realize how many sets are going to still be 3-0? In a seeded tournament, early rounds are full of them because of all the mismatches. Trust me, we've tried stuff like this before. If 90% of matches are going to go 2-0 in the early seeded portion, why why why would you waste the time to go to 3-0?

Think of it this way. The other guy has already tried his best counterpick and lost. Why would his second-best counterpick give him that much more of an advantage? It doesn't demonstrate any more depth until you get to the later rounds, when players are much closer in skill on average.

Like I said, I'm fine with people doing whatever they want at their tournaments, and if you have time, by all means do it. But it's really really inefficient.

 

[Divinokage]

It's not going to add any hype. It's going to make it drag. Do you realize how many sets are going to still be 3-0? In a seeded tournament, early rounds are full of them because of all the mismatches. Trust me, we've tried stuff like this before. If 90% of matches are going to go 2-0 in the early seeded portion, why why why would you waste the time to go to 3-0?

Think of it this way. The other guy has already tried his best counterpick and lost. Why would his second-best counterpick give him that much more of an advantage? It doesn't demonstrate any more depth until you get to the later rounds, when players are much closer in skill on average.

Like I said, I'm fine with people doing whatever they want at their tournaments, and if you have time, by all means do it. But it's really really inefficient.

I guess that makes sense since the stronger players will normally always get top 10 anyways regardless of Bo3 or 5.

 

[HyugaRicdeau]

Yeah, we have to think about how often the results would be different in a Bo5 versus a Bo3. I notice that in pools, usually around 80% of the sets are 2-0. It's very unlikely that a Bo5 would have changed the results of sets like that given that the losing player already lost on his best counterpick, and then would have to win on his opponent's two best counterpicks. So there's 20% leftover, and even if the skill level is totally even, only a quarter of those would change results in a Bo5 set. So we're talking about on average 5% of the sets in pools that would change in Bo5 sets. I don't really think all the extra time is worth that.

Plus, I think there is some merit in allowing for upsets. Sets don't determine who is the better player, they determine who the better player was during the course of the set. Shorter sets emphasize consistency as part of what we consider "skill." I mean if we go too far in making sure the best player wins, why doesn't everyone just give Mango their money and have a day-long smashfest instead?

 

[SwiftBass]

Plus, I think there is some merit in allowing for upsets. Sets don't determine who is the better player, they determine who the better player was during the course of the set. Shorter sets emphasize consistency as part of what we consider "skill." I mean if we go too far in making sure the best player wins, why doesn't everyone just give Mango their money and have a day-long smashfest instead?

besides the fact that I hate the basis of the thread's argument........I like this point and fully agree.

 

[KishPrime]

I tried to make that exact point like 3 times already...thanks Hyuga for apparently summing it up far better than me.

 

[SwiftBass]

an argument like this can't be systematically approached(due to uncontrollable factors) and I believe that hygua best captured the essence of what defeats the systematical approach of the "best of" argument:

varying consistencies in skill and the desire to not make odds fulfill their intended outcomes on a more strict basis(i.e upsets being an incentive to play)

amirite hygua?

 

[AXE 09]

But if we want more upsets to happen, shouldn't we allow more counterpick stages to be legal? Or shouldn't we allow items to be on?

I always thought that the current ruleset was made so that the best player out of the ones who are playing each other will win. I think that upsets should only happen if someone who normally would lose to one person ends up winning because he got better, and not because he got lucky since his opponent accidentally SD'd the first match, won the second match, and lost on a counterpick stage that limited his character and benefited the other player for the 3rd match.

Bo3 in pools and first round in WB/LB (depends from tournaments to tournaments).

I honestly think Bo5 WOULD be the best way to go if time was not an issue. However, since almost every tournament organizer is worried about time constraints, this doesn't happen. If time was not an issue, I feel that Bo5 would be much more fitting than Bo3.

 

[HyugaRicdeau]

I think so, Swift, if I understand you. If we hypothetically could have sets be best of 101, and player fatigue wasn't an issue, it would essentially remove any test of consistency, because all the inconsistency would have averaged out. Now, not everyone is going to agree on where to draw the line as far as how strongly it should be a factor, I'm just pointing out that it is an important part of being "good."

Axe:
I don't think anyone is necessarily saying we want more upsets to happen, just that "the best player should win" is not the be-all end-all principle in rule-making. We would want the upsets to happen within the confines of what we regard as "competitive gameplay." Getting lucky from a mistake the opponent made is different from getting lucky from a random stage effect, like say falling blocks on Green Greens, because it was still something within the opponent's control.

 

[SwiftBass]

I think so, Swift, if I understand you. If we hypothetically could have sets be best of 101, and player fatigue wasn't an issue, it would essentially remove any test of consistency, because all the inconsistency would have averaged out. Now, not everyone is going to agree on where to draw the line as far as how strongly it should be a factor, I'm just pointing out that it is an important part of being "good."

exactly the fundamental argument is soooo incoherent with what we are trying to accomplish that I feel that people should make some other statement along with it. simply implying that statistically it seems better doesn't fly. I do not mind ppl wanting bo5 because they believe that it will effect other factors like hype or maybe they just want more grueling sets or maybe they just like the number 5....but the basis of the "best of" argument is not stand-alone imho.

 

[Aldwyn McCloud]

I completely agree on the Bo5 idea. Competitively speaking, best of 3 doesn't have any advantage at all. The only reason it's used is because of time issues. Too bad that's a big ****ing issue lol. Generally speaking, my opinion is that sets should be made Bo5 as much as possible. Unfortunately, tournament size plays a big role. I believe European tournaments have more Bo5 sets because they also have less attendants. Italian events are even smaller, so that Bo5 is used as soon as the top seeds meet in the bracket. Since Bo5 sets improve the competition, my belief is that TO should behave in a stricter way in order to have tournaments run on time. To me, this seems the fairest way. The whole tournament shouldn't suffer from a few people not being quick enough to do their stuff (of course, there should be exceptions but you got the point).

