Tournament Pre-OMPL XIII Thread

Status
Not open for further replies.
UT

UT

I don't like that falling feels like flying
is a Tournament Directoris a Site Content Manageris a Battle Simulator Administratoris a Top Social Media Contributoris a Member of Senior Staffis a Top Community Contributoris a Top Smogon Discord Contributoris a Top Contributoris a Top Smogon Media Contributoris a Dedicated Tournament Hostis a Top Team Rater Alumnusis a Top Metagame Resource Contributor Alumnusis a Tiering Contributor Alumnus
Appeals + PS Admin
With the tiers now set and considering how much fun last year's was, we're doing it again!

For a refresher, last year's admin thread can be found here with the changes specific to OMPL XII called out here. With one notable exception, by default we will be running with the rules and procedures we used last year, but we're happy to entertain changes and improvements if people have suggestions!

Things that are not up for discussion: number of teams (eight), the tiers (AAA1 / AAA2 / BH / GG / MnM / PiC / STAB / bo3), no trades, no retains.

The main thing that will need to change, as is tradition, is the tiebreaker format! Now that we have an odd number of tiers (counting bo3), we have new options! The two simplest options:
  • AAA, as the double meta, is the default meta in tiebreak. Both teams pick a meta.
  • Both teams strike two metas. The remaining three metas are our tiebreaker.
Both of the above options still have edge cases to figure out (mostly what if teams pick / strike the same thing), and obviously we're open to other options as well.

Other things might be worth discussing:
Budget and manager selfbuy (100k and 13.5k)
Signups (including binding "tiers not played")
Whatever else y'all want to talk about, idk
 
1740157638472.png

The tiebreaker slots should be
SV Balanced Hackmons
SV Godly Gift
SV Mix and Mega

Pros:
- no delays while teams submit picks/strikes, easy to schedule and play quickly, you can start prepping immediately
- no randomness like picks/strikes can have in the event of both teams choosing the same tiers.
- teams are guaranteed to have capable subs, good builder support since these tiers have the extra slot in bo3
- the tiers are well-liked as evidenced by winning the poll in the previous thread, arguably among the most competitive
- more BH (unbiased)

Cons:
- no PiC
 
UT said:
With the tiers now set and considering how much fun last year's was, we're doing it again!

For a refresher, last year's admin thread can be found here with the changes specific to OMPL XII called out here. With one notable exception, by default we will be running with the rules and procedures we used last year, but we're happy to entertain changes and improvements if people have suggestions!

Things that are not up for discussion: number of teams (eight), the tiers (AAA1 / AAA2 / BH / GG / MnM / PiC / STAB / bo3), no trades, no retains.

The main thing that will need to change, as is tradition, is the tiebreaker format! Now that we have an odd number of tiers (counting bo3), we have new options! The two simplest options:
  • AAA, as the double meta, is the default meta in tiebreak. Both teams pick a meta.
  • Both teams strike two metas. The remaining three metas are our tiebreaker.
Both of the above options still have edge cases to figure out (mostly what if teams pick / strike the same thing), and obviously we're open to other options as well.

Other things might be worth discussing:
Budget and manager selfbuy (100k and 13.5k)
Signups (including binding "tiers not played")
Whatever else y'all want to talk about, idk
Click to expand...
I am surprised you didn't link any Taylor Swift songs
 
TB edge cases

Picks option 1 - Backup pick
Team A and B, both submit two picks, call these A1 A2 and B1 B2 (can pick AAA)
Case 1 - If A1 =\= B1 (didn't pick same tier), TB is AAA/A1/B1
Otherwise A1 == B1 (picked same tier). Check second picks. Note that second picks can't be the same as first pick so no issues here.
Case 2 - If A2 =\= B2 (didn't pick same for second pick), TB is A1=B1, A2, B2
Case 3 - Otherwise same picks again, TB is AAA, A1=B1, A2=B2

Eg 1 - A: AAA, BH. B: MnM, GG -> TB is AAA, AAA, MnM (Case 1)
Eg 2 - A: AAA, BH. B: MnM, BH -> TB is AAA, AAA, MnM (Case 1, second pick match irrelevant)
Eg 3 - A: AAA, BH. B: AAA, MnM -> TB is AAA, BH, MnM (Case 2)
Eg 4 - A: AAA, BH. B: AAA, BH -> TB is AAA, AAA, BH (Case 3)

+ Fairly straightforward, edge case is rare. Picks ensures best players will be featured.
- Picks are more likely for skewed matchups, subjectively reducing suspense.

Picks option 2 - Double tier

Both teams submit 1 pick. TB is these picks + AAA. Can allow for 2 or even 3 of the same tier (if both pick AAA)

+ Straightforward. 1 AAA locked for AAA enjoyers. Same picks positives as before.
- Allows for double or even triple tiers which can be impact picks. Same picks negatives as before.
Elaboration - A team might be incentivized to not pick their strong tier because of the possibilities of doubling tiers.
Example - A team thinks they have stronger PiC player, but no second PiC player. Opposing team has 2 PiC players. First team picking PiC would risk double PiC which would be highly favourable for second team.

