Latest updates

Check the Important info page for latest updates! (15 February 2024)

Monday, July 16, 2018

FIFA Ranking: July 2018 final preview (new method)

FIFA has adopted a new calculation method for their ranking. It will be used for the very next ranking, the first ranking after the World Cup, to be published on July 19th.

Here is the final preview of the July 2018 ranking as I think it should be calculated given the information FIFA has provided sofar. There are some points of discussion though that could alter the result considerably:


- what will be the starting values for each team ? I use the unrounded June 2018 ranking points.
- which I (or K) factors to use ? FIFA published two different sets within one week, with the second set a factor 10 smaller than the first set. I use the first set because otherwise the exchanged points per match will be really minimal, even for World Cup knock-out matches.
- what to do with a win in extra time or a PSO win ? I use the same approach as classic elo: a win in extra time counts as a normal win, a PSO win/loss counts as a draw for both teams.
- when aren't points subtracted for the losing team in a knock-out match ? In line with the assumption made in the previous point: I don't subtract points when a team really loses a knock-out match (so in regular time or in extra time). For a PSO win or loss points are added/subtracted for both teams.

Especially the first point about the used starting values will be of huge influence on the resulting ranking. Let's wait and see what FIFA comes up with this Thursday before jumping to conclusions.


Probable July 2018 rank - Team - Probable July 2018 points - +/- Ranking - +/- Points


