✔️

HARD AGREE, the 3rd vs 2nd match to qualify for the final was a way better format

✔️

tick

.

This format leaves more room for upsets, like in season 28 where the fourth place seed Ascent.EU beat Lowpander and Lego and ran Svift really close in the grand finals. Although I realize thats rare, imo thats worth the delay.

unless those teams were forced to use ringers why is this inherently a bad thing?

This is the first time I've even heard people in public actually say that the current playoff structure is bad or boring at all hence why we haven't even considered on any change yet beforehand, but I do agree with the points you've listed here except for the double elimination itself, but I'm not exactly defending it either.

Pretty sure the implementation of the Double elimination in the first place was done to make the playoffs less anti-climatic where the 1st seed team was already a guaranteed winner pretty much (duh) regardless of the 2nd/3rd seed team having just battled out for a spot in the grand finals and being as warm as they could be, can't remember these things exactly from 3 years back. However I do agree that the current format does drag on for too long and changing up things for it would make things at least a bit more fresh so the playoffs won't need to last for 3+ weeks.

Something that could work in place of this would be to have a playoff structure similar to Season 23 instead of Season 21, where 4th seed team was also taken into the playoffs and had to play their match before the semi-finals and grand finals, in which case the 4th seed team won the 3rd seed in this example. I'm aware however that this can't happen in every season and this was also the only one where this did happen with single elimination.

❌ I am enjoying the current system other then that it drags on for quite a while.
I like having the feeling of a lower bracket where you have a second chance & play against the losers from the previous one. (I've not experienced the other way of playoffs so I guess my opinion doesn't really matter at the end of the day:))

I disagree with most points made here
"It's boring" : It's more matches of top-level TF2 action, which is always a big + personally. It also adds the opportunity to build a narrative, with the first games being like a appetizer, the upper bracket final a potential preview of the grand finals, and every game in the lower bracket being a fight for survival.
"makes the season irrelevant" : In my opinion, the better team should win regardless of any advantage given but I understand that there need to be some sort of incentive for finishing first and being the first to reach the Grand Final. The BO4 could be re-instated, I didn't really mind it or perhaps we could say that the picks and bans are always determined using the finishing positions of the regular season, except for the Grand Final where it would be the first to reach it. From a purely spectator standpoint, I never felt that the top seed needed any sort of advantage but this is obviously an issue on the players' side so my opinion isn't really that important.
Bloodis covered the part about double elim pretty well, so I'll just conclude by saying that the s21 format was interesting when the top 3 teams are all quite close, but recent seasons have less of that, and removing very entertaining games like between 3 and 4 would be a shame.

Quick disclaimer : I only really watch TF2 (or any other sport) for the narratives that build themselves and I understand that other people might enjoy watching it for other reasons.

Under Starkie's system we wouldn't have had https://gyazo.com/09bca92183f9bf677162fc409c147258

The only thing I really agree with is how long playoffs take to be played, but that could be fixed by giving teams a shorter time frame. Right now playoffs goes from March 17 – April 7th. That's three weeks, and make teams treat playoffs as regular season games, playing one playoff a week.
I'm aware my team is one of the culprits here, but I think this is the main issue atm. Even shortening it by a week would make for a much more dynamic system.

As someone who has played in both the old and new system I vastly prefer the way things happened before. I'm speaking for myself at the very least when I can say that having to play potentially 3 more bo3s and a bo5 and all the scrims on top of that actually produces worse games to watch because I feel burnt out at the end of week 7 and all this on top is honestly a chore that is not needed.

Double elim is cringe, agreed. No reason to double elim it in prem/div 1 when already played a season. However, being boring/the season meaning nothing isnt true. It isnt as boring if you actually are pushed to try. If you are on a team that just rolls their games and doesnt really need to be pushed to try that hard once a week, and scrim fairly hard, then i see how it is boring, and how the season is irrelevant. I guarantee for svift, and faint (and also the 5th-7th place teams) the season was not irrelevant.

give higher seed a 1 map advantage again that was cool

If you want a compromise you can soften up the double elimination a bit. 4 team double elim requires 6 matches, but the frist two matches are usually garbage. Map pick advantage pales in comparision to #1 playing #4, so it's usually only a formality, while #2 vs #3 can be close but all that #2 gets for being better than #3 all season long is getting one more map pick. Beyond that there's no advantage for having the higher seed, you still need to play the same number of matches.

So other than reducing it to 3 teams you can use an asymmetric bracket. E.g.
A: #3 vs #4, loser is eliminated (so not strictly double elimination), skipping straight to what's usually the Lower Bracket Round 1.
B: #1 vs #2, essentially what will almost always happen in the Upper Bracket Finals anyway, but now #2 is guaranteed a shot at this for being better than #3 in the season.
C: Winner of A vs loser of B, basically Lower Bracket Finals
D: Winner of B vs winner of C, Grand Finals
Advantages in B and D can be adjusted depending on how badly you want #1 to win.

That's only one of the options. 4 teams, but only 4 matches. #3 and #4 are eliminated after one loss, #1 and #2 after two. #4 doesn't have to play #1 in their first match though and #3 plays a lower seed first instead of a higher one so their chances are not significantly worse than in true double elimination. If "higher seed beats lower seed" holds true you get the exact same matches as with double elimination, only two less.

