Proposal Odd number of starting slots for WCoP going forward

[Lily]

So as much as I enjoyed this year's WCoP, it's no secret that it was REALLY long. Like, five months long.

I'd like to propose that going forward, WCoP uses an odd number of slots, either 7 or 9, for playoffs at least. I don't mind it in pools either but if that's logistically harder then it's no big deal.

This year was obviously pretty exceptional, between the 1 week delay due to the PS outage which was nobody's fault ofc and the fact that every single possible week that could have a tiebreaker did have a tiebreaker, but I think fixing the latter problem would be good since it's actually within our control. I played in a tiebreaker every week in playoffs and the fatigue is really overbearing. One of the great parts of WCoP relative to other official tournaments is that it's usually pretty short, but this year was anything but; it was slightly longer than even SPL was, which is kinda nuts.

I'll admit there are definitely some benefits to tiebreaks. They're entertaining to watch (although I'd also argue that it makes most of the week a snoozefest) and for someone like me who's still fairly up-and-coming it was beneficial to get experience in those high-pressure environments. But there are lots of issues too, between the fatigue for the tiebreak players and how much the hype is killed for other participating teams; I wasn't on team Spain for instance but I can't imagine it was very interesting for them to just have to wait for us to finish our tiebreak vs Germany. In SPL/SCL I don't really mind it as much because those tours are kind of expected to be way longer and they have a larger variety of tiers too, but in WCoP which is more rapid in general I think it's just a bit much.

Personally I'd prefer 9 starting slots but I understand it can be hard to fill up a roster for a lot of teams as is so I'm fine with 7. Would love to hear input on that, and would love to hear input from players on other teams who participated in tiebreaks/had to wait out another team's tiebreak to see how they feel etc.

 

[Finchinator]

Personally I think the format of playoffs should mirror the format utilized in the regular season; 8 is a good number for pools, and by extension feels like appropriate for the playoffs as the playoffs are meant to determine who the best team in any given tournament format is. I don’t love the differentiation in formats within the tour, especially when it comes to the biggest stage (if it must occur, absolutely go 9 rather than 7 though).

EDIT: if pools work with 9, then I have no issue with universalizing it then. Just don’t think format should differ.

Tiebreakers are also hype, but admittedly generate diminishing returns after each round forces another full week, doubling the elapsed time — perhaps a more prompt tiebreak with newly staggered deadlines could be in order.

I think that, to the OP’s credit, WCOP this year was a bit too long. A lot of people agree on this and I am glad this was posted. I think the best way to fix this would be to shift the dynamic surrounding qualifiers and the first round by either a slight format shift or numbers shift, potentially expanding the first round pool to limit the length of qualifiers while maintaining the length of the main tournament (meaning we could shave at least a week off of the cumulative length of the tournament). This also could give more people new opportunities.

Couple this with not wasting a week on PS outages, hopefully not having to juggle around a HOME/DLC release every year, etc. and we are looking at a much briefer tournament.

 

[blank]

I completely disagree with the OP and with some points in Finch's post.

First, it's completely illogical to say "SPL/SCL are expected to be long but WCoP should be short." Why? Even with these tiebreaks, WCoP is still a far shorter tournament than SPL/SCL for the average team. The only time WCoP is comparable in length to those tours is when your team is making a run from play-ins to playoffs... with tiebreaks.

Second, it is foolish to remove tiebreakers from a team tour. Tiebreaks have a lot of subjective value - they're fun, they're competitive, they're intense... really cool all around. Beyond that, when your team is down and out, they give your team further motivation. It isn't just win the week or bust - if your team has a rough start to the week, you can rally and pull for a tiebreak, something we see often in team tours. Given the nature of WCoP regions, tiebreaks also help to make less populated regions more competitive, as these regions are often top-heavy.

Third, the notion in Finchinator's post of "absolutely go with 9 rather than 7" is extremely silly. Sorry, not all of our regions have 10 million players like US East. Speaking for Midwest, we could barely field a roster of 8 starters lol... yet we're still one of two teams capable of making 4x playoff appearances in a row. I'm sure this is the same for other smaller regions, if not worse. It's already ridiculous to consider going to an odd number of slots, but to also act as if we need to go to a higher number? You're basically asking to go back to the WCoP of old - an uncompetitive disaster of a tournament.