1 Germany 1469 0 -89
2 Brazil 1438 0 7
3 Belgium 1380 0 82
4 France 1313 3 115
5 Portugal 1252 -1 -22
6 Switzerland 1202 0 3
7 Argentina 1196 -2 -45
8 Chile 1136 1 1
9 Poland 1115 -1 -68
10 Spain 1104 0 -22
11 Peru 1086 0 -39
12 England 1084 0 33
13 Uruguay 1074 1 56
14 Denmark 1071 -2 20
15 Croatia 1031 5 86
16 Mexico 1007 -1 18
17 Colombia 1000 -1 14
18 Netherlands 981 -1 0
19 Sweden 957 5 77
20 Wales 953 -2 0
21 Italy 951 -2 0
22 Tunisia 893 -1 -17
23 USA 874 2 1
24 Iceland 863 -2 -45
25 Costa Rica 861 -2 -23
26 Austria 844 0 -1
26 Senegal 844 1 6
28 Northern Ireland 803 1 0
29 Slovakia 801 -1 -3
30 Romania 783 0 1
31 Republic of Ireland 777 0 0
32 Serbia 771 2 20
33 Paraguay 770 -1 -3
34 Venezuela 755 -1 0
35 Iran 744 2 36
36 Ukraine 733 -1 0
37 Australia 718 -1 0
38 Congo DR 706 0 0
39 Turkey 705 -1 -1
40 Bosnia-Herzegovina 693 0 0
41 Morocco 680 0 -6
42 Scotland 660 0 0
43 Montenegro 652 0 0
44 Greece 650 0 0
45 Czech Republic 643 1 2
46 Nigeria 638 2 20
47 Ghana 625 0 1
48 Bulgaria 614 1 0
48 Cameroon 614 1 0
50 Hungary 609 1 -3
51 Burkina Faso 604 1 0
52 Norway 598 1 2
53 Jamaica 588 1 0
54 Japan 574 7 53
54 Korea Republic 574 3 30
54 Egypt 574 -9 -75
57 Slovenia 558 -1 0
58 Russia 552 12 95
59 Panama 548 -4 -23
60 Albania 538 -2 0
61 Bolivia 535 -2 -1
62 Ecuador 524 -2 0
63 Honduras 503 -1 0
64 Finland 487 -1 1
65 Mali 484 -1 0
66 Cape Verde Islands 478 -1 0
67 Algeria 474 -1 0
68 Saudi Arabia 468 -1 3
69 Cote d'Ivoire 463 -1 0
69 Guinea 463 -1 0
71 FYR Macedonia 451 0 0
72 El Salvador 442 0 0
73 Syria 440 0 0
74 South Africa 431 0 3
75 China PR 423 0 0
76 Zambia 419 0 -1
77 United Arab Emirates 418 0 0
78 Belarus 416 0 -1
79 Lebanon 411 0 0
79 Canada 411 0 0
81 Curacao 408 0 0
82 Uganda 403 0 0
83 Congo 393 0 0
84 Luxembourg 392 1 2
85 Oman 391 -1 0
86 Gabon 390 -1 0
87 Cyprus 384 0 0
88 Benin 379 0 0
89 Iraq 377 0 0
90 Faroe Islands 376 0 0
91 Trinidad and Tobago 370 0 0
92 Kyrgyzstan 363 0 0
93 Israel 362 0 0
94 Estonia 358 0 -1
95 Uzbekistan 354 0 0
96 India 351 1 1
97 Georgia 349 -1 -3
98 Qatar 348 0 0
99 Palestine 347 0 0
100 Armenia 337 0 -1
101 Libya 336 0 0
102 Vietnam 334 0 0
103 Niger 323 0 0
104 Haiti 322 0 0
105 Azerbaijan 321 0 0
106 Mauritania 304 1 0
107 Korea DPR 303 1 0
107 Central African Republic 303 1 0
107 Madagascar 303 -1 -5
110 Jordan 296 0 0
111 Sierra Leone 292 0 0
112 Bahrain 289 1 0
112 Kenya 289 0 -2
114 Mozambique 282 0 0
115 Philippines 280 0 0
116 Kazakhstan 276 1 3
117 Namibia 275 -1 -2
118 Zimbabwe 267 0 2
119 Tajikistan 264 0 0
120 New Zealand 261 0 5
121 Guinea-Bissau 255 0 0
122 Thailand 248 0 0
123 Malawi 247 0 0
124 Chinese Taipei 242 -1 -5
125 Togo 241 0 0
126 Antigua and Barbuda 239 0 0
127 Lithuania 238 -1 -1
128 Sudan 232 0 0
129 Turkmenistan 229 1 0
129 Andorra 229 1 0
131 Latvia 228 -2 -2
131 Nicaragua 228 1 0
133 Yemen 227 0 0
134 Swaziland 220 0 -2
135 St. Kitts and Nevis 218 0 0
136 Rwanda 216 0 0
137 Angola 209 0 0
138 Botswana 206 1 1
138 Myanmar 206 0 0
140 Tanzania 200 0 0
141 Kosovo 197 0 0
142 Hong Kong 195 0 0
143 Equatorial Guinea 190 0 0
143 Solomon Islands 190 0 0
145 Afghanistan 188 0 0
146 Guatemala 180 0 0
147 Lesotho 179 0 4
148 Burundi 174 0 0
149 Comoros 172 0 0
150 Maldives 171 0 0
151 Ethiopia 166 0 0
152 Dominican Republic 162 0 0
153 Suriname 154 0 0
154 New Caledonia 150 0 0
155 Mauritius 140 0 0
156 South Sudan 132 0 0
157 Tahiti 130 0 0
158 Liberia 129 0 0
159 Kuwait 128 0 0
160 Barbados 127 0 0
161 Nepal 118 0 0
162 Vanuatu 117 0 0
163 Belize 115 0 0
164 Indonesia 111 0 0
165 Cambodia 103 1 0
165 Papua New Guinea 103 1 0
167 Grenada 102 1 0
168 Fiji 101 -3 -3
169 Singapore 99 0 0
170 Malaysia 96 1 3
171 St. Lucia 95 -1 0
172 Gambia 92 0 0
173 Moldova 89 2 1
173 St. Vincent and the Grenadines 89 0 0
173 Puerto Rico 89 0 0
176 Chad 88 -1 0
177 Dominica 86 0 0
178 Bermuda 82 0 0
178 Laos 82 0 0
180 Liechtenstein 80 0 0
181 Cuba 75 0 0
182 Guyana 73 0 0
183 Bhutan 68 0 0
184 Malta 65 0 0
185 Macau 60 0 0
186 Mongolia 51 0 0
186 Sao Tomé e Príncipe 51 0 0
188 Seychelles 48 0 0
188 Aruba 48 0 0
190 Guam 45 0 0
190 Timor-Leste 45 0 0
192 American Samoa 38 0 0
192 Cook Islands 38 0 0
194 Bangladesh 35 0 0
195 Gibraltar 34 0 0
195 Brunei Darussalam 34 0 0
197 Samoa 32 0 0
197 Djibouti 32 0 0
199 US Virgin Islands 18 0 0
200 Sri Lanka 17 0 0
201 Pakistan 10 0 0
202 Cayman Islands 9 0 0
203 San Marino 8 0 0
204 Montserrat 4 0 0
204 British Virgin Islands 4 0 0
206 Turks and Caicos Islands 0 0 0
206 Tonga 0 0 0
206 Eritrea 0 0 0
206 Somalia 0 0 0
206 Bahamas 0 0 0
206 Anguilla 0 0 0

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.

