PDA

View Full Version : European NT Coefficient Ranking



nikosextra
10-06-2013, 08:28 PM
I hope this is a correct location for this thread.

Introduction
I think there is a general consensus in this forum and among most basketball fans that the FIBA World Ranking does not depict the actual strength of most teams in the world, especially in Europe, where many good teams fail to qualifyfor big tournaments and end up too low in the table.

Inspired by UEFA national teams coefficients, I tried to come up with a new formula of providing the ranking for European national teams. The general rules of my ranking are:

Coefficients are counted during a 4 year cycle. So the current coefficient is the outcome of the results in the years 2010-2013.
Older results tend to count less towrads the final coefficient. 2010 results are multiplied by 10%, 2011 by 20%, 2012 by 30% and 2013 by 40%. For the next coefficients ('11-'14) 2010 results will not count at all, the rest will be lowered by 10% and 2014 results will be multiplied by 40%.
Matches counted consist of: World Cup, Olympics, pre-qualifying for the Olympics, Eurobasket, Eurobasket qualification (all phases)
According to the strength of the competition and of the particular stage of the competition the match was part of, different bonus are awarded to both teams.
Points scored and conceded also count (a lesser amount)


The only exception to the above was the Additional Qualifying Round for Eurobasket 2011, which was not counted. All the other matches played were counted (World Cup 2011, Eurobasket 2011 Qualification, Eurobasket 2011, pre-qualifying for the Olympics and the Olympic tournament 2012, Eurobasket 2013 Qualification, Eurobasket 2013 and Eurobasket 2015 Qualification A). In the parentheses above, different colours are used for the different years those matches were played (so, a different weight was applied to them).
For 2010-2011, teams that participated in the now defunct Division B, were awarded a base coefficient of 5000.
Teams that did not participate in any competition for a particular year, did not earn any points.
In the case of Slovenia 2012, when they played no games because they had auto-qualified for the Eurobasket as hosts, but had not qualified for the Olympics, their 2011 coefficient was also applied to 2012.

Points awarded
For every match, both teams get 10,000 points for playing. The winner gets an extra 20,000 points. For every point scored, 101 points are added. For every point conceded, 100 points are deducted. There are no draws (overtime result is the only one counted).

Bonus points are awarded to both teams, depending of the competition and the stage of the competition.
Eurobasket
Qualification A Groups -10000
Qualification A Play-offs 0
Qualification 0
Group Phase match 5000
Top-12 10000
QF 15000
5th-8th place matches 10000
SF 20000
3rd place Final 25000
Final 30000

World Cup
Group Phase match 7000
Eighth Finals 15000
Quarter-Finals 20000
5th-8th place matches 12000
Semi-Finals 30000
3rd place Final 35000
Final 50000

Olympics
Pre-Qualifying Tournament 2000
Group Phase 12000
Quarter-Finals 20000
Semi-Finals 30000
Bronze medal match 35000
Final 50000

If a match contains a European team playing a non-European team, points are counted normally.

For example: World Cup 2010 Group Phase, Angola-Serbia 44-94
Serbia gets 10,000 for playing + 20,000 for winning + 7,000 bonus points + 101*94 points scored - 100*44 points conceded = 42,094 points
For the same competition stage: France-Spain 72-66 : France:37,672, Spain: 16,466
For the Olympic Games Final: USA-Spain 107-100 : Spain got 59,400 points despite losing, because of the +50,000 bonus

A team's coefficient for one year is the average points that team collected from the matches they played in that particular year.

2010-2013 Coefficients
http://s.neepic.net/preview/6d8baab6cfc65023ad92ea1d7e5db343.jpg (http://neepic.net/15111)

Joško Poljak Fan
10-06-2013, 09:57 PM
Definately the rankings should be re-done :) Some great ideas imo, however I'd love a system where for example, the number of points would also depend on the opponents ranking as well, meaning that winning vs. Qatar brings less points as losing vs. USA... or simmilar.

Alyosha12
10-07-2013, 07:32 AM
Great job!


Definately the rankings should be re-done :) Some great ideas imo, however I'd love a system where for example, the number of points would also depend on the opponents ranking as well, meaning that winning vs. Qatar brings less points as losing vs. USA... or simmilar.

I like this idea. A win by 50 against Lichtenstein shouldn't count more then a win against Spain by 2 IMO.

nikosextra
10-07-2013, 10:02 AM
It usually won't, since you are probably going to meet Liechtenstein in Eurobasket Qualification A, while Spain in the final tournament, so you get to have bigger bonuses. For example, Austria's win against Luxembourg this year (with a 41 pt margin) gave roughly the same points as losing a top-12 match.

