Prediction game - Rules and discussion thread

[Jackson Harrison]

@Lyonnais @fabiomh

What about we use a real Swiss system? We don't need to mess with TV pairings etc.

Alternatively this may be of use https://en.m.wikipedia.org/wiki/2019–20 ... qualifying

[fabiomh]

@Lyonnais @fabiomh

What about we use a real Swiss system? We don't need to mess with TV pairings etc.

Alternatively this may be of use https://en.m.wikipedia.org/wiki/2019–20 ... qualifying

This is exactly the goal. The only constraint is matching two opponents from each pot.

Some comment of mine:
- Of course we need to have an even number if total players - the virtual average can help.
- The schedule is simple if each pot has the same number of players and with an even number of players per pot.
if not, we may need some correction.

some example and some hypothesis of solution:
48 partecipants:
- CL/EL: 4 pots x 12 players per pot: no issue
- ECL: 3 pots x 16 players per pot: no issue

47 partecipants:
- CL/EL: (+1 virtual avg) like 48 partecipants
- ECL: (+1 virtual avg) like 48 partecipants

46 partecipants:
- CL/EL:
sol 1: (+2 virtual avg) like 48 partecipants
sol 2: no virt avg: 4 pots (A 11 players, B 12, C 12, D 11) : schema similar to the one in the link suggested by @Jackson Harrison
- ECL:
sol 1: (+2 virtual avg): like 48 partecipants
sol 2: no virt avg: 3 pots (A 16 players, B 14, C 16) : schema similar to the one in the link ? to be found for 3 pots

45 partecipants:
- CL/EL:
sol 1: (+3 virtual avg) like 48 partecipants
sol 2: (+1 virtual avg): like Sol 2 for 46 partecipants
- ECL:
sol 1: (+3 virtual avg): like 48 partecipants
sol 2: (+1 virtual avg): like Sol 2 for 46 partecipants

44 partecipants:
- CL/EL:
sol 1: (+4 virtual avg) like 48 partecipants
sol 2: (no virtual avg): a "simple schema corrected", no possibility to have MDs with only intra-pot matches
- ECL:
sol 1: (+4 virtual avg): like 48 partecipants
sol 2: (+2 virtual avg): like Sol 2 for 46 partecipants
sol 3: (no virtual avg): 3 potes (A 14 players, B 16, C 14): schema similar to the one in the link ? to be found for 3 pots

...

[Jackson Harrison]

@Lyonnais @fabiomh

What about we use a real Swiss system? We don't need to mess with TV pairings etc.

Alternatively this may be of use https://en.m.wikipedia.org/wiki/2019–20 ... qualifying

This is exactly the goal. The only constraint is matching two opponents from each pot.

Some comment of mine:
- Of course we need to have an even number if total players - the virtual average can help.
- The schedule is simple if each pot has the same number of players and with an even number of players per pot.
if not, we may need some correction.

some example and some hypothesis of solution:
48 partecipants:
- CL/EL: 4 pots x 12 players per pot: no issue
- ECL: 3 pots x 16 players per pot: no issue

47 partecipants:
- CL/EL: (+1 virtual avg) like 48 partecipants
- ECL: (+1 virtual avg) like 48 partecipants

46 partecipants:
- CL/EL:
sol 1: (+2 virtual avg) like 48 partecipants
sol 2: no virt avg: 4 pots (A 11 players, B 12, C 12, D 11) : schema similar to the one in the link suggested by @Jackson Harrison
- ECL:
sol 1: (+2 virtual avg): like 48 partecipants
sol 2: no virt avg: 3 pots (A 16 players, B 14, C 16) : schema similar to the one in the link ? to be found for 3 pots

45 partecipants:
- CL/EL:
sol 1: (+3 virtual avg) like 48 partecipants
sol 2: (+1 virtual avg): like Sol 2 for 46 partecipants
- ECL:
sol 1: (+3 virtual avg): like 48 partecipants
sol 2: (+1 virtual avg): like Sol 2 for 46 partecipants

44 partecipants:
- CL/EL:
sol 1: (+4 virtual avg) like 48 partecipants
sol 2: (no virtual avg): a "simple schema corrected", no possibility to have MDs with only intra-pot matches
- ECL:
sol 1: (+4 virtual avg): like 48 partecipants
sol 2: (+2 virtual avg): like Sol 2 for 46 partecipants
sol 3: (no virtual avg): 3 potes (A 14 players, B 16, C 14): schema similar to the one in the link ? to be found for 3 pots

...