Playing every thing in a day can make it end very late and also your system doesn't allow the 4th team to participate in the playoffs.
You could allocate less time for playoffs (as ombrack said) and force teams to play within this period of time to make it less boring.
I think we can fuse the s21 idea and the current system as:

Game A 1 vs 2 (with map pick/ban advantage to 1)
Game B 3 vs 4 (with map pick/ban advantage to 3)

Game A winner goes to final (and has pick/ban advantage in the finals)
Game B looser is out
Game B winner plays Game A looser in Game C
Game C winner goes to final

This format can be made not boring. You can force all games to be played in 1 week. A and B on day 1, C on day 2 and final on day 3, days do not need to follow each other. We can even now merge onto your original idea and play game C and final on the same day.
This makes the season relevant because ending at the top of the tables gives you advantages in the picks and/or a direct chance to reach finals.

edit: heh, pretty much Setsul's idea but 10 minutes too late...

I prefer only 3 teams in playoffs. 4 teams is usually more than 50% of the whole div considering how many fold now

Why do you think this is problematic ?

i'll try and make my point clearer on double elim because i was being brief in the first post

the only fair way to do double elim is for the final to be 2 bo3's as the upper bracket team hasn't lost a match. if it's 1 bo3 (or bo5) then if the lower bracket team wins both teams have lost 1 match but the lower bracket team are the winners. a random map advantage or map pick advantage just seems arbitrary. i understand why it's a necessary evil to lengthen tournaments for LAN but it's not required here as there's already been 28 season matches.

so a double elim final is either incredibly lengthy (and one-sided) or it lacks competitive integrity

also i thought bloodis' suggestion was a good idea. personally i prefer mine since it makes the season a proper battle for teams contending for 3rd but his system makes a miracle 4th place run possible

whats the point of playing 7 weeks just to eliminate 3 or 4 teams? every league ever has a minority of teams making playoffs, not half or more

Ok, I get your point that winning game A doesn't give the team as much advantage as it should.
Isn't it possible to properly balance the map advantages, by, for example, allowing the winning team to pick and ban first ?

Also,

So, both systems are valid and whether s21 or s23 is the best really depends on which teams are in. I don't know if this is actually doable, but maybe the playoffs system/structure should be decided depending on how close the 3rd/4th places are...

That's incorrect.
1. It's still not fair. 3-2, 0-3, 3-2 means the first team wins with a map score of 3-5 or 6-7 total, both clearly negative. If the second match is a Bo3, Bo5, Bo7 or even Bo9 the second team wins. Instead the first team effectively gets an arbitrary 3 map advantage. Is that better?
2. Double elimination is never fair. If two teams meet only twice then the team that wins the second match, the one in the losers bracket, advances and the other is eliminated. Even if it's 1-1 in matches. The second match always counts more. That's how double elimination works.
That's why match length generelly increases in later rounds.

Seems to me that you don't understand double elimination brackets. I guess the "unfairness" of the losers bracket match is easily overlooked if you never have to play one.

Playing two games in the Grand Finals is usually used for competitions with fixed match length. E.g. if everything else was Bo1 then two Bo1s are played because playing three or four halves sounds retarded and is in some games not even possible.

Try to think about that please.
Because "#1 needs to win 1 match, everyone else needs to win 2" is an awfully simplistic and #1 centered view. By your own logic it also lacks "competitive integrity" because #2 gets no advantage over #3.

ive read your post multiple times and i still dont understand what youve said

I'll try to make it as simple as possible:
In double elimination brackets if two teams play each other twice the first match doesn't matter.
Whoever wins the second match advances.
That's how double elimination works.
When two teams meet in the Lower Bracket Finals it doesn't matter one bit if they've played against each other before and who won. Whoever wins goes to the Grand Finals, whoever loses is out.

Do you understand that?

If you got that then you should understand why most double elimination brackets use e.g. Bo1 for the first round, Bo3 for the second and Bo5 for the final.
Winning Bo5 > Winning Bo3.

Your 3 arguments are all the same "why isn't #1 the winner after winning one playoff match?".
Gauntlet/stepladder: #1 has to play one match ✔️
Everything else: #1 has to play more than one match ❌

Now the complicated part:
You complain about any advantage #1 could be given as "random" and "arbitrary", claim that two Bo3s for the Grand Final, one of the less fair options, is in fact the only fair option and complain that if the the exact same rules apply to #1 as to everyone else (first match doesn't matter, second match is longer and/or higher seed gets advantage) then double elimination "lacks competitive integrity".

I'd try to explain again why you generally play two matches in the GF if all other matches were Bo1 and why Bo5 > 2xBo3 but it'll probably just go over your head again.

i gave you the benefit of the doubt that maybe you were just coming off as a condescending twat unintentionally in your first post but now you've confirmed it

What do you mean by most? If there's a scene in which double elim is more prevalent than FGC, I've yet to see it, and bracket reset in GF is a pivotal feature of almost every FGC tournament. The principle reason other scenes don't use this format is because it is not pragmatic due to length of sets.

Well what do you want me to do?
You said you didn't understand anything so either you weren't trying or I have to make it simpler to understand.

Dota, SC, although there's obvious bias towards tournaments where scaling works. It makes more sense to scale both single and double elimination. Also double elimination groups with Bo3 for everything. No bracket reset, no scaling, only the second match matters.

I guess when you're starting with a Ro256 scaling doesn't really work (edit: yeah, just saw your edit and you guessed it) and at that point it's more consistent to keep everything Bo3 and use that as winners bracket advantage instead of doing Bo6 or having no winners bracket advantage at all.

so what? if the two teams meet again then that means that they both have 1 loss. of course the loser is eliminated?? 2 losses = DOUBLE elimination??? who cares if they played against each other before, one of them now lost 2 games and is out.
unless you're talking about the grand final which is exactly why stark says you need 2x bo3