Round Robin Bracket Hybrid Format

Hey everyone,

In Michigan we've decided to mix things up for an upcoming tourney by running a round robin/bracket hybrid for loser's bracket of Melee singles. We think this format puts more weight on ability to defeat a variety of players and playing styles, resulting in more accurate final placings, not to mention more playing time for people who don't make it very far into the bracket. What do players and TOs in other regions think about this?

Blank Bracket:


Example:

AzN_Lep said:

This is a pretty neat idea, I'll give it a go at my next tournament.

I recently gave the Swiss RR format a try, and most people seemed to enjoy it at first, but it ended up becoming pretty taxing by the end of the tournament. I did 6 rounds of Swiss into an 8 man bracket and even the folks who didn't make brackets were pretty exhausted. That's the only real problem I can see with this format also.

So in the example provided here you have a 32 man bracket. Let's say Kels loses his first round but then proceeds to win every round into grand finals. He'd have to play 14 sets by the end of the tournament. While this particular case is pretty unlikely, that's a LOT of smash for anyone to play in tournament. I guess this isn't necessarily a problem, but it's definitely something that you might wanna keep in mind.

Click to expand...

I don't think this is an issue at all. It's true that someone who pulls a Mango at Pound 3 would have to labor through 14 sets with all those round-robins, but they would have to play 11 sets through grand finals under the traditional double-elimination bracket, too. That difference isn't quite significant enough to be addressed, and the scenario is also extremely unlikely to happen like you mentioned. On the TO's part, I guess seeding/placement would have to be done extremely carefully for winner's bracket to minimize the chances of a top player making an early exit. But there should already be enough incentive for those top players to stay alive in winners.

AzN_Lep said:

Oh but for things that could/should change - I feel bad for the players that go 0-2. A lot of times those kids are the ones that want to play the game the most, but then get completely wrecked and never wanna come back. It'd be really discouraging to attend a tournament that promotes more playtime for less experienced players, and then end up going 2 and out. My only suggestion would be to play an initial round of pools during the first round of losers and have the top 2 advance. Of course I'm sure you've noticed the problem here would be - what to do if the top 2 players happen to be from same branch of the bracket (top or bottom). In this case I'd just take the simple route and rename the players. Yes this'll change the perception of "who beat who" for the first round of the tournament, but since we're already tinkering with the bracket for the new tournament format, I wouldn't be too concerned with the bracket image looks like at the end.

Click to expand...

You read my mind. I've thought about running eight round-robins as the 1st round of losers to ensure no two-and-outs, but you're right that if the top two of a round-robin were to come from the same branch it would complicate things. I mean, this tourney format is pretty much incompatible with tio to begin with, and I was planning on just typing out the names onto a blank bracket in Photoshop... so with a little extra manual TO work, it could easily be done. Definitely still considering it. Thanks for bringing up some excellent points. Let me know if you'd like a copy of my blank bracket, I have a .psd file ready to go.

Yeah, I'm writing a 2010 Melee Year in Review. The thing keeping me from completing it is the disturbing lack of photos from national events last year. If someone could link me to any online album where some kinds of pictures were taken at Pound 4, APEX, or RoM 3, it would be greatly appreciated.

MK26 said:

Am I getting this right?

For the modified idea, the losers of Round 1 are put into groups of 4 and the top two advance.
The winners of LR1 and the losers of WR2 go to LR2 and once again are grouped in fours, but now only the top player advances.
Rinse/repeat: winners of LR2 and losers of WR3 go to LR3

In other words:

Round 1: unchanged
Winners: unchanged
Losers R1: Losers grouped into fours and play a RR; top two advance
Losers R2: Two who advanced from LR1 (I assume it'd have to be different pools or else two players would play each other in two consecutive RRs) and two who lost in WR1 are grouped into fours and play a RR; top two advance
Losers R3: Winners of two different LR2 pools and losers of two different WR3 matches are grouped into fours and play a RR; top two advance
Repeat until losers RR after winners' semis; the winner of this final pool plays the loser of Winners' Finals (in Losers' Finals, and the winner of this match plays the winner of WF in GF)