WCoP has now had 4 very different winners in a row, two of which made it all the way from qualifiers to the trophy. Hell, even Spain this year had a relatively rough start at the very beginning of pools (grats to you guys btw, great run). Teams people think are shoe-ins to win it all can miss or make playoffs, teams people are counting out can do well. The tournament has proven itself to be very competitive for 4 years in a row now. Does that mean it's perfect? Of course not, but drastic changes like the ones discussed here are totally unnecessary when the tournament simply works.

The funniest part about this is that these frustrations are just completely misplaced. It is actually very annoying to be on a non-tiebreaker team, having to sit there and wait for others to play it out. The solution? Go back to shorter tiebreakers. TDs should take this thread and the rescheduling thread as wakeup calls to repair Smogon's chronic johning/fishing culture and go back to a more honorable time.

 

[Finchinator]

Third, the notion in Finchinator's post of "absolutely go with 9 rather than 7" is extremely silly. Sorry, not all of our regions have 10 million players like US East. Speaking for Midwest, we could barely field a roster of 8 starters lol... yet we're still one of two teams capable of making 4x playoff appearances in a row. I'm sure this is the same for other smaller regions, if not worse. It's already ridiculous to consider going to an odd number of slots, but to also act as if we need to go to a higher number? You're basically asking to go back to the WCoP of old - an uncompetitive disaster of a tournament.

We have not had an official with below 8 slots in over a century, so I am not sure how proposing to keep with that trajectory in our most inclusive format with the largest playerbase ever is "extremely silly". 7 is too few slots, 8 is the perfect amount of slots, and 9 is just better than 7. I prefer keeping it at 8 as I mentioned in my post.

I agree that not all regions have a ton of players, but I think the awesome results from some newcomer teams speak volumes as to how possible it is to grow playerbases and manufacture talent in WCOP settings. And shockingly: not everything is a shot at Midwest or a call to benefit my own team. We sucked this season and another slot was in no way going to shock that, so implying some bias is just bad posting etiquette on your end. I think we both agree that deviating from 8 is not optimal, but you managed to do so while implying I was bias and trying to prompt "an uncompetitive disaster of a tournament"...which is something lol

 

[ninjadog]

Get rid of tiebreaks in quarter finals, keep them for semis and finals. QFs have the greatest possibility for tbs, the most teams that would be delayed by it, and the least hype.

Also it absolutely should be 7 slots, especially given that 7 slots is literally being used in qualifiers to avoid tiebreaks. expanding it just gives an advantage to the teams with bigger player pools to pick from (who are already advantaged by having said larger population).


Shorter tiebreak deadlines also just aren’t realistic because shockingly not everyone lives in the same timezone, and the player base is on average a lot older than it used to be so people have less time.

 

[Luigi]

earlier today lyd joked in discord that one random matchup should be worth 2 wins, which is obviously a silly system BUT

what if the teams picked their best player to go in the Champion Slot™ and that's worth 2 points? Should keep the regular 8 games and TB for the finals cause it's iconic and there are no teams being held up.

Pros:

• Preserves 8 slots
• You get guaranteed tiebreak quality matches every round, for every matchup.
• Solves the time constraints

Cons:

• It's kind of arbitrary and there might be teams that have a roster without a standout that would be hurt by this, but like, skill issue? such a team would be in an even worse spot in a TB anyway.
• Does it count as 2 wins/losses on the sheet?
• Gets stressful for each team's best player as the tournament goes on, but those are people that are accustomed to high pressure games anyway, and it's less stressful than a TB game.

Also, entirely separate but since we're addressing how long this WCoP was and a shorter tournament is desired, round 1 should be 2 weeks long. A grand total of 5/192 games got played in week 1 this year. Two weeks is ample time to make 3 teams.