Or maybe include more virtavg to make up numbers

For example 45 player:
Pot 1: Top 12
Pot 2: Next 11+virgavg of all pot 2
Pot 3: Next 11+virgavg of all pot 3
Pot 4: Next 11+virgavg of all pot 4

What are your thoughts on a real Swiss System?

[fabiomh]

the "simple schema corrected" for 44 partecipants divided in 4 pots (A, B, C, D)

based on simple schema posted by @Lyonnais (perfect for 48 partecipants):
I renamed 1,2,3,4 into A,B,C,D. The teams into A1, A2, ... A11, B1, B2, ... and so on

...
MD1/MD5: pot 1 v pot 4, pot 2 v pot 3
MD2/MD6: pot 1 v pot 3, pot 2 v pot 4
MD3/MD7: pot 1 v pot 2, pot 3 v pot 4
MD4/MD8: pot 1 v pot 1, pot 2 v pot 2, pot 3 v pot 3, pot 4 v pot 4

MD1/MD5: A v D, B v C
MD2/MD6: A v C, B v D
MD3/MD7: A v B, C v D
MD4/MD8: A v A, B v B, C v C, D v D

Correction:
MD1: A1-D1, A2-D2, ... A10-D10, same for B-C (A11-D11, A6-D6, B11-C11, B6-D6 replaced by A11-A6, B11-B6, C11-C6, D11-D6)
MD2: A1-C8, A2-C9, ..., A4-C11, A5-C1, ..., A11-C7, same for B-D
MD3: A1-B10, A2-B11, A3-B1, ... A11-B9, same for C-D
MD4: A1-A2, A3-A4, ..., A9-A10, [no A11] same for B-B, C-C, D-D (+ A11-D11 + B11-C11)
MD5: A1-D4, A2-D5, ..., A8-D11, A9-D1, A10-D2, A11-D3, same for B-C
MD6: A1-C5, A2-C6, ..., A7-C11, A8-C1, ..., A11-C4, same for B-D
MD7: A1-B9, ... A3-B11, A4-B1, ..., A11-B8, same for C-D
MD8: A1-A7, A2-A9, A3-A11, A4-A8, A5-A10, [no A6], same for B-B, C-C, D-D (+A6-D6 + B6-C6)

[fabiomh]

@Lyonnais @fabiomh

What about we use a real Swiss system? We don't need to mess with TV pairings etc.

Alternatively this may be of use https://en.m.wikipedia.org/wiki/2019–20 ... qualifying

This is exactly the goal. The only constraint is matching two opponents from each pot.

Some comment of mine:
- Of course we need to have an even number if total players - the virtual average can help.
- The schedule is simple if each pot has the same number of players and with an even number of players per pot.
if not, we may need some correction.

some example and some hypothesis of solution:
48 partecipants:
- CL/EL: 4 pots x 12 players per pot: no issue
- ECL: 3 pots x 16 players per pot: no issue

47 partecipants:
- CL/EL: (+1 virtual avg) like 48 partecipants
- ECL: (+1 virtual avg) like 48 partecipants

46 partecipants:
- CL/EL:
sol 1: (+2 virtual avg) like 48 partecipants
sol 2: no virt avg: 4 pots (A 11 players, B 12, C 12, D 11) : schema similar to the one in the link suggested by @Jackson Harrison
- ECL:
sol 1: (+2 virtual avg): like 48 partecipants
sol 2: no virt avg: 3 pots (A 16 players, B 14, C 16) : schema similar to the one in the link ? to be found for 3 pots

45 partecipants:
- CL/EL:
sol 1: (+3 virtual avg) like 48 partecipants
sol 2: (+1 virtual avg): like Sol 2 for 46 partecipants
- ECL:
sol 1: (+3 virtual avg): like 48 partecipants
sol 2: (+1 virtual avg): like Sol 2 for 46 partecipants

44 partecipants:
- CL/EL:
sol 1: (+4 virtual avg) like 48 partecipants
sol 2: (no virtual avg): a "simple schema corrected", no possibility to have MDs with only intra-pot matches
- ECL:
sol 1: (+4 virtual avg): like 48 partecipants
sol 2: (+2 virtual avg): like Sol 2 for 46 partecipants
sol 3: (no virtual avg): 3 potes (A 14 players, B 16, C 14): schema similar to the one in the link ? to be found for 3 pots

...