This does make a lot of sense, but could potentially double the length of Losers'...where three matches would normally be played (ie to eliminate 3 players in losers) now six or more must be played (ie to crown a champion of a four-player RR and eliminate three players, even when not accounting for tiebreaker matches/sets
And also, it would decrease the hype of matches...regular player A could beat amazing player B, which would normally be incredibly hype and be a massive upset and so on, but if regular player A loses to good players C and D while amazing player B beats good players C and D, amazing player B would still advance...the standings would be
1. B 2-1 (advances on h2h)
2. C/D 2-1
3. D/C 1-2 (ahead on h2h)
4. A 1-2

Click to expand...

What I bolded is the only incorrect part. Losers 1st Round remains unchanged and the round robins only begin in Losers 2nd Round.

It does NOT double the length of losers, not even close. Normally someone plays one set for the chance to advance one round in losers. Under this format, you play three sets for the chance to advance two rounds in losers. That only adds seven extra sets total to the entire tourney.

It also does not decrease hype at all. We tried this yesterday at a tourney in Michigan and I think it's safe to say the format actually increased hype by a considerable amount. It pretty much guarantees competitive pools due to the way they're created, so the final couple round robins were extremely interesting to play out and/or watch for us.

And we already addressed the tiebreaker situation by making clear that:

Ties are broken by
1) number of sets won
2) head-to-head matchup
3) overall winning percentage in games

If still tied (only possible with a 3-way tie), each player will face off against their two opponents again in a one-game playoff. Make sure to keep track of stock counts; if there is yet another tie in sets and winning percentage, the player who loses the fewest stocks is the winner of the round-robin.

-Jumpman- said:

The idea is interesting, but flawed. A regular DE bracket is based around the number of sets won by each player. If a player goes to losers, he has to win two sets in GF, because otherwise he'd be able to win the tournament while winning just as much sets as the player who won WF.

As easily conjectured, the problem lies in the number of sets necessary to win the tournament. If a player wins the tournament by advancing through the losers bracket, which in your case is a combination of a bracket and a RR tournament, he has won more sets than any player who advanced to GF through winners. Thus factually speaking, this player has proven himself to be capable of beating more players than the person thought to be the best. My point: the idea is unfair for the players in losers bracket, for the winner is better than the best in WB, and the loser is disadvantaged more than what is currently considered fair.

I very much like the idea, but for the outcome of the tournament to be valid or fair to any degree, the winner's bracket needs to have the same format. Or... using a regular DE bracket.

Click to expand...

Your entire post makes no sense. Under a normal double-elim bracket, you also have to play several more sets in losers than you do in winners. Under this hybrid format, it's almost the exact same number of sets in losers for someone who is a threat to win the tournament unless they get eliminated in one of the very first rounds in winners, which is extremely unlikely. You say beating people in losers takes more skill than beating people in winners... that's completely wrong. Players who are in losers bracket have already lost a set in winners bracket; the average skill level in losers bracket is lower. Someone who beats more people in losers does not deserve anything more than someone who beats fewer people but stays alive in winners.

Rat said:

Ok I'm not 100% sold on this.

The main problem is after you lose a set in RR, it's pretty much pointless. You are guaranteed more matches (compared to being Knocked out right away.) Yet without the chance to move on any extra matches you play feels like a glorified friendly.

I'd almost rather be Knocked out, then have more time to play friendlies and MM than be forced to play a few extra matches.

There should be some incentive to fight for second seed.

I still want to try it a couple more times.

Click to expand...

A much more valid criticism of the hybrid format. We found that this is a non-issue in the later round robins (3rd and 4th round losers) due to the highly competitive nature of the round robin matchups - the bracket is designed to produce a group of four players who should be very close in skill level if seeding is done correctly. However, it can be an issue in the 2nd round losers round robin. It's true that players who lose their first set of a round robin might not completely throw in the towel, but even so, they don't have the same sense of urgency to advance like they would in a traditional double-elim bracket match. I think this can be solved in future hybrid format tourneys by having everyone play at the same time. Since there are four round robins in 2nd round losers, make each round robin play their first two sets at the same time. TOs could even go a step further and have pool monitors record results and hide them from the players until after the round robin is over, but I don't think that's quite necessary.

