Implemented changing the maximum amount of byes

[TheFranklin]

Recently there have been a number of tours where a significant amount of people could not participate despite wanting to, simply because the current rules with respect to maximum percentage of byes did not allow it.
For example: https://www.smogon.com/forums/threads/the-nu-open-xi-round-1.3699736/
The tour started with a huge sublist being close to 1/4 of the sign ups. A lot were still be subbed in, but in the end still 50 people were denied the chance to play in the tournament.
Similar situations occured during ST where about 20 people couldn't play: https://www.smogon.com/forums/threads/season-33-week-1-ss-ou-3-won-by-yelodash.3699046/ and https://www.smogon.com/forums/threads/season-33-week-3-oras-ou-2-won-by-z0mog.3699641/

In my opinion we should enable as many people to play as possible. With the current policy of a maximum of 10% of the total bracket being byes, it is highly likely that similar situations will repeat in the future.

My solution would be to increase the maximum percantage of byes from 10% to 15%.

 

[sugar ovens]

can't we just get *everybody* in with some of the players having to play two matches in a round, the first one having a shorter deadline, similarly to how extensions are done? in case of non-official tours prolong the first round to two weeks if needed

i genuinely don't see any reason why is this not done this way. if you happen to not be able to play your match in a week - well that sucks, but like every other round will be only one week long so. what is a bigger blow to your competitiveness and prestige - randomly remove a part of the playerbase or give some players the advantage equivalent to a single activity win/flip that people get all the time?

mildly inconvenience a couple of people or remove possibly over a hundred players from the tournament, what sounds better to you?

 

[Wigglytuff]

My solution would be to increase the maximum percantage of byes from 10% to 15%.

The 10% number currently used seeks to limit the # of players that receive a bye in R1 while also attempting to avoid situations like this year's NU Open, where a large number of players aren't allowed in due to bracket size limitations. How 10% was arrived at, I don't know, but I suspect it's arbitrary. Arbitrarily increasing it (in this case, to 15%) doesn't guarantee that this year's NU Open won't happen again.

boring math sh*t

I did the lavoscalc on the numbers: for a 512 size bracket, a 15% sub limit would be 76.8, or 76 rounded down. That means if less than 437 players sign up, then the bracket size will have to size down to 384 and up to 53 players will miss out on the tournament, or 14% of players. For reference, this year's NU Open (256 size bracket) allegedly had 1/4 (the current thread sublist has added people who signed up after R1 went up so calculating it now would be inaccurate)

Unless the TDs are willing to entertain rearbitrating an acceptable % of players that miss out that is balanced against an acceptable % of players that receive a Bye R1, then we shouldn't bother too much with the boring math sh*t.


What should the cutoff % of Byes in R1 be increased to?
The absolute range of % of Byes in R1 is 0-25%. 0% is obviously undesirable and more than 25% means that Byes will face Byes in R1, so there will still be Byes in r2. Given that Slam/Classic tend to be 256 - 512 size brackets, creating the possibility of someone to win the tournament while only having won upwards of 3/4s the series of someone else would raise some serious questions about competitive integrity.

In light of money prizes seeming to be the wave for Slam/Classic Opens/Cups, the cutoff should be increased to 25%. With money coming into play, dicking a large portion of the playerbase out of the chance to play and win is far more distasteful than the idea of more players getting an R1 Bye. I don't see people complaining about the winner of the last official having received a bye so keeping that status quo is probably fine.

Increasing to 15% will only last until the next time an NU Open moment happens and then the process will repeat (rightfully) due to a large # of players missing out.

Activity wins/coinflips are last resort for hosts (or at least they should be) and should be considered separately from bracket size. There are things that can be done to try to decrease the number of AWs and coinflips that would directly do nothing to address bracket size problems.

 

[Boat]

Gonna bump this, cause I think we can kill two birds with one stone here.

You can solve the above issue by increasing the maximum bye % to 25%, and that makes it very, very easy to remove RR finals in their entirety.

https://www.smogon.com/forums/threa...in-non-live-tournaments.3665073/#post-8663952

 

[Lalaya]

bump on the proposal to make it 25% (yes this happens every smogtour)
I largely agree with what said above, and even if we can't prevent situations such as having to drop anywhere from 10 to 30-40 people from a tour depending on size, we can make it easier without compromising the integrity of said tour and quite frankly I haven't saw any drawback in doing so (especially since it also means getting rid of 3-way finals which can reset and prolong a tour for no reason)

 

[dhelmise]

Gonna bump this, cause I think we can kill two birds with one stone here.

You can solve the above issue by increasing the maximum bye % to 25%, and that makes it very, very easy to remove RR finals in their entirety.

https://www.smogon.com/forums/threa...in-non-live-tournaments.3665073/#post-8663952

I know the focus of this thread and the one linked in this post is on live tours and single elimination tours, but I'd like to bump this again because RoA almost hit a round robin tour number for the RBY seasonal, which is gross in its own way, but we actually missed the threshold by 3 which means we have a 128-man with 42 subs (unless more people sign up after this post...?). Neither solution is really friendly to the people who signed up nor the host, and raising the bye % would at least solve issues like the NU open situation from last year of having a large number of subs in double elim tournaments.

 

[Amaranth]

I know the focus of this thread and the one linked in this post is on live tours and single elimination tours, but I'd like to bump this again because RoA almost hit a round robin tour number for the RBY seasonal, which is gross in its own way, but we actually missed the threshold by 3 which means we have a 128-man with 42 subs (unless more people sign up after this post...?). Neither solution is really friendly to the people who signed up nor the host, and raising the bye % would at least solve issues like the NU open situation from last year of having a large number of subs in double elim tournaments.

For the RBY seasonal, and any upcoming double elimination tournaments, we will be attempting a slightly different approach.
Double Eliminations currently follow the guidelines outlined here. While they have been mostly effective, they have some edge cases - such as this RBY Seasonal - where a lot of people would be forced to be substitutes.
We are adjusting the threshholds for maximum subs to be 1/4th of bracket size, instead of 1/2. This means that the threshholds will be adjusted as follows:
Size 16: 11 to 20 sign ups
Size 32: 21 to 40 sign ups
Size 64: 41 to 80 sign ups
Size 128: 81 to 160 sign ups
Size 256: 161 to 320 sign ups
Size 512: 321 to 640 sign ups

Note that this applies to double elimination only, and is a tentative change that will be given a few trial runs - we may revert to earlier values if they are found to be preferred.
The new values can result in some scenarios with a lot of byes in the bracket, but we think they are worth experimenting with to reduce the number of people who don't get a chance to play in the tournament at all. We'll be receptive to both players and hosts about their experiences with these bracket sizes.

A post on adjusting bye threshholds for single elimination tournaments is in the works and should be coming soon, too.

 

[Merritt]

As we go into a new year of tournaments, the TD team will be adjusting the maximum bye percentage for single elimination tournaments to 25%. What this means is that up to 25% of the participants in Round 1 of a tournament can be byes.

By increasing the bye percentage to 25%, this means that Round 1 of tournaments will generally not have an initial substitute list (or if it does, the substitute list will be relatively small), as all players who initially signed up will be paired with either an opponent or a bye for Round 1 for most sizes. We are not currently pairing this with a removal of round robin final sized brackets for single elimination tournaments as mentioned in Boat's proposal, as that would increase the potential number of players who are put on the substitute list for Round 1, which is counterintuitive to this change. We will potentially revisit this in the future, particularly after seeing how the double elimination tournament size adjustment described in the post above goes.

This change will go live in the bracketmaker today.