Strikes option 1 - Random
Both teams strike 2. If no duplicate strikes then left with 3, all is good. If duplicate then left with more than 3, randomly select from remaining.
Random selection is fine here because of strikes, remaining tiers should not be too biased in one team's favour.

Eg 1 - A: AAA, BH. B: MnM, GG -> TB is PiC, STAB, Bo3
Eg 2 - A: AAA, BH. B: MnM, AAA -> TB is pick 3 from (GG, PiC, STAB, Bo3)
Eg 3 - A: AAA, BH. B: BH, AAA -> TB is pick 3 from (GG, MnM, PiC, STAB, Bo3)

+ Extremely straightforward, no complicated process. Strikes balance out matchups, adding to suspense.
- Teams don't get to play their strongest tiers.

Strikes option 2 - More strikes
Both teams strike 2 + 1 backup.
Case 1 - If 4 first strikes are distinct then left with 3 tiers for TB.
Otherwise there are dupes, add backup strikes.
Case 2 - Combined strikes have 1 dupe and leave 2 tiers. Use dupe + the 2 tiers for TB.
Case 3 - Combined strikes have 2 dupes and leave 3 tiers for TB.
Case 4 - Combined strikes are 3 pairs of dupes, use these 3 for TB.

TLDR process lists in order until either left with 3 unstriked tiers for TB, or all 3 strikes are dupes and use them for TB.

Eg 1 - A: AAA, BH, w/e. B: MnM, GG, w/e -> TB is PiC, STAB, Bo3 (Case 1, backup irrelevant)
Eg 2 - A: AAA, BH, GG. B: MnM, AAA, STAB -> TB is AAA, PiC, Bo3 (Case 2, dupe is AAA)
Eg 3 - A: AAA, BH, GG. B: MnM, AAA, GG -> TB is STAB, PiC, Bo3 (Case 3)
Eg 4 - A: AAA, BH, GG. B: MnM, AAA, BH -> TB is STAB, PiC, Bo3 (Case 3)
Eg 5 - A: AAA, BH, MnM. B: MnM, AAA, BH -> TB is AAA, BH, MnM (Case 4)
Eg 6 - A: AAA, BH, MnM. B: BH, AAA, GG -> TB is STAB, PiC, Bo3 (Case 3)
Eg 7 - A: AAA, BH, MnM. B: BH, AAA, MnM -> TB is AAA, BH, MnM (Case 4)

+ (Not actually as complicated as it seems check TLDR), objective process no randomness. Same strike positives as before.
- Intimidating process to look at, need a backup strike from both teams, same strike negatives as before.

Manager selfbuy prices
I still am a fan of QT's proposed selfbuy price idea (TLDR managers get to allocate a set number of credits across all other managers, and the combined credit is the manager's selfbuy price). However I know that people aren't a big fan of this and it also messes with moccs which is the most important thing so yeah.

Tiers not played
I know a lot of other places don't do it but I'm not sure why the bind breaks during playoffs, feels extremely arbitrary and potent to giving unfair advantages to teams that make playoffs while having players with bound not played tiers that they are very strong at. Would like to see the bind be for the entire season.

Signup dates
Idk if this is up for discussion but please keep managers signups a week before players so we can have the full 2 weeks of player signups to mocc.

AAA slots
Completely random just throwing out as a possibility: AAA1 player has to be no cheaper than AAA2 player.
 
Chessking345 said:
Manager selfbuy prices
I still am a fan of QT's proposed selfbuy price idea (TLDR managers get to allocate a set number of credits across all other managers, and the combined credit is the manager's selfbuy price). However I know that people aren't a big fan of this and it also messes with moccs which is the most important thing so yeah.
Click to expand...
My biggest issue with this is manager self-buys are necessary to encourage quality manager signups; if managers do not know their own self-buy price until they're already locked into the tour, this will discourage them from signing up if they also want to play, as they might get bid up to a price they cannot justify.

I am open to adjusting the self-buy price if people think the current one is an issue, but anything that leaves potential managers with incomplete information I think causes more harm than good by reducing manager sign ups.
 
akira is on the right track but not having the biggest tier (the only one with two specialised slots) is criminal. chessking you're probably right but im not gonna read it there's too much maths, sorry

one of the tiebreaks should always be aaa, then both teams pick a tier for the remaining games.
if both managers happen to pick the same one (unlikely but possible) then the third tier should be decided at random between the remaining eligible bo3 tiers:
- team 1 picks bh, team 2 picks partners in crime
the tiebreak is AAA, BH, PIC
- team 1 picks stab, team 2 picks stab
the tiebreak is AAA, STAB, and a randomly selected tier between BH, GG, and MnM
- team 1 picks gg, team 2 picks gg
the tiebreak is AAA, GG, and a randomly selected tier between BH and MnM

no double slots allowed even for aaa because that's boring.