How is that any different than in a traditional double-elim bracket, where M2K would go to losers and have to face Amsah and Hungrybox anyway? Losers bracket has always been "more difficult" because you have to play more sets. This format effectively creates damage control for unfortunate matchups in losers by requiring a comprehensive round-robin for the chance to advance two traditional rounds. Yes, it comes at the expense of an extra set here and there, but with proper execution by the TO and proper understanding by all players involved, there is no groundbreaking chance in philosophy behind the bracket. The best and most deserving player almost assuredly comes out of each round-robin to advance two traditional rounds, and in the process it is extremely fun to watch a pool of evenly matched players fight it out.

I mostly ignored your first post because almost everything you said in it is wrong. If you really want me to break it down, I guess I can...

-Jumpman- said:

The idea is interesting, but flawed. A regular DE bracket is based around the number of sets won by each player. If a player goes to losers, he has to win two sets in GF, because otherwise he'd be able to win the tournament while winning just as much sets as the player who won WF.

Click to expand...

No it isn't. A regular double-elim bracket requires someone to win 6 straight sets in winners to place 1st, or 11 straight sets in losers to place 1st if they get knocked out early. You keep stating that the hybrid format is unfair to those in losers... yet under a traditional double-elim format, it's just as "unfair" to those in losers.

-Jumpman- said:

As easily conjectured, the problem lies in the number of sets necessary to win the tournament. If a player wins the tournament by advancing through the losers bracket, which in your case is a combination of a bracket and a RR tournament, he has won more sets than any player who advanced to GF through winners.

Click to expand...

There is no problem in the number of sets necessary to win the tournament. Someone who is a threat to win the entire tournament is either unaffected (because they make it to winners finals) or has to play one or two extra sets in losers. At most, they'd have to play three extra sets, but that scenario is extremely unlikely and can only happen if a player who is a threat to win the tournament is defeated in 1st round winners. 99.9% of the time, someone who is defeated in 1st round winners is not a threat to win the tournament, and is not qualified to worry about having to win three extra sets. The other .1% of the time either comes from awful seeding by the TO or a complete sandbag effort by the "top" player, and in that case it would be justified to force him/her to win extra sets in losers to get to grand finals.

-Jumpman- said:

Thus factually speaking, this player has proven himself to be capable of beating more players than the person thought to be the best. My point: the idea is unfair for the players in losers bracket, for the winner is better than the best in WB, and the loser is disadvantaged more than what is currently considered fair.

Click to expand...

A player who runs through losers is capable of beating more players, but not better players, so he/she does not deserve anything more than a player who survives through to winners finals. The idea is not unfair for players in losers bracket because they are playing on average one extra set in the entire tourney. Even in the absolute worst case scenario, they are playing 14 sets instead of 11 sets, which is a small difference.

-Jumpman- said:

I very much like the idea, but for the outcome of the tournament to be valid or fair to any degree, the winner's bracket needs to have the same format. Or... using a regular DE bracket.

Click to expand...

The hybrid format has been tested twice and TOs on both sides have all agreed that the hybrid format is more fair than a double-elim bracket. Placings are nearly impossible to argue against; results truly indicate who the best players are, and that's the entire purpose of hosting a tournament. Late matches in losers bracket are more hyped because they take place inside a round-robin that is inherently designed to be highly competitive. And there are fewer consequences to unfortunate character-specific or region-specific matchups in losers bracket because the round-robin requires a comprehensive test against a variety of playing styles, resulting in the best player advancing two "traditional" rounds.

Please don't post in my thread again unless you have valid points to discuss and can actually back them up with arguments. I don't mean to attack you personally, but based on what you've said in this thread so far, the only conclusion I can come to is that you simply don't understand how this hybrid format works.