And this is generally the case; losing high-profile matches almost guarantees your average is not hurt much.
See 2012 for example: The Greek failure in the pre-quals was enough to guarantee them more points than every Eurobasket qualifying team, except those with a perfect record. Therefore, a team is rewarded for taking part in more difficult competitions, as long as they have justified their place there.

A system that works with relative strengths is the ELO Ranking, but it has the disadvantage that it counts all matches since the start of time and teams' positions are much more "stable". This would not have worked well if the ranking was used to decide seeds or award wildcards for example.

nikosextra
09-07-2014, 10:42 PM
I 'm not really sure if anyone's interested, but what the heck?

I updated the ranking with this year's Eurobasket qualifiers and the World Cup. So pretty much, 8 more matches to go until the final ranking. The top3 remains the same, but Lithuania is closing in on France. Russia is on a free fall while Bosnia and Latvia are going up.

http://s.neepic.com/dqEGJN28.png (http://neepic.com/dqEGJN28)

nikosextra
09-15-2014, 05:10 PM
Final rankings after World Cup 2014. WC2010 results are now discarded.

http://s.neepic.com/e2R1GKWB.png (http://neepic.com/e2R1GKWB)

Dtown
09-23-2014, 01:42 AM
An excellent ranking I prefer the standards and timeline to the actual FIBA

nikosextra
06-04-2015, 12:52 PM
Here is a mock draw using my coefficients to determine pots and following FIBA's draw rules regarding hosts, co-hosts and draw order.

Procedure
1. Teams ranked according to their coefficients and divided into 6 pots (1-6) of 4 teams
2. Hosts and Co-hosts France: Pot 1 (Finland: Pot 4), Croatia: Pot 2 (Slovenia: Pot 1), Latvia: Pot 3 (Estonia: Pot 6) & Germany: Pot 3 (Turkey: Pot 4)
3. POT 1: France drawn first (Group A), the rest randomly drawn
4. POT 2: Croatia paired with Slovenia (Group D), the rest randomly drawn
5. POT 3: Latvia & Germany drawn first in Groups B & C, the rest randomly drawn
6. POT 4: Turkey paired with Germany (Group C), Finland paired with France (Group A), the rest randomly drawn
7. POT 5: All randomly drawn
8. POT 6: Estonia paired with Latvia (Group B), the rest randomly drawn

http://i61.tinypic.com/4r5ena.png (http://i61.tinypic.com/4r5ena.png)

Since FIBA is changing their competition formats and cycles, I'm going to make some changes and come up with a new coefficient-system, under the same general guidelines.

Victorious
06-04-2015, 04:39 PM
Nice and fair draw. But FIBA wants weak groups so that strong countries can eliminate each other and weak countries can reach far. Hence grow the game of basketball.

nikosextra
10-28-2015, 08:47 PM
Update after Eurobasket 2015 final results.
Question: Does anyone know when the qualifiers for EB2017 will start and with what system? I couldn't find anything, so the countries that did not qualify for EB2015 have their 2014 result copied in 2015.
Note: Yes, I know. It's weird that Spain won the Cup but they were overtaken by 3rd France. But my system punishes Spain a lot for those 2 losses in the first phase. Probably in next editions I will lower the bonus for those games or increase the bonus for knock-out games.

https://s.neepic.com/LqeZN5WB.png (https://neepic.com/LqeZN5WB)

Modis
11-15-2015, 09:47 PM
How many years does your ranking calculation table go back to ? For example did you start from 1990 or even earlier ?

nikosextra
11-15-2015, 09:53 PM
If you read the first post you will more or less understand it. Only the last four seasons count towards the coefficient, reducing in effect as you go back in time.

Modis
11-15-2015, 10:06 PM
If you read the first post you will more or less understand it. Only the last four seasons count towards the coefficient, reducing in effect as you go back in time.

I understand the system but what was the year you started the count. The oldest one I see is 2010.

nikosextra
11-15-2015, 10:13 PM
Yes, 2010 is the first year I calculated, since my first ranking was posted in 2013. If you are interested, I could maybe do a calculation of previous years and see what I come up with, when I have some free time (say Christmas holiday)

Modis
11-15-2015, 10:32 PM
Yes, 2010 is the first year I calculated, since my first ranking was posted in 2013. If you are interested, I could maybe do a calculation of previous years and see what I come up with, when I have some free time (say Christmas holiday)

Yeh. Nice work keep it up ;).

nikosextra
11-15-2015, 10:43 PM
Thanks for the feedback! I appreciate it.