 

[Dj Breloominati♬]

Please do not remove tiebreaks. The point of fatigue for the tiebreak players is valid, but the other side of the coin also needs to be seriously considered before any decision is made. A lot of the time, trailing teams on the cusp of playoff qualification try and realistically aim for tiebreak atleast, because it offers them a fresh slate ( to start again in a sense - you can reset your mental from bad gameplay / unlucky rng ). I say so being from such a region myself (India) - the latter half of Wcop this year was us focusing on trying to hit the magic number of 13-11/12-12 (just taking the example of wcop quals here), and we would be missing out on the entire experience if not for tiebreaks. Another more recent example is RoW coming back from a 1-4 deficit to reach tiebreak and ultimately beat US East in UUWC Quarterfinals. These are just some of many examples but yeah, tiebreaks are a blessing to the "weaker" teams, they keep us in the race and also provide "experience in high-pressure environments" as lily mentioned ( and this experience can't really be substituted by other tours given wcop is the "grandest" tour for a lot of us )

The pros outweigh the cons, in my eyes.

Tldr : Tiebreaks give trailing teams a chance to get back in the competition/round and promotes better gameplay overall, because there's still something to play for a lot of the time.

 

[Isa]

change wcop to 9 players (or 7, idc which)

this automatically removes tiebreaks in the section where one is not needed (playoffs) while keeping them in the section where strength of schedule could have a higher negative impact on teams and thus serves as a small "equalizer" (group stage)
no tiebreaks in playoffs means better tournament flow and more predictable tournament length. one could easily imagine a player missing finals because wcop suddenly ended three weeks later than intended; if you knew this when drafting your team, this could have changed your squad selection at the start of the tournament. predictability matters.
smogon barely has an off-season right now and player fatigue is a real thing as several players have alluded to over the years. cutting wcop down by three weeks would help here too by allowing more space in the calendar.

spl and scl should keep even numbered slots because ties are important to the outcome of the group stage, but wcop does not play by the same rules. make wcop better.

 

[Chaitanya]

Idc much about the rest of it but one slot counting for more than the others is a meme for sure. Main thing I wanted to say tho is 3 weeks for pools is def a waste. Everyone will john to the end regardless of how long pools are and the 2 week rule to play a game is a huge meme. 2 weeks is plenty to play 3 games. Goes for the lower tier world cups that do pools too.

 

[Amaranth]

WCoP main event lasts 10 weeks total. 6 of them being actual tournament, and 4 of them waiting for 3 games to happen in 7 days each time. This is blatantly unacceptable from a logistics/timing point of view. A solution to speed up the timing is mandatory IMO.
Given that this is a logistic issue and not a competitiveness issue, it would be weird to address it by changing the actual format of the tournament. There is nothing wrong with tiebreakers as a system for determining which team advances - it's only the logistics that need addressing.

I would be interested in Saturday deadlines for regular rounds with Sunday spared for potential tiebreaks. It feels logistically viable to run any eventual tiebreaks Sunday-Tuesday at most, even accounting for the existence of Oceania.
For easing logistics, I also think it should become standard for teams to submit 3 players with a description of their full availability so that any incompatibilities within the shorter Sun-Tue deadlines can be avoided by... just not pairing the two incompatible people together.

Preserving the function of the tiebreaks and Just Speeding Them Up feels like the way to go imo. Keep the full extra week for any finals tiebreaker if it gets there.

 

[Star]

I think Saturday deadlines for regular rounds is a truly terrible idea tbh (as is shorter tb deadlines in general). Weekend is often the only time ppls availabilities even line up given the timezone differences in WCoP and taking away one of those days completely is literally guaranteed to lead to problems. Not to mention that running the tiebreak Sunday - Tuesday is quite often just forcing ppl to play it Sunday and making ppls seasons come down to one day of prep after weeks of the tournament is just extremely dumb.

Tiebreaks are the most high-stakes games we have in team tours and I’m personally very against any idea that eliminates them bc I think they’re both rewarding for players and fun for spectators, speaking as someone who’s been in both positions. I get people feeling this WCoP went on for too long, but having a tiebreak in literally every possible round is an exception, not the norm. I really do not think the above ideas of making quarters 7 or 9 slots for a literal one week POSSIBLE reduction in tour length is remotely worth it. Playoffs should always be the full teams battling it out and I think people can cope with waiting a week if needed as they have done for the past decade in every team tournament.

