UEFA Seeding & Qualification based on ClubElo Values - Updates & Feedback

[amirbachar]

Question regarding how coefficients are inherited across multiple qualification rounds:

- Partizan Belgrade (coeff 25.5) starts in NCQ2 where they lose to Servette and inherits their coefficient of 9.0 into ELQ3. They are now unseeded in ELQ3, based on Servette's coefficient.

- Nordsjelland (coeff 7.5) starts in ELQ2 where they win against Cluj and inherits their coefficient of 26.5 into ELQ3. They are now seeded in ELQ3, based on Cluj's coefficient.

- Nordsjaelland draw Partizan in ELQ3 and wins. What coefficient to they have in EL Play Off? Partizan's of 25.5 (even though Partizan came with a lower coefficient from a previous round)? Their own of 7.5? Surely not Servette's or Cluj's?

They use 25.5, the higher one of the actual teams.

[amirbachar]

:

[Update]

https://docs.google.com/spreadsheets/d/ ... XcSfrTFvk/

The spreadsheet has now been updated as below:

- All access list and rebalancing changes based on UEFA's latest regulations (and based on @eye's excellent spreadsheet.

- Losers relegated to the tournament below take the lower coefficient of the pair: now also for CHQ1->ECLQ2cp, CHQ2->ELQ3cp, CHQ3->ELPO, ELQ3cp->ECLPOcp.

- Seeding based on coefficient (either actual or inherited) is now used across all rounds and tournaments

From a group stage participation perspective it yield the following forecast:

Thank you
BTW, There is a bug with the average and median Elo.

[Stadion]

They use 25.5, the higher one of the actual teams.

That's easy -- then no changes required to the spreadsheet.

Thank you
BTW, There is a bug with the average and median Elo.

Thanks for flagging - fixed.

[amirbachar]

They use 25.5, the higher one of the actual teams.

That's easy -- then no changes required to the spreadsheet.

Thank you
BTW, There is a bug with the average and median Elo.

Thanks for flagging - fixed.

Thank you for all your efforts!

[Stadion]

With so much coefficient ‘inheritance’ between qualification rounds and tournaments, I’m planning on implementing a label that shows from what team and round a club’s coefficient has been inherited.

[Stadion]

With so much coefficient ‘inheritance’ between qualification rounds and tournaments, I’m planning on implementing a label that shows from what team and round a club’s coefficient has been inherited.

[Update]

I have now implemented this for UCL and UEL. I've added a column for coefficient, and for those clubs that have inherited their coefficient a number indicates from which club in the previous round, and the colour indicates from which tournament (e.g. blue for UCL, orange for UEL).

[amirbachar]

With so much coefficient ‘inheritance’ between qualification rounds and tournaments, I’m planning on implementing a label that shows from what team and round a club’s coefficient has been inherited.

[Update]

I have now implemented this for UCL and UEL. I've added a column for coefficient, and for those clubs that have inherited their coefficient a number indicates from which club in the previous round, and the colour indicates from which tournament (e.g. blue for UCL, orange for UEL).

That's great, thanks!
I noticed that you let the lowest Elo unseeded winner inherit the coefficients of the highest Elo seeded loser. I believe it would make more sense to let the highest one inherit the highest one's coefficients instead.
Would that be hard to implement?

[Stadion]

With so much coefficient ‘inheritance’ between qualification rounds and tournaments, I’m planning on implementing a label that shows from what team and round a club’s coefficient has been inherited.

[Update]

I have now implemented this for UCL and UEL. I've added a column for coefficient, and for those clubs that have inherited their coefficient a number indicates from which club in the previous round, and the colour indicates from which tournament (e.g. blue for UCL, orange for UEL).

That's great, thanks!
I noticed that you let the lowest Elo unseeded winner inherit the coefficients of the highest Elo seeded loser. I believe it would make more sense to let the highest one inherit the highest one's coefficients instead.
Would that be hard to implement?

Actually my draw logic is currently based on coefficients at the moment. I.e. the seeded team with the highest coefficient that is predetermined to progress will be "drawn" against the unseeded team with the lowest coefficient that's predetermined to lose. Coefficients are then inherited accordingly.

If you could put forth a compelling reason for why i should revise this logic I'm happy to consider it.

