Elo Ratings
This article is about the Elo Ratings. For the old rankings, see Rankings.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The Elo Ratings for the 4chan Cup were created in the attempt to create a more precise ranking system to reflect the strength of the 4chan Cup teams. This system is an alternative to the Rankings which is the based on the official FIFA ranking system for men's national football teams.
A teams's Elo rating is represented by a number which increases or decreases based upon the outcome of games between rated teams. After every game, the winning team takes points from the losing one. The difference between the ratings of the winner and loser determines the total number of points gained or lost after a game. If the high-rated team wins, then only a few rating points will be taken from the low-rated team. However, if the lower rated team scores an upset win, many rating points will be transferred. The lower rated team will also gain a few points from the higher rated team in the event of a draw. This means that this rating system is self-correcting.
The 2014 Autumn Babby Cup was the last tournament where the elo ratings was updated. Following a test run between the 2014 4chan Summer Cup and the 2014 Autumn Babby Cup, it was clear that the elo ratings had many more inaccuracies when compared to the original ranking model. The only upside to the elo ratings was that winning teams like /mu/ folliwing the 2014 4chan Summer Cup does not get shafted because of rankings that have not been updated. Besides that, previous winning teams like /trv/ was not punished enough, and newly created teams had an unfair advantage. Basically, not enough matches were played per team and tournament for the elo ratings to converge to a value that made sense.
Current calculation method
This ranking system makes use of the following factors from matches to determine change in points:
- Match Result
- Match Quantifier
- Goal Difference
- Predicted Result
A clear difference between this system and the Rankings is that a points to not deteriorate over time. A team can only gain and lose points by playing matches.
Match Result
Between the two teams in a match a total of 1.0 have to be distributed among the two teams for the match result quantifier(W).
| Result | Match Result |
|---|---|
| Win (no penalty shootout) | 1.0 |
| Win (penalty shootout) | 0.7 |
| Draw | 0.5 |
| Loss (penalty shootout) | 0.3 |
| Loss | 0.0 |
Match Quantifier
The match quantifier(K) is a value which determines how many points is transferred between teams. The higher the match quantifier, the higher transfer of points. This value is not final, and can change if different values is determined to generate more precise rankings.
| Competition Level | Match Quantifier |
|---|---|
| Elite Cup | 150 |
| Babby Cup | 100 |
| Friendlies | 50 |
| Babby Cup Qualifiers | 50 |
Goal Difference
The goal difference value(G) is based on how many goals the winning team won with. Draws and penalty shootouts have a default goal difference value of 1.0.
| Goals | Value |
|---|---|
| +1 | 1.0 |
| +2 | 1.2 |
| +4 | 1.6 |
| +8 | 2.4 |
Predicted Result
Predicted result(WE) is a value which is calculated based on the respective teams points prior to the match. A team with an advantage in points will have a higher predicted result value. The values ranged from 0.0 to 1.0, where a 0.5 value means that the teams is predicted to be equal in strength.
- 1 / 10(-(Team - Opposition) / 400) + 1)
Team and Oppositionis, respectively, the point tally of the team which is having they predicted result value calculated, and the opposing team. The same calculation is done for the opposing team, with the Team1 and Team2 values switched. The two predicted result values for the team should always equal 1.0 when added together.
Ranking Formula
Putting the previous factors together, you get the following formula, which calculates the point tally folliwng the result of the match.
- Team + (K x G) x (W - WE)
The main thing to take from this is that the predicted result(WE) is subtracted from the match result(W). This means that so long a team scores a result better than the predicted outcome, they will gain points.
As an example, lets take the final match for the 2014 Summer Cup between /mu/ and /hr/.
| /mu/ | /hr/ | |
|---|---|---|
| Points | 826.91 | 778.27 |
| K | 150 | 150 |
| G | 1.2 | 1.2 |
| W | 1.0 | 0.0 |
| WE | 0.57 | 0.43 |
| Change | +77.48 | -77.48 |
Other Factors
A disqualification penalty is used whenever a team disqualified from a higher tier tournament. For example, if a team advances from a babby cup, but is disqualified from the elite cup that follows, the team will be subjugated to the disqualification penalty. This is done because it is not possible to lose of defend your points following your disqualification, so it would wrongly bloat your points. A prime example of this is /fa/. /fa/ reached the final in two consecutive Babby Cups, and were disqualified from the following Elite Cup in both cases. Without a disqualification penalty, /fa/ would have been in the top 10 of the rankings following the 2014 Summer Cup.
When a new team enters the 4chan Cup, they all start with exactly 400 points. The only exception to this is all the founding teams of the original summer and winter cups. This is done to prevent new teams entering the cup from immediately being in the top 10.
The point tranfer in 3rd place matches is always cut in half for any level of competition, to prevent the 3rd place team from bloating their points.
External links
Current rankings file updated November 10th 2014
http://www.mediafire.com/?d99s2g4p884b893
| |||||||||||
| |||||||||||||||||