Saturday, November 23, 2019

EURO 2020 qualifying play-offs simulations (November 2019)

After the confusion about the pairing of play-off paths, necessary to grant play-off path winners who are also host in EURO 2020 their own pre-designated group, the draw yesterday was pretty straight forward.


The draw-result:

Path A
Iceland - Romania
Bulgaria - Hungary
final played in Bulgaria or Hungary.

Path B
Bosnia-Herzegovina - Northern Ireland
Slovakia - Republic of Ireland
final played in Bosnia-Herzegovina or Northern Ireland.

Path C
Scotland - Israel
Norway - Serbia
final played in Norway or Serbia.

Path D
Georgia - Belarus
Northern Macedonia - Kosovo
final played in Georgia or Belarus.

The 8 single match path semi finals are played on March 26th 2020 and the 4 single match path finals are played on March 31st 2020.

For the draw of the groups in the EURO 2020 finals next week, the following teams (or groups of teams) are already certain of their group:

Group A: Italy*, pot 2, pot 3, Wales/Finland
Group B: Belgium, Russia, Denmark*, Wales/Finland
Group C: Ukraine, Netherlands*, pot 3, Romania or the winner of path D
Group D: England*, pot 2, pot 3, the winner of path C
Group E: Spain*, pot 2, pot 3, the winner of path B
Group F: Germany*, pot 2, pot 3, Iceland/Bulgaria/Hungary or the winner of path D

Teams to be drawn from pot 2 are France, Poland, Switzerland and Croatia.
Teams to be drawn from pot 3 are Portugal, Turkey, Austria, Sweden and Czech Republic.
* The host that plays three home matches in the group

Here are the probabilities (in %) -generated over 10.000 simulations- for each participant in the play-offs to win their semi final and to win their path-final.
All match-results in each simulation are based on ClubElo prediction formula's for goals scored in a match depending on elo home win expectancy.



path A     semi final              final
Iceland 71,10 48,45
Bulgaria 70,06 20,86
Romania 28,90 20,35
Hungary 29,94 10,34
path B     semi final              final
Slovakia 74,22 48,22
Bosnia-Herzegovina 73,47 25,85
Republic of Ireland 25,78 16,58
Northern Ireland 26,53 9,35
path C     semi final              final
Norway 68,30 50,02
Serbia 31,70 24,74
Scotland 80,77 21,09
Israel 19,23 4,15
path D     semi final              final
Georgia 77,57 38,58
North Macedonia 72,13 35,04
Kosovo 27,87 18,06
Belarus 22,43 8,32

About me:

Software engineer, happily unmarried and non-religious. You won't find me on Twitter or other so called social media. Dutchman, joined the blog in March 2018.

20 comments:

  1. As an N.I. fan, is there any chance of your laying odds on those simulations?
    With a predicted 9.35% chance of qualifying, I'd quite like a bet - shall we call it a nice round 10/1?
    Anyone?

    ReplyDelete
    Replies
    1. I don't bet.
      And I understand that, as a N.I. fan, such a rather low predicted success rate for your team can be a bit of a cold shower. But hey, it is only mathematics and... it is not 0% :)

      Having seen your team play twice against the Dutch recently, I do have the feeling that 10% chance to qualify sounds about right though: I wasn't impressed at all with your skills and tactics. But again, I don't bet.

      Lots of success in March !

      Delete
    2. For me 10% seems too low, especially considering that N.I. is estimated to win only 35% of their final home game if they win the semis, taking into account how poorly the opponnents in path B have played recently.

      But granted, it is nitpicking...

      Delete
  2. Iceland, Bulgaria and Hungary will be unimpressed after the draw for Euro2020 Group F. If they qualify they play Portugal, Germany then France, wow.

    Romania would be much happier with the prospect of 2 home games against Austria and Ukraine first up if they qualify.

    The winners of Georgia, Belarus, Kosovo and North Macedonia will be very hopeful that Romania gets knocked out in the playoffs and they are locked into Group C. Iceland will therefore have a lot of neutral fans when they host Romania on March 26.

    In the Tournament proper the Group D Winners face the Group F Runners Up in Dublin in the Round of 16. I would much prefer to finish Group D runner up and face the Group E Runners Up in the Round of 16 in Copenhagen, as long as that doesn't end up being Spain.

    Similarly the Group B and Group C Winners both have around a 1/3rd chance to face the Group F 3rd placed team in the Round of 16 in Balbao (Group B) or Budapest (Group C) respectively.

    Switzerland, Belgium, Austria, then Playoff Winning team in Group D, Poland and Portugal all end up as the teams having to travel twice in their Euro2020 groups.