 

[Amaranth]

having a tiebreak in literally every possible round is an exception, not the norm.

if every game is a 50/50, the odds of a 4-4 over 8 games is 27.34%. this means you should expect a quarterfinal tiebreak week much more often than not, a semifinal tiebreak week almost half the time, and a final tiebreak week over 1 in 4 times.
over the last four years, in 12 total rounds (quarters+semis+final x4), we have seen 6 of these rounds go to tiebreaker.

this is not a "possible" reduction. it's a mathematically expected reduction of 1.5 weeks on the playoffs alone, which you can round up to 2 weeks when you include the likelihood of a tb at the end of round 1 as well. the math is rough around the edges but it's pretty clear to me that saving about 2 weeks of waiting, every year, on average, is VERY worth the effort.
e: i guess slightly under two weeks if you consider that you'd keep the full week for any finals tb, but the general point stands


re: the rest of the post. I think taking away sundays is nowhere near "guaranteed to lead to problems", and I think "making ppls seasons come down to one day of prep" is disingenuous when your team has literally already prepped for the opponent during the regular series. I think you are severely overestimating the dangers/drawbacks of shortening tiebreak time, and severely underestimating the logistical benefits of doing it.

 

[Nat]

I feel like this thread is going to go in many different directions (and already has) but one thing I wanted to touch on is I'm very much against shorter tiebreaks. Even if someone wouldn't necessarily know from experience about the prep that goes into TB being separate from prior prep about an opponent, finding times with someone in a 1 day weekend slot can easily be impossible. Scheduling issues over a 1 week period are all too common, and there's no subs in tiebreaks as of now. I have trouble scheduling with ppl from west, and i'm sure other regions have similar issues within themselves.

This could be exacerbated further when a round could end 9pm -4 saturday, and people in certain regions don't even know if they'll have to be available for TB or not until they wake up that day sunday, and then are busy monday/tuesday. idk, just seems like a recipe for disaster.

 

[Finchinator]

Please do not take away Sunday from any week, especially not playoff weeks. It is already challenging for anyone with a job to schedule with people with big timezone differences, but taking away the most active weekend day will hurt matters a lot. A Sunday (or Monday/Tuesday) tiebreak is not even a good solution as is with all of the potential uncertainty leading up; this also dwarfs the hype and potentially the spectator component.

After a lot of discord discussion, I think keeping tiebreaks as is is best. You run into a pickle between maximizing quality for the sake of sacrificing time or sacrificing quality for the sake of saving time; in a tiebreak setting, you should always go with the latter, allowing the best competitors to participate even if they can’t find a perfect schedule within the first 3 or 5 days as weekend almost always works out.

I really think keeping WCOP 8v8 all tour is best. I really think keeping playoffs a full week and tiebreaks a full week is best.

We can improve on the margins by looking at qualifiers and even the initial field. Hopefully we do not have any delays with releases or outages as well. Tiebreaks are very possible and inevitable, but they are not what we should be cutting down on at this point. Playoffs are a part of every tournament and compromising a full week to play shouldn’t be on the table.

 

[lydian]

I should note I'm commenting only as a spectator, and ofc the input of the players should be held at a higher value, but I wanted to touch on some things:

On the importance of tiebreakers:​

ABR brought up an important point in the discord, that tiebreaks help make the last game matter more often, as you have two possible outcomes that would lead to the last game mattering: 3-4 and 4-3, as opposed to just one with 7 or 9 players: 3-3 or 4-4, respectively. I think it's important to preserve this because it makes series come down to the wire more often, which makes the tour much more exciting and engaging. I wrote some python code to showcase this:

Results

7 players (assuming 50-50 odds for each game):​

4-0: 6.25%
4-1: 12.50%
4-2: 15.62%
4-3: 15.62%
3-4: 15.62%
2-4: 15.62%
1-4: 12.50%
0-4: 6.25%
3 games not played: 12.50%
2 games not played: 25.00%
1 game not played: 31.25%
0 games not played: 31.25%