Or maybe include more virtavg to make up numbers

For example 45 player:
Pot 1: Top 12
Pot 2: Next 11+virgavg of all pot 2
Pot 3: Next 11+virgavg of all pot 3
Pot 4: Next 11+virgavg of all pot 4

What are your thoughts on a real Swiss System?

48 is the perfect number; from 45 to 44 virtual avg can easily used to increase the number to 48.
Below 45, you cannot use too many virtual average players, considering also that probably some players will retire during the competition and you are forced to add some more virtual avg player.


I don't know very well the real Swiss System, I have just read a bit on wiki. It should have not pre-fixed schedule, but it should be determined every MD, based on the ranking, and avoiding that two palyers match twice. Too complex,and it needs some specific computer program to be run after each MD.
IMHO not suitable for our simple game and goals. We want to enjoy and need a clear, simple and transparent method.

[Jackson Harrison]

This is exactly the goal. The only constraint is matching two opponents from each pot.

Some comment of mine:
- Of course we need to have an even number if total players - the virtual average can help.
- The schedule is simple if each pot has the same number of players and with an even number of players per pot.
if not, we may need some correction.

some example and some hypothesis of solution:
48 partecipants:
- CL/EL: 4 pots x 12 players per pot: no issue
- ECL: 3 pots x 16 players per pot: no issue

47 partecipants:
- CL/EL: (+1 virtual avg) like 48 partecipants
- ECL: (+1 virtual avg) like 48 partecipants

46 partecipants:
- CL/EL:
sol 1: (+2 virtual avg) like 48 partecipants
sol 2: no virt avg: 4 pots (A 11 players, B 12, C 12, D 11) : schema similar to the one in the link suggested by @Jackson Harrison
- ECL:
sol 1: (+2 virtual avg): like 48 partecipants
sol 2: no virt avg: 3 pots (A 16 players, B 14, C 16) : schema similar to the one in the link ? to be found for 3 pots

45 partecipants:
- CL/EL:
sol 1: (+3 virtual avg) like 48 partecipants
sol 2: (+1 virtual avg): like Sol 2 for 46 partecipants
- ECL:
sol 1: (+3 virtual avg): like 48 partecipants
sol 2: (+1 virtual avg): like Sol 2 for 46 partecipants

44 partecipants:
- CL/EL:
sol 1: (+4 virtual avg) like 48 partecipants
sol 2: (no virtual avg): a "simple schema corrected", no possibility to have MDs with only intra-pot matches
- ECL:
sol 1: (+4 virtual avg): like 48 partecipants
sol 2: (+2 virtual avg): like Sol 2 for 46 partecipants
sol 3: (no virtual avg): 3 potes (A 14 players, B 16, C 14): schema similar to the one in the link ? to be found for 3 pots

...

Or maybe include more virtavg to make up numbers

For example 45 player:
Pot 1: Top 12
Pot 2: Next 11+virgavg of all pot 2
Pot 3: Next 11+virgavg of all pot 3
Pot 4: Next 11+virgavg of all pot 4

What are your thoughts on a real Swiss System?

48 is the perfect number; from 45 to 44 virtual avg can easily used to increase the number to 48.
Below 45, you cannot use too many virtual average players, considering also that probably some players will retire during the competition and you are forced to add some more virtual avg player.


I don't know very well the real Swiss System, I have just read a bit on wiki. It should have not pre-fixed schedule, but it should be determined every MD, based on the ranking, and avoiding that two palyers match twice. Too complex,and it needs some specific computer program to be run after each MD.
IMHO not suitable for our simple game and goals. We want to enjoy and need a clear, simple and transparent method.

Websites like challenge.com operate these systems smoothly.

It's less complex than this, how do we plan to determine fixtures on this? I'm going to run a mock based on this season to see how it would've worked.

[fabiomh]

I would prefer a method with:
a) pots, as per continuity with the past
b) prefixed schedule for all the 8MDs, known as soon as possible, before the first MD

[Jackson Harrison]

I would prefer a method with:
a) pots, as per continuity with the past
b) prefixed schedule for all the 8MDs, known as soon as possible, before the first MD

Agree. Keep it in line with real comps.