Since that FA already used the list to get an opponent, I thought I should post some of that info. Who knows, maybe other FAs will use it... :D
Probable August 2010 FIFA Ranking - Team
| 34 | Peru |
| 16 | Greece |
| 36 | Slovakia |
| 45 | Burkina Faso |
| 48 | Latvia |
| 21 | Paraguay |
| 58 | Benin |
| 9 | Portugal |
| 52 | Belgium |
| 15 | Russia |
I double-checked. Peru will have 726 points in August 2010, good enough for 34th place.
Isn't the best opponent country dependent, that is for a given team the best opponents list is different, considering their strength differences?
ReplyDeleteWhat system did you use? Was it using ELO to evaluate the points expectancy?
Thanks,
Amir
It's a different approach. I used Elo Ratings to compute the FIFA points/ actual effort ratio.
ReplyDeleteWith countries like Greece, Portugal, Paraguay and Slovakia which will play in the world cup and might have a run of 7 games would not that change very much their points in August? World cups are unpredictable, if there is any rule is that home nations do well, and that usually a team looking hopeless ( Argentina?) will win it.
ReplyDeleteThis is based on Elo ratings - and the teams you mentioned didn't have very good results.
ReplyDeleteSlovakia - 2 draws and a loss. Out in the group stage.
Paraguay - a win, a loss, a draw. Out in the second round.
Greece - 2 draws and a loss. Out in the group stage.
Portugal - 2 wins and a loss. Out in the second round.
If they perform better than expected, it will be even better for their August 11th opponents.
The Elo rating has its flaws, but until I'll have the time to come up with a better one (see Voros or ESPN), that's the one I use.
As I already explained this is based on (FIFA points for a win) / (actual effort from Elo rating) ratio.
Slovenia, Gabon, Algeria were also in the list, but are no longer available on August 11th.
If you go by eloratings to predict world cup rankings, then you have to consider as well that the team picked to play a friendly might very well beat their opponents. Portugal would only be a good opponent for teams 1-8 then. Or is it better to loose to a team ranked higher than beat a team ranked lower?
ReplyDeleteTeams important enough will have their own agenda for friendlies than just being asked, they might want teams similar to next opponents, euro 12 qualification for example, or teams which will guarantee a full stadium or please their sponsors.
By the way, while I admire your dedication and analysis, I got to warn you that using eloratings to try to predict a world cup is better taken with not a grain, but a few tons of salt.
You are right. That's why I warn the FAs when sending the lists.
ReplyDeleteThis is a new approach - I mainly draw up "ideal opponents" lists based on individual rankings - as Amir pointed out.
You are also right about choosing their opponents for friendlies. There are some FAs (not many) also interested in climbing in the rankings.
Thanks for your nice words.
I'm totally lost as to what this post is about? Best opponent by ranking for a friendly?
ReplyDeleteYes - best opponent based on (FIFA points for a win) / (actual effort to win ratio) for the August 11th FIFA friendly match date.
ReplyDeleteI'm guessing it was Australia who approached for this data...hence why they have lined up a friendly with slovenia in august.
ReplyDeleteNo, clifton, it wasn't Australia. Although, I'm pretty sure they do have someone working towards improving their ranking. I've had many Australians asking questions about the calculation method. Who knows...
ReplyDelete