8 players (assuming 50-50 odds for each game):​

5-0: 3.12%
5-1: 7.81%
5-2: 11.72%
5-3: 13.67%
4-4: 27.34%
3-5: 13.67%
2-5: 11.72%
1-5: 7.81%
0-5: 3.12%
3 games not played: 6.25%
2 games not played: 15.62%
1 game not played: 23.44%
0 games not played: 54.69%

9 players (assuming 50-50 odds for each game):​

5-0: 3.12%
5-1: 7.81%
5-2: 11.72%
5-3: 13.67%
5-4: 13.67%
4-5: 13.67%
3-5: 13.67%
2-5: 11.72%
1-5: 7.81%
0-5: 3.12%
4 games not played: 6.25%
3 games not played: 15.62%
2 games not played: 23.44%
1 game not played: 27.34%
0 games not played: 27.34%

7 players (assuming 66.7-33.3 odds for each game):​

4-0: 19.75%
4-1: 26.34%
4-2: 21.95%
4-3: 14.63%
3-4: 7.32%
2-4: 5.49%
1-4: 3.29%
0-4: 1.23%
3 games not played: 20.99%
2 games not played: 29.63%
1 game not played: 27.43%
0 games not played: 21.95%

8 players (assuming 66.7-33.3 odds for each game):​

5-0: 13.17%
5-1: 21.95%
5-2: 21.95%
5-3: 17.07%
4-4: 17.07%
3-5: 4.27%
2-5: 2.74%
1-5: 1.37%
0-5: 0.41%
3 games not played: 13.58%
2 games not played: 23.32%
1 game not played: 24.69%
0 games not played: 38.41%

9 players (assuming 66.7-33.3 odds for each game):​

5-0: 13.17%
5-1: 21.95%
5-2: 21.95%
5-3: 17.07%
5-4: 11.38%
4-5: 5.69%
3-5: 4.27%
2-5: 2.74%
1-5: 1.37%
0-5: 0.41%
4 games not played: 13.58%
3 games not played: 23.32%
2 games not played: 24.69%
1 game not played: 21.34%
0 games not played: 17.07%

Code

import math
import sys

def simulateMatch(scores, team1Strength, team2Strength, winReq):
    newscores = {}
    for key in scores.keys():
        if max(key) < winReq:
            if (key[0]+1, key[1]) in newscores.keys():
                newscores[(key[0]+1, key[1])] += scores[key]*team1Strength
            else:
                newscores[(key[0]+1, key[1])] = scores[key]*team1Strength
         
            if (key[0], key[1]+1) in newscores.keys():
                newscores[(key[0], key[1]+1)] += scores[key]*team2Strength
            else:
                newscores[(key[0], key[1]+1)] = scores[key]*team2Strength
        else:
            newscores[key] = scores[key]*(team1Strength+team2Strength)
    return newscores

def simulateRound(playerCount, team1Strength, team2Strength):
    winReq = int(playerCount/2) + 1
    scores = {(0,0): 1}
    notPlayedDict = {}
    for i in range(playerCount):
        scores = simulateMatch(scores, team1Strength, team2Strength, winReq)
 
    for key in scores.keys():
        notPlayed = playerCount - sum(key)
        if notPlayed in notPlayedDict.keys():
            notPlayedDict[notPlayed] += scores[key]
        else:
            notPlayedDict[notPlayed] = scores[key]
        print(str(key[0]) + '-' + str(key[1]) + ': ' + '{0:.2f}'.format(100*scores[key]/(team1Strength+team2Strength)**playerCount) + '%')
 
    for key in notPlayedDict.keys():
        if key == 1:
            print("1 game not played:", '{0:.2f}'.format(100*notPlayedDict[key]/(team1Strength+team2Strength)**playerCount) + '%')
        else:
            print(key, "games not played:", '{0:.2f}'.format(100*notPlayedDict[key]/(team1Strength+team2Strength)**playerCount) + '%')