68 comments:

  1. Well that wad underwhelming, the two finalists moving up a combined 8 places while Germany and Brazil who underperformed maintained their positions.

    ReplyDelete
  2. France 115 points for a World Cup win! And Germany number 1. Yikes for this new ranking system.

    ReplyDelete
  3. Wouldn't the base point value be converted to the new point system too? That way the results would be more accurate?

    Keeping the old results points and moving to the new point method as an input means that Germany and Brazil maintain their lead. It doesn't seem correct to me

    ReplyDelete
    Replies
    1. I just read the first line **

      All good, the initial points will then change :)

      Delete
    2. I suggest this method should be used next time. I'd want that we (Croatia) go to a No.4, while afterward this new system would protect us not to fall to quickly. If they already now use this new system we'll never be closer to the top ten, while in event they use what I've suggested we would stay a while among the top ten :)
      Just joking. Seriously speaking - I understand the idea that one occasion under performance don't really mean that some big football nation should fall for ten or so places. But - on the other hand - if they are a big team, they should soon return to the former place. Applying these rules mean protecting the big ones - further differentiation of big and small. I'm only amused to imagine Italy and Netherlands trying to reach same place among top ten, maybe even among top five :)

      Delete
  4. I hope this won't be the actual method. It is utterly ridiculous. It would basically ensure that everyone remains forever where they are.

    ReplyDelete
  5. I've read ELO ratings needs about 30 matchs per team to be quite accurate.
    They perhaps should start with Fifa ranking from 4 or 8 years ago and alter it with all results since then.
    Otherwise, we'll have to wait for many years to have an accurate FIFA Ranking !
    By now, Germany have a huge amount of points because of winning WC 2014 and Confederations Cup 2017 (with a B Team) !

    ReplyDelete
    Replies
    1. They could just as easily start with the very first England-Scotland game and add every result since.

      Delete
  6. What a complete load of crap!

    France 4th after winning the World Cup and they already were #7 before the competition.

    Germany #1 after failing to make it pass the group stage in the World Cup?
    Croatia #15 after making the World Cup final?
    Chile #8 after failing to qualify for the World Cup?
    Portugal #5 in what world?
    USA #23 when they can't even qualify for the world cup out of the CONCACAF?

    What an absolute joke!

    You can ask 100 experts and none would have these teams ranked anywhere near where they are.

    ReplyDelete
    Replies
    1. Totally agree.... how could Croatia be ranked lower than Poland !...etc

      Delete
  7. All in favour of bringing the ELO system in, just don't know why they wouldn't wait until August to do so for a fairer reflection going forward.

    ReplyDelete
  8. Actually Portugal in 5th is not that far fetched Sylvain. They are still the Euro Champions.

    ReplyDelete
    Replies
    1. And that's exactly why they still are ranked so high with the new system, just like Germany. With the old system, the results of Euro2016 would be devaluated from 50% (timeframe 2) to 30% (timeframe 3), which would mean a drop to 10th for Portugal. That is imho a far more better reflection of their current strength.

      Delete
  9. I would be interested to see the simulations for the CONCACAF Nations league qualifying competition

    https://en.wikipedia.org/wiki/2019%E2%80%9320_CONCACAF_Nations_League_qualifying

    Which 6 teams make League A and play in the 2019 Gold Cup ?
    Which 4 teams make League B and play in the 2019 Gold Cup ?
    Which 8 teams make League B but do not play in 2019 the Gold Cup ?
    Which 12 teams make League C ?

    ReplyDelete
    Replies
    1. Ted, that is on my to-do list :) Will be somewhere in the coming weeks.