    ReplyDelete
  3. What a horribly imbalanced Euro draw that was. The holders and winners of the last 2 World Cups all in one group, but Group C likely to feature 4 teams who never even made it to Russia if Iceland don't win their play-off path.
    Surely it would have made more sense to have the holders (Portugal) among the Pot 1 teams, as is tradition in such events.

    ReplyDelete
    Replies
    1. I think there were so many scenarios involved in this format and that one suggested tweak would have made a key difference to the draw dynamics as it would have dropped Ukraine to Pot 2 and Russia to Pot 3.

      Delete
    2. Yes, and with all due respect to Ukraine who had an excellent qualifying campaign, that would have given the pots a more balanced look.
      One other tweak could have been not having teams that met in the qualifiers meeting again in the group phase. As it is now, Bel/Rus, Eng/Cze & Spa/Swe are all drawn to meet yet again.

      Delete
    3. I don't recall when the holders of either the World Cup or the Euros were automatically placed in Pot 1. Must have been quite some time ago.

      In recent times pots were always based some type of ranking (like world ranking) or coefficients that weighted recent results much higher than results from four years ago.

      Delete
    4. Sorry, my bad. Missed out that in previos Euros holders were indeed placed in pot 1 (in contrast to recent World Cups, I beleive).

      Delete
    5. You are wrong. Previous WC Champ was Germany and they were in pot 1 as I remember every WC winner and Euro Champion since 1982.

      Delete
    6. Yes and no. They might have been in Pot 1 but not based on them being the holders but on FIFA Ranking (skewed towards recent results).

      Delete
    7. Ironically, for once, the (November) FIFA rankings would have been a very useful basis for the Euro draw.
      The pots would have looked like this:
      Pot 1: Belgium, France, England, Croatia, Portugal, Spain
      Pot 2: Switzerland, Italy, Netherlands, Germany, Denmark, Sweden
      Pot 3: Poland, Wales, Ukraine, Austria, Turkey, Russia
      Pot 4: Czech Rep, Finland, Play-Off Winners (4)

      That would have produced 6 pretty well-balanced groups, and would also guaranteed to keep Ukraine & Russia apart in the group stage.



      Delete
    8. Nothing ironic there Anonymous, however flawed the FIFA rankings would be fine to seed the competition. Any improvements to the ranking system would be welcomed but would only improve things further.

      The system UEFA did go for put a Pot 1 strength team in Pot 3 because they dropped more points than other teams in qualifying, however it could definitely be argued that they played stronger opposition than the best 2 ranked group runners up who scraped into Pot 2.

      Delete
  4. UEFA had decided the Euro 2022 qualification system

    https://www.uefa.com/insideuefa/about-uefa/news/newsid=2635809.html

    The above is a little unclear but 10 runners up and 2 more teams will make 12 teams playing off for the final 3 European WC spots. Hopefully the 2 spots linked to the 2020-21 Nations league are only from 3rd ranked teams so that there is still an incentive for teams who end up well back from the Group Winners and Runners up.

    ReplyDelete
  5. Ted, what's unclear about:
    "and the two best group winners of the 2020/21 UEFA Nations League overall ranking that have neither qualified directly for the final tournament as European Qualifiers group winners, nor entered the play-offs already as European Qualifiers group runners-up" ?
    Nothing about 3rd ranked teams...

    ReplyDelete
  6. It is clear. 2 spots from UNL will be UNL group winners teams which won't be 1st and 2nd place in classic qualification. It means they could be teams even from league C.

    ReplyDelete
  7. @Anonymous. Clear how?

    What if all 14 UNL group winners qualify or finish 2nd in WCQ. Do the 2 spots then go to teams that were 2nd in their UNL Group.

    Why not just take the 3rd placed teams in WCQ only and give playoff spots to the 2 who were ranked higher in the overall UNL standings with the possible exception that teams being relegated by finishing last in the UNL groups would not be eligible.

    ReplyDelete
    Replies
    1. Sorry but this is not my idea, I've read this info on the Uefa website. But...Do you belive eg. Latvia and Faroe Islands win or take 2nd place in classic qualification? Probably uefa executive comitee think this way.

      Delete
    2. Anonymous - I do not believe the Two Group D Winners whoevver they end up being will come close to finishing top 2 in their Group. However if I was determining rules then I would be clear on all possible outcomes, however unlikely they may be.

      I suspect that it is possible that 2 of the 4 UNL League B Group Winners could be the beneficiaries but there is definitely a likelihood that the top ranked team/s from League C could end up benefitting if they do not finish in the top 2 in WCQ.

      Delete