this system should guarantee high quality games in every possible scenario as you will always have either a specialist or a BO3 player drafted to be put into a tiebreak

UT said:
Other things might be worth discussing:
Budget and manager selfbuy (100k and 13.5k)
Signups (including binding "tiers not played")
Whatever else y'all want to talk about, idk
Click to expand...
"Budget and manager selfbuy (100k and 13.5k)"
keep manager selfbuys static until we have even a single instance of a 13.5k manager having a crazy run (bonus point if the rest of the team doesn't noticeably suffer from having less dedicated support). introducing variable pricing based on performance throws a lot of other annoying questions into the pot like "do playoff games count the same" and "what if my team is already out and my manager starts throwing games to lower their price next year" and the idea of other managers voting on your price relies on the other managers having good analytics of what that manager-player is worth. dont make me type !overpay more than i have to.

"Signups (including binding "tiers not played")"
keep "tiers not played", there's no drawback to it. i dont really care if the lock stays on for playoffs and/or tiebreaks though.

"Whatever else y'all want to talk about, idk"
GHROmVQXUAARsy9.jpg


im middle left shaq today what about you guys
 
Greybaum said:
akira is on the right track but not having the biggest tier (the only one with two specialised slots) is criminal. chessking you're probably right but im not gonna read it there's too much maths, sorry

one of the tiebreaks should always be aaa, then both teams pick a tier for the remaining games.
if both managers happen to pick the same one (unlikely but possible) then the third tier should be decided at random between the remaining eligible bo3 tiers:
- team 1 picks bh, team 2 picks partners in crime
the tiebreak is AAA, BH, PIC
- team 1 picks stab, team 2 picks stab
the tiebreak is AAA, STAB, and a randomly selected tier between BH, GG, and MnM
- team 1 picks gg, team 2 picks gg
the tiebreak is AAA, GG, and a randomly selected tier between BH and MnM

no double slots allowed even for aaa because that's boring.

this system should guarantee high quality games in every possible scenario as you will always have either a specialist or a BO3 player drafted to be put into a tiebreak


"Budget and manager selfbuy (100k and 13.5k)"
keep manager selfbuys static until we have even a single instance of a 13.5k manager having a crazy run (bonus point if the rest of the team doesn't noticeably suffer from having less dedicated support). introducing variable pricing based on performance throws a lot of other annoying questions into the pot like "do playoff games count the same" and "what if my team is already out and my manager starts throwing games to lower their price next year" and the idea of other managers voting on your price relies on the other managers having good analytics of what that manager-player is worth. dont make me type !overpay more than i have to.

"Signups (including binding "tiers not played")"
keep "tiers not played", there's no drawback to it. i dont really care if the lock stays on for playoffs and/or tiebreaks though.

"Whatever else y'all want to talk about, idk"
View attachment 715623

im middle left shaq today what about you guys
Click to expand...
To avoid the scenario where you need to randomize a tier, you could have it set up so AAA is set, and then have it so one team is randomly designated to pick first. Then the team designated to go first chooses one tier, and then the team designated as second chooses a tier that isn't AAA or has been chosen by the team that selected first.
 
Hello!

Picks + Strikes led the poll above, and so we have come up with what we believe to be the optimal solution for tiebreakers.

In the case of a lower seed tiebreaking with a higher seed, the Pick/Strike order will follow this order:

Metas: AAA, BH, GG, PiC, MnM, STAB, Bo3

Lower Seed Picks 1
Higher Seed Picks 1 and Strikes 1
Lower Seed Strikes 2
Higher Seed Strikes 1

This will result in 4 strikes and 2 picks, with the final meta being the unpicked + unstriked meta.

In the case of an absolute tiebreaker (tied in all metrics), one of the captains of each team involved must play each other in a Bo1 Randbats Mayhem. The winner's team will be the higher seed in the tiebreaker selection process above.

Let us know what you think about this solution!
 
goldmason said:
In the case of an absolute tiebreaker (tied in all metrics), one of the captains of each team involved must play each other in a Bo1 Randbats Mayhem. The winner's team will be the higher seed in the tiebreaker selection process above.
Click to expand...
can the managers gentleman's to something else
 
bo5 means we have to leave a meta out, bo7 means we double up or have to include bo3 in the bo7, and both of those suck
 
Last call (ish) for any other format changes you want to see, otherwise I think we're close to locking this in.

Changes from last year are the metas, the tiebreaker, and making tiers not played binding for playoffs as well (not tiebreaker).
 
mayonnaise. mayo. said:
Allow me to elaborate.

USUM is the gen of PH with a permaladder. If none of the PH gens had permaladders, I would go with ORAS or DPPt. Furthermore, having a gen of PH would attract a lot of the community and lead to a MASSIVE tour. Just saying.
Click to expand...
yo we only have competitive metas in ompl :baseballed:
 
Status
Not open for further replies.
Back
Top