      Delete
  10. I don't thik this will be the new ranking.
    With the constants listed on the first Fifa document (60, 50, 40...), I tried some simulation and the new system works well with a range of 600 points beetwen the first ranked team and the last. But even a range of 1000 points (the same of the Fifa women World Ranking) can be work decently. I expected there will be some formula for the conversion from actual points to new ratings, mantaining the same positions.
    However, if my assumption will be correct, some consideration about the new system:
    - The choice to not consider the goal difference (different to the women world ranking), makes the system quite different to the traditional ELO. Wins and losses will be more important respect the strenght of the confederation you play in. European and south american nations of medium level can be ranked worse than nations who plays continental finals in Asia and Africa, and minnows european nations who loss almost all their matches could be ranked very low. In this aspect, the new system will be similar to the old system, but with the positive aspect that it eliminates the problem of the distorted ranking due to friendlies.
    -draws will not move many points, respect wins and losses.
    -In this system without goal difference every loss, even with strong teams, costs points. I expect there will be a minimum (maybe 1000 points like women ranking) that teams at the bottom of the ranking can't exceed. Otherwise teams like San Marino or Gibilterra, playing more than 20 matches every 4 years, will end far to Tonga who plays only 3 match in the same period. And the occasional draws they can achieve are not enough to regain. IRL I think Gibilterra or San Marino can beat Tonga every time they plays together.
    However I don't like very mutch the ELO formula apllied to football, the biggest problem is the difference between the number of matches played in each confederation, and excessive slowness to adapt to recent results.
    In my system, after world cup, I have now France, Brazil, Belgium, Croatia, Uruguay, England, Colombia, Spain, Argentina, Germany in top 10, and I continue to prefer it.

    ReplyDelete
  11. What actually is the ultimate purpose of such a ranking? Isn't it to reflect the current strength of the teams, and to give an idea about who would be most likely to win if two given teams were to play right now? If so, it would make sense if more emphasis was put on recent results than on how teams did 2+ years ago.

    ReplyDelete
    Replies
    1. For seedings in qualifiers and tournament finals

      Delete
  12. Obviously, we'll find out what FIFA has done soon enough, but it seems utterly bizarre to think that they would decide a team shouldn't lose points for a knockout-round loss, but they could lose points for a knockout-round draw where they advance on penalties.

    ReplyDelete
    Replies
    1. Yes, quite unbelievable that no one at FIFA saw this as they were designing the new system.

      Delete
    2. It would be bizarre, but the previous ranking system had a similar flaw where a worse result on the pitch could be better for the ranking. If a two-legged tie went to penalty shoot-out, the PSO result overrode the 90-minute result of the second leg. http://www.football-rankings.info/2009/09/fifa-ranking-flaw-in-calculation.html

      I do assume that FIFA means that you can never lose points in a knockout match, but they've exhibited enough incompetence over the years for me to not be sure.

      Delete
  13. No updates on FIFA site yet...

    ReplyDelete
  14. Even in that case Russia will gain 22 points for the match vs. Croatia despite being eliminated, and Croatia will in best case not lose any points. I don’t know, it may be better to award a PSO winner 0.75, and loser 0.25.

    ReplyDelete
  15. Next FIFA ranking will be published on August 16th. The July ranking will be skipped.

    ReplyDelete
  16. Thus if I understand there aren't publications of Fifa Ranking this month

    ReplyDelete
  17. LOL Did FIFA realise their July ranking would be seen as a big joke? What are they hoping to achieve by skipping a month?

    ReplyDelete
    Replies
    1. Probably to change the ranking system in the meantime

      Delete
  18. Ed in what this postponement might do to the ranking for August

    ReplyDelete
  19. There aren't rankings in July I think . I don't hope anymore

    ReplyDelete
  20. It was a blessing that neither Germany nor Brazil won the World Cup. Had there been an expected winner, we would have been stuck with a ranking that never changes for the next couple of World Cup cycles at least. The previous system was used for 12 years despite its obvious flaws.

    Now FIFA simply cannot release a ranking that keeps Germany on top. The ridiculousness of that would be obvious to everyone. It still baffles me that apparently FIFA didn't test or simulate their system at all. Why else would they flake now? It's been known for nearly two weeks now that Germany would stay on top with the parameters as stated. It doesn't take two weeks to re-calibrate the parameters if you have any foresight.