 

[forward]

I believe that the best rule to determine a player would be a best of 5 but you have to win by two matches, similar to what they do in Tennis (and I think volleyball).

There is no denying that having more matches gives a more accurate result of the better player. By definition a better player is a player who wins. If both players win then the definition needs to be clarified, who wins more? OR who wins quicker?

Or maybe it's both?

To be honest guys I've thrown myself into a direction I wasn't even thinking about when I decided to reply.

What's more important in determining who the better player is? Long term consistency or short term consistency?

Assuming that player fatigue is not an option, I would believe more long term gives an accurate answer.

But player fatigue is not something that can be ignored.

I honestly can't define what makes a player better. I want to say that having more matches is key but I can't seem to prove it.

Perhaps if you take Armada's idea. Can a one stock match determine the better player?

Kish proved that it can. The results would be less consistent, but look at poker, it's not consistent but it's loved and the skill is admired.

I think the reason we do Bo3 is for convenience, and that is all. You could technically use whatever rules you want and in the end the "better player" will still emerge victorious.

Better player will always be subjective, but it feels like we all have this intuition of what it is and should be but nobody can really explain it in a clear and objective way.

 

[Fly_Amanita]

I honestly can't define what makes a player better. I want to say that having more matches is key but I can't seem to prove it.

The ruleset defines who is the better player. If there are two people and one of them tends to win best of 19s and the other tends to win best of 3s, then the first guy is better at the version of Melee with bo19 sets while the other guy is better at the version with bo3 sets. I think the question should be which skills are we more interested in testing and which ones are we able to test (e.g. a bo19 wouldn't work for time reasons, although somebody may like how it lets to players adapt to each other over an extended period of time), rather than what ruleset lets the "better player" win, since the "better player" is a consequence of the ruleset.

 

[forward]

Does that mean it's impossible to determine who is a better player outside of a tournament set?

 

[Fly_Amanita]

Well, what I'm trying to get at is that the ruleset is a part of the game we play and thus any way of defining who is the better player should depend on the ruleset. We should try to adjust the ruleset to test the skills we're most interested in; you could still, at least in some cases, be reasonably confident about which of two people would perform better in tournament from the results of serious friendlies.

For what it's worth, I doubt that adjusting the number of rounds played will often significantly alter tournament results, so I think concerns about time should dictate the number of rounds played more than anything else.

 

[KishPrime]

Does that mean it's impossible to determine who is a better player outside of a tournament set?

More accurately, it is impossible to determine who is a better player outside of a written determination of an "ideal." For us, that was defined by winning tournaments. It could have been defined by BTT or HRC. It could've been defined by how fast one can suicide.

Better can only be as objective as it relates to the subjective definition of criteria.

All that said, Bo5 really doesn't make that big of a difference 95% of the time.

 

[SwiftBass]

Does that mean it's impossible to determine who is a better player outside of a tournament set?

dunno I would say yea given that the margin of skill is not overwhelmingly great. but even then how can you tell if the person is actually "playing?"


Tourneys provide that common ground with the material and mental incentives. relateively speaking I'd say that friendlies have little to offer when it comes to determining who is better.

 

[Samochan]

What about a compromise?

Add in more Bo5 matches, like say Quarterfinals or if time and other factors allow it, round 2/3 in bracket. Don't the tops of the pools face each other around this point? Depending of the size of the tourney and bracket ofc.

If time and tournament size is no issue, Bo5 is reasonable for the whole bracket. Imo, Bo5 should be in effect for quarters and onwards.

 

[LunInSpectra]

The less matches there are, the better the chance one may have at "not being found out".

 

[Fly_Amanita]

I wouldn't object to introducing Bo5 sets earlier into a tournament if time permits, although whether there is enough time isn't something the TO will necessarily know beforehand.

 

[Divinokage]

The less matches there are, the better the chance one may have at "not being found out".

I disagree with this statement.

You cannot play the same vs anyone you face since yes they will adapt to you but then you can adapt to them back... and then they will have to adapt to you again to counter you. It's a whole spiral of change so to speak. I think it's definitely possible to change your playstyle while playing a regular or bo5 tournament set. It's to remove habits that makes you unpredictable, it's hard but as long as you don't go into a auto-pilot mode then you can win vs anybody.

 

[KishPrime]

they will adapt to you but then you can adapt to them back... and then they will have to adapt to you again to counter you. It's a whole spiral of change so to speak.

I think this is actually a brilliant way of getting to the heart of the matter. You want every match to allow both players to have this opportunity to adapt to each other while simultaneously not getting adapted to for several turns around the cycle.

For an extremely lopsided matchup, there is no spiral because the other player can't adapt. In close matches, how long does it really take? In my opinion, there is enough time for players to go 8 or 9 times around this spiral in the context of a single Best of 3. If one player can't get ahead of the curve going that many times around, he doesn't deserve to win.

In a Best of 5, you're basically making sure that a player didn't ride only those 8 or 9 sets of tricks, and testing the ability and depth of knowledge (or in this case, ability to counter). It's not necessary when players are unequal in skill to begin with, as they likely won't even make it through 8 or 9 revolutions. We tended to force this by using the "Each Stage Only Once" rule.

In a perfectly even match, of course, it doesn't matter how many rounds you play, because it'll come out basically even every time.