[amirbachar]

[Update]

I have now implemented this for UCL and UEL. I've added a column for coefficient, and for those clubs that have inherited their coefficient a number indicates from which club in the previous round, and the colour indicates from which tournament (e.g. blue for UCL, orange for UEL).

That's great, thanks!
I noticed that you let the lowest Elo unseeded winner inherit the coefficients of the highest Elo seeded loser. I believe it would make more sense to let the highest one inherit the highest one's coefficients instead.
Would that be hard to implement?

Actually my draw logic is currently based on coefficients at the moment. I.e. the seeded team with the highest coefficient that is predetermined to progress will be "drawn" against the unseeded team with the lowest coefficient that's predetermined to lose. Coefficients are then inherited accordingly.

If you could put forth a compelling reason for why i should revise this logic I'm happy to consider it.

Yes, I do have a good reason, the whole premise of the Sheet is that a team with more lower Elo opponents is more likely to win in that round. But in practice it's probabilistic, and the more such teams there are, the more likely it is, including against more seeded teams in the next round.
What I suggested is equavallent to that.

[Stadion]

Yes, I do have a good reason, the whole premise of the Sheet is that a team with more lower Elo opponents is more likely to win in that round. But in practice it's probabilistic, and the more such teams there are, the more likely it is, including against more seeded teams in the next round.
What I suggested is equavallent to that.

Got it -- so to clarify. There are really three steps that determine which club progress to the next round, and whose coefficient they bring with them:

1. The first step determines which clubs on the seeded and unseeded side progress based on their Elo values. We won't touch this.
2. The next step is what should decide which clubs should be paired for the "draw" (this is really what determines who inherits whose coefficients) -- ClubElo values, or coefficients?
3. Depending on what we picked in the previous step, how do we determine who draws who? The highest seeded Elo/coefficient draws the highest unseeded Elo/coefficient? Or highest draws lowest?

Let me know your thoughts on points 2 and 3!

[amirbachar]

Yes, I do have a good reason, the whole premise of the Sheet is that a team with more lower Elo opponents is more likely to win in that round. But in practice it's probabilistic, and the more such teams there are, the more likely it is, including against more seeded teams in the next round.
What I suggested is equavallent to that.

Got it -- so to clarify. There are really three steps that determine which club progress to the next round, and whose coefficient they bring with them:

1. The first step determines which clubs on the seeded and unseeded side progress based on their Elo values. We won't touch this.
2. The next step is what should decide which clubs should be paired for the "draw" (this is really what determines who inherits whose coefficients) -- ClubElo values, or coefficients?
3. Depending on what we picked in the previous step, how do we determine who draws who? The highest seeded Elo/coefficient draws the highest unseeded Elo/coefficient? Or highest draws lowest?

Let me know your thoughts on points 2 and 3!

That's a good formulation! I believe that the best way to go about 2+3, is to let the the unseeded the the highest Elo draw the seeded with the highest coefficients, since it is the most likely out of all unseeded teams to beat that seeded team (or other seeded teams as well).

[Stadion]

That's a good formulation! I believe that the best way to go about 2+3, is to let the the unseeded the the highest Elo draw the seeded with the highest coefficients, since it is the most likely out of all unseeded teams to beat that seeded team (or other seeded teams as well).

This is now implemented. The unseeded club with the highest ClubElo value is 'drawn' against the seeded club with the highest coefficient. This ensures that strong teams that are unseeded have better chances progressing through the rounds by inheriting the highest coefficient possible from eliminated seeded clubs.

[amirbachar]

That's a good formulation! I believe that the best way to go about 2+3, is to let the the unseeded the the highest Elo draw the seeded with the highest coefficients, since it is the most likely out of all unseeded teams to beat that seeded team (or other seeded teams as well).

This is now implemented. The unseeded club with the highest ClubElo value is 'drawn' against the seeded club with the highest coefficient. This ensures that strong teams that are unseeded have better chances progressing through the rounds by inheriting the highest coefficient possible from eliminated seeded clubs.

Thank you very much, it's great how attentive you are about it

[Stadion]

Spreadsheet updated with the same rebalancing logic as @eye's excellent spreadsheet.

https://docs.google.com/spreadsheets/d/ ... =217607255