    ReplyDelete
  21. This comment has been removed by the author.

    ReplyDelete
  22. Using the Webpage Archive, the June Rankings article was changed sometime recently from "The next FIFA/Coca-Cola World Ranking will be published on 19 July 2018." to "The next FIFA/Coca-Cola World Ranking will be published on 16 August 2018."

    ReplyDelete
  23. How possible is it that FIFA converts the previous results (maybe from the past 2 years) to their new ELO system and start from there, instead of starting fresh from the June rankings?

    ReplyDelete
    Replies
    1. What they need to do instead is either cut down the initial gaps or change the parameters of the calculation method or both. Say, if the June ranking points were divided by three, the July top 4 would be 1st Belgium, 2nd France, 3rd Brazil, 4th Germany (if my math is correct). Drop the rule that a team can't lose points in a knock-out game and Belgium and France swap places.

      Delete
  24. I think the new formula will work fine in the long run since it will close the gaps at the top over time. The problem is the gaps at the top were too big to close over a single competition resulting in no major changes near the top. By 2022, the top 10 will probably be close enough that the 20 to 40 points changing hands will result in actual ranking changes.

    ReplyDelete
    Replies
    1. I'm afraid it will not. The parameters are flawed in a way that it takes decades rather than years for any catching up to happen. Top teams play roughly a dozen high-value matches (World Cup or continental finals) in a four-year cycle and can't lose points in half of them! Qualifiers and friendlies are nearly meaningless.

      Someone on Bigsoccer calculated what the pre-World Cup points would have been had the new system been adopted after the 2014 World Cup.

      http://www.bigsoccer.com/threads/fifa-rankings-world-cup-seeding-2022-edition.2078808/page-3#post-36871596

      Compare them with the starting values (July 2014) if you dare: https://www.fifa.com/fifa-world-ranking/ranking-table/men/rank=240/index.html

      For example, Netherlands would have gone from 3rd in July 2014 to 5th in June 2018 after not qualifying to either Euro or the World Cup. Greece would have gone from 13th to 19th after likewise twice failing to qualify and finishing last in their Euro qualifying group. The ranking never changes with the parameters as announced.

      Delete
    2. BTW I'm the person who made that post on BigSoccer. :)

      If you double the k values, the Netherlands would have entered this World Cup in 11th. Still not great, but a lot better than 5th.

      Delete
  25. It would only be fair to use the old formula for July, as otherwise the results of the 2014 World Cup will effectively have more weight than those of the 2018 WC, which really makes no sense.

    ReplyDelete
  26. Keep in mind there are no matches scheduled until the publication of the August ranking.

    ReplyDelete
  27. Why's it being skipped this month?

    ReplyDelete
  28. Likely because FIFA figured out that the new ranking would have not been representative and would have been highly criticized so they have to reconvene. :)

    ReplyDelete
  29. If they wanted to do it right, they should scrap current rankings altogether, and calculate using Elo formula starting from scratch from a date 10 or 20 years ago.

    If they use current points as a starting point it will take decades before we get reasonable rankings.


    Also, there should be some discounting of old games, perhaps in 4 year cycles, ie from one World Cup to the next.

    ReplyDelete
    Replies
    1. If FIFA is considering this (and that would be a good starting point) they should do this from 1993 onwards. In december 1992, the FIFA-ranking was born, so when you want to do it right, calculate the points retrospecively from January 1993 onwards. Only thing to figure out is what starting points you're giving to the countries: all the same, or do you take the positions of the Dec 1992 ranking as a measuring point for the relative strength of a country. I think I would to the latter.

      Delete
    2. Yes, why not. My guess is that it wouldn’t matter much on such a long time scale.

      Delete
  30. Guys, keep in mind that the PDF where FIFA describes the new method says that it will seamlessly convert form the old ranking system into the new one, so no recalculation of previous games.

    I think the problem lies with the parameters for Elo. They went for 600 points as a difference for a team being 10x better. That means if team A has 1600 and team B 1000 points, it will beat team B 10 times for every 1 time B wins.