def check(value, name):
    try:
        value = int(value)
        if value > 0:
            return value
        else:
            print(name, "is not a stricly positive integer")
    except:
        print(name, "is not a stricly positive integer")
    return False

if len(sys.argv) == 4:
    playerCount = check(sys.argv[1], "playerCount")
    if playerCount == False:
        sys.exit(1)
    team1Strength = check(sys.argv[2], "team1Strength")
    if team1Strength == False:
        sys.exit(1)
    team2Strength = check(sys.argv[3], "team2Strength")
    if team2Strength == False:
        sys.exit(1)
 
    teamGCD = math.gcd(team1Strength, team2Strength)
    if teamGCD != 1:
        team1Strength = int(team1Strength/teamGCD)
        team2Strength = int(team2Strength/teamGCD)
    simulateRound(playerCount, team1Strength, team2Strength)
else:
    print("expected arguments: playerCount team1Strength team2Strength")

As you can see, having 8 players and a chance to tie about doubles the chances that all games matter and are played at tournament stakes, which is a huge improvement! And this seems to hold true even if teams are evenly matched (assuming 50-50 odds) or if one team is significantly stronger than the other (assuming 66.7-33.3 odds).

On top of that, after some reflection, tiebreakers feel to me absolutely worth the extra hassle (be it logistic or waiting), because they're very hype and lead to top player matchups. Maybe it could be considered to remove them just from quarterfinals and possibly semis too, but I agree with Finchinator it seems incredibly weird to switch the format (be it number of players or any other solutions to avoid tiebreakers) mid-tournament, especially so mid-playoffs. So I think we should try to find a way to maintain tiebreakers in some nature.

If we do deem that tiebreakers need to go:​

I would be disappointed with this outcome, but if it had to happen, I think changing the player pool to an odd number would be the least bad option. Probably 7 as many smaller teams claim they'll be disproportionally affected by an increased number of starting slots, and it should apply to the whole tournament, changing the number of starters mid-tournament would be bad, and especially so if it increases (i.e. 8 -> 9) because in that case managers would need to purposefully sub players in so that they get playing experience and aren't thrown directly into a playoff. Pools stage should still work so long as the number of teams and players per pool isn't changed.

The option I jokingly suggested on Discord that apparently is picking up traction (have a 2-point slot) would be "hype" and help address the issue of tiebreakers often having stronger players match up against eachother, and in a way you could see it as a "pre-emptive tiebreak" (maybe you could have three slots worth 2-points?). But it still feels uncompetitive and unfair, and it would absolutely trivialize several matchups, as it'd be incredibly hard to come back from losing a double-pointer game. Maybe someone could test this in a tier PL and see how the community receives it, but I think this idea is too unserious for a tournament like WCoP.

On the 3-week pools:​

This is off-topic for the thread, but it's being discussed here so I want to leave my two-cents. As is, I don't think it makes sense to have three week pools when so few games happen in week 1. I think you should either decrease it to two weeks and continue to enforce the "you must schedule 1 game for before the last week" rule, or, my preferred option, assign a week for each pairing. So A vs. B and C vs. D in week 1, then A vs. C and B vs. D in week 2, and so on... This helps spread out the games and also leaves everyone at a fair advantage of you and your opponent having played the same number of games up until that point. In the current system, it's possible for someone who's played 0 games to go up against someone who's played 2, and one player has a clear advantage in scouting and prepping in this situation. I've heard counterarguments for this that it doesn't give teams enough time to prep and grow accustomed to their setting, but that hasn't stopped SPL or SCL from using this format.

In conclusion:​

I strongly prefer tiebreakers to be kept for the reasons I outlined above (mostly that they make the games actually matter much more often, thus maintaining the level of competitivity high throughout the week), we can either maintain the status quo, or try to find some other possibility to expedite the process, I'm not sure on that yet so I'll leave that up to people with more experience hosting, managing, and playing in tournaments to decide (although I do think pairing people based on availability could be problematic for several reasons, like schedules changing, transparency, and potential issues arising when there is more than one viable option).