Hepcat
06-15-2016, 04:16 PM
Good stuff! I'll be interested in seeing this table again after the Olympics.

:cool:

amirb
09-23-2016, 02:59 PM
Any update after Eurobasket 2017 Qualifying?

Modis
11-07-2016, 12:23 AM
Any Update after 2016 Olympics.:)

judasmartel
12-31-2016, 07:07 PM
@OP

I think the ranking points are too big, like, 100k points for a game? Maybe divide everything by a power of 10 to make it less intimidating? Other than that, I think it's a good idea to adopt a ranking system similar to UEFA or FIFA.

What are your thoughts about the Elo Rating System with points scored in a game taken into account? One common criticism of the Elo is that it doesn't care whether you lost by 1 or by 50, if you lost, you lost, period.

nikosextra
08-19-2017, 10:05 PM
After a long time, I have updated this ranking for 2017, of course still missing the Eurobasket results. Some of the bonus points have been altered, also as a result of the coming changes to the system. A lot of 'technical scores' due to the absence (to my knowledge) of any action for many teams in 2015. I am thinking of changing the cycle to reflect competitions instead of years in the next edition.

Here's a a taste:

https://s30.postimg.org/w8aikk6xd/Basket_Ranking.png (https://postimg.org/image/p52n4y1hp/)

Modis
08-26-2017, 01:34 AM
Netherlands is 6th ?

nikosextra
08-26-2017, 06:42 PM
They are, since all Eurobasket trams haven't got any points yet for this year :) The final rankings will be completed after Eurobasket.

Mindozas
08-26-2017, 06:46 PM
Netherlands is 6th ?

Better let's get drunk again :D

Modis
08-26-2017, 07:51 PM
Better let's get drunk again :D

round 2. Its Saturday night :cool:

nikosextra
09-18-2017, 08:48 PM
Update after the Eurobasket games. I have slightly changed the bonus points. What do you think?

https://s26.postimg.org/k0e24qb8p/Basket_Ranking.png (https://postimg.org/image/s5w42vzhh/)

CoachZ
09-19-2017, 08:41 AM
Spain coming ahead of Serbia in both 2016 and 2017, while winning bronze and Serbia winning silvers makes little sense. I suppose you are focusing more on wins and losses, then the actual end result.

nikosextra
09-20-2017, 09:12 PM
Hey guys, thank you both for your input.

@LuDux. Yes, Fiba have their own ranking but it is very flawed. It involves very old achievements that cannot be considered current. Looking at the ranking you provided is enough to make this obvious.

@CoachZ. It is true that sometimes having many victories in the early phases of a tournament distorts the final score a bit, even after an early qualification. See Lithuania and Greece for example this year. This is something that can be changed by increasing the bonus points for playing a knock-out match. Regarding the example you mentioned, though, Spain-Serbia, keep in mind that Spain was both times 3rd because they lost to the final winner, whereas Serbia only beat the eventual 4th-placed team. We cannot know what would happen in a Spain-Serbia game during those competitions.

CoachZ
09-20-2017, 09:38 PM
I get that, but it is like saying that Spurs should be ahead of Cavs, since they lost to the champs :D

I think that the final placement should still carry more weight, maybe not a huge amount but it's very curious to have a lower placed team twice ahead in consecutive years.

nikosextra
09-20-2017, 10:03 PM
Hmm okay. I think that one main factor here is that every victory gives 20,000 points, regardless of the match. Probably I should lower this fixed value and hence have higher weight on the bonus points for the match importance. Not sure if this would make Serbia have a better score than Spain, but it wouldn't hurt to try and see how the two systems compare.

LuDux
09-21-2017, 04:01 AM
Didn't realize it's not FIBA rankings thread


Hey guys, thank you both for your input.

@LuDux. Yes, Fiba have their own ranking but it is very flawed. It involves very old achievements that cannot be considered current. Looking at the ranking you provided is enough to make this obvious.



It's ranking of last 20 years, it's supposed to involve very old achievements

amirb
09-27-2017, 09:15 PM
Hmm okay. I think that one main factor here is that every victory gives 20,000 points, regardless of the match. Probably I should lower this fixed value and hence have higher weight on the bonus points for the match importance. Not sure if this would make Serbia have a better score than Spain, but it wouldn't hurt to try and see how the two systems compare.


If the goal of your rankings is showing who the bet team is, then as you said both Serbia and Spain won and lost against the same teams in the top 4, so the other results should obviously distinguish between them. What are the bonus points you used?