    Then they set the weight of World Cup final to 60. So if suddently team B started to beat team A game after game, it would take them more than 5 games to meet at 1300. And if they played a friendly game (weight 10), team B would need to win 50 times in a row, to bridge the gap to team A... There are just not enough games for these weights. Also, not losing Elo points in knockout stages is very controversial.

    ReplyDelete
  31. Lots of valid comments here, but in short the main problem is the k-factor being used. You could start with the same old ranking and points, but allow for bigger points exchange per match - that would improve the ranking considerably.

    Regarding the no point loss in the knockout rounds, I think that it is a valid point to try and balance more points for teams that get far and win competitions, and avoid taking out too many points from teams that get far in the knock-out rounds and could suffer a significant point drop due to the k coefficient of every subsequent match being higher.

    It should be a matter of tweaking the draw, penalty shootout cases, but completely doable in my opinion.

    ReplyDelete
    Replies
    1. If the FIFA would double the k-factor, the rankings would look a lot better:

      (Rk PrevRk Cntry Pts Diff)
      1 3 Belgium 1460 162
      2 2 Brazil 1445 14
      3 7 France 1429 231
      4 1 Germany 1381 -177
      5 4 Portugal 1231 -43
      6 20 Croatia 1206 261
      7 6 Switzerland 1204 5
      8 13 England 1167 116
      9 5 Argentina 1155 -86
      10 9 Chile 1137 2
      11 14 Uruguay 1127 109
      12 10 Spain 1123 -3
      13 12 Denmark 1095 44
      14 11 Peru 1055 -70
      15 8 Poland 1053 -130
      16 24 Sweden 1033 153
      17 15 Mexico 1018 29
      18 16 Colombia 1012 26
      19 17 Netherlands 981 0
      20 18 Wales 953 0

      And as far as the knockout-rounds are concerned: FIFA just needs to be consistent in their approach. If the case is that losing teams in a knockout-game do not lose points, no matter if the loss is after regular or extra time or after PSO, that would be clear. And winning in regular or extra time is regarded as a win, but winning after PSO is regarded as a draw, that would be clear too. And even if they regard winning after PSO as a win, that would be fine too. But be consistent and transparent about it.

      Delete
    2. I can't see a good justification for the rule that you can't lose points in knockout games. The same justification that is used for that could also be used for World Cup group stage games. If you can lose points in the group stage, narrowly qualifying and losing three games in the group is worse for the ranking than narrowly failing to qualify. Hence, you shouldn't lose any points in the group stage either.

      We can go on. The same thing applies to all continental championships except Copa América (which has no qualifiers). And that's still not the end of it. In any multi-round World Cup or continental qualifiers, it should be only possible to lose points in the first round in order to not penalise any team for advancing in the competition. In Nations League, everything except the lowest league should have ranking protection. So... no. Just, no.

      If you have to reward teams for advancing in the tournament, I'd rather give them bonus points for final placings (eg. 100 poins for World Cup winners, 75 points for runner-ups, 60 for 3rd place, 50 for 4th, 40 for QF, 30 for R16, 20 for 3rd in group, 10 for 4th in group). The teams would keep these points until the next World Cup. This would come with its own set of problems, but at least this would not cause ranking inflation because the number of bonus points would stay constant. Knock-out matches wouldn't create ranking points out of thin air (a very un-Elo-like feature).

      Delete
    3. And creating ranking points out of thin air slowly diminishes the impact of the elo formula. The teams benefiting from it (the teams playing the KO rounds, which means the stronger teams) will slowly drift away from the lesser teams and stay seeded for ever.

      Delete
  32. What rank India
    After won intercontinental cup

    ReplyDelete
  33. This ranking is hahahaahahaha.Greatings from Croatia

    ReplyDelete
  34. This ranking is hahahaahahaha.Greatings from Croatia

    ReplyDelete
  35. This is going to sound crazy but why not start everybody with the same number of points? A fresh start for all the countries. If the new point system works, let’s let it play out. You get points if you win. You get more points if you beat good teams. You get more points if you advance in big tournaments. The past is the past. So let’s see who is good now.

    The standings with fluctuate a lot at the beginning but won’t that be kind of fun to watch? After a year or two they will reflect who is good now and isn’t that what they are suppose to do? If associations need to rank teams for upcoming tournaments they can use FIFA rankings if they want or use their own rankings or whatever they want. (I think UEFA does that for some things).

    If our complaint was that the old FIFA rankings were biased why not remove all the bias and start over. The good teams should rise to the top and if they don’t maybe they’re not that good.

    ReplyDelete
    Replies
    1. http://www.goalie365.com/show_fifa_worldcup_news.php

      Delete
  36. Has anybody run the calculations from June to the August ranking? I'm no Excel expert but I always found it easy to calculate Australia's points in the old method but am struggling to get it right after just 4 games in the new method. Anyone happy to cast their eye over this?

    9/6 v Hungary (PBefore AUS=1460 / HUN=1400) P=1464.43
    16/6 v France (PBefore AUS=1464.43 / FRA=1575.31) P=1444.67
    21/6 v Denmark (PBefore AUS=1444.67 / DEN=1585.82) P=1451.28
    26/6 v Peru (PBefore AUS=1451.28 / PER=1512.67) P=1429.21

    FIFA points for AUS at August release were 1430.71.

    If I use the same PBefore for all sides and all games I get 1430.16. Closer, but still not accurate.

    I know that's limited data but can anyone see where I might have gone wrong? Thanks in advance.

    ReplyDelete
    Replies
    1. Nigel,

      your PBefores seem incorrect:
      2018-06-09; AUS = 1464,000; HUN = 1411,508; I = 5
      2018-06-16; AUS = 1466,249; FRA = 1575,676; I = 50
      2018-06-21; AUS = 1446,422; DEN = 1583,524; I = 50
      2018-06-26; AUS = 1452,852; PER = 1512,611; I = 50; P = 1430,706

      Delete
    2. Hungary has played on June 6th against Belarus already, their PBefore can't be their starting value. Also their starting value is not 1400 (that you use), but 1412.

      It seems your starting values are not correct. Read the way FIFA determined the starting values carefully: if 2 teams have the same ranking position then they get the same starting value and the next team gets that same starting value minus 4, not minus 8.

      Delete
    3. Excellent. Thanks mate. Really appreciate your help and the work you do on this site.

      Delete
    4. I'm having trouble calculating the expected result of the game from the last game played by Australia. Can you demonstrate on how to calculate that part of the equation to get the final value of 1430.706?

      Delete
    5. Australia played their last match until now against Peru in the World Cup group C. The I-factor for the match = 50. The PBefore for Australia = 1452,852 and Peru = 1512,611.

      This leads to a win expectancy (We) for AUS of 1 / (1 + 10 ^((1512,611 - 1452,852) / 600)) = 0,443 and for PER of 1 / (1 + 10 ^((1452,852 - 1512,611) / 600)) = 0,557.
      The actual result of the match AUS-PER was 0:2, so W for AUS = 0 and for PER = 1.
      This leads to a points difference of (W - We) * I = (0 - 0,443) * 50 = -22,146 for Australia and a points difference of (1 - 0,557) * 50 = 22,146 for Peru.
      Resulting points AUS = 1452,852 - 22,146 = 1430,706 and PER = 1512,611 + 22,146 = 1534,757.

      Delete
    6. I'm trying to get the accurate result of the June 9 match up between Latvia and Azerbaijan. I'm getting the wrong result for Latvia. Do you know what the before ranking value for both countries before the match up is I'm came up with the after ranking for Latvia of 1121.93 instead of getting 1121.92. Here's how I'm calculating it.

      1205.289-1124.038= 81.251/600= .135418333
      10^.135418333 +1= 2.365898202
      1/2.365898202= .422672454
      1124.038 +5(0-.42267)
      1124.038- 2.113 = 1121.925 1121.93
      Can you show me the me the correct math calculation and show me where I went wrong?

      Thanks

      Delete
    7. Nothing wrong here. Latvia's points after the match are 1121,925. When rounding to two decimals any normal person would come up with 1121,93. Not FIFA though. In their August ranking Latvia has 1121,92 because they apply an exotic rounding method: rounding to the nearest even number.

      Delete
    8. In case of rounding exactly a half, that is. 1121,9251 is 1121,93 for FIFA also.

      Delete