Welcome to World Footy News
Tuesday, August 11 2020 @ 06:13 am ACST

The following comments are owned by whomever posted them. This site is not responsible for what they say.
IC2011 - Divisions
Authored by: Brett Northey on Wednesday, April 21 2010 @ 12:51 pm ACST


I agree avoiding 1st playing 15th or 20th is a useful / important asset of divisions, and I think they need to be used.

And I can see that in terms of the rankings it helps sort out teams that are close together, rather than subject them to big jumps from playing teams much higher.

However, I still think there is a weakness in the case I mentioned. Maybe that negative doesn't outweight the positives.

But let's look at a simplified case of a 4 team world. In order of ranking, A down to D. So split them into 2 divisions: A, B and separately C, D.

Assume all games go to form and also to emphasise the point A play B twice and C play D twice. A edges further up in ranking points, B edge down. Similarly C edge up and D edge down.

We may now have the situation that C edges above B. Yet all they have done is beat a lower ranked team. And B's only "crime" is to have been good enough to play in the top division, which has caused their ranking to drop.

So taking that back to the true IC2011 case, yes the teams at the bottom of Division One would have the chance to rise by beating teams above them, but if things go to form/ranking, their points will drop, and similarly the top teams of Division Two will rise. By the end of the tournament, assuming no cross-over games between Divs One and Two, we could see the lower middle teams rise above the upper middle teams, without really having earnt it.

So when I say it our rankings need a wider cross-over of games between higher and lower ranked teams, I really mean across that division divide. Say it separates into 1 to 12 and 13 to 24. Then not 1st playing 13th, but games like 10th playing 13th or 14th.

I'll test some scenarios out. I suspect the effect will be subtle enough that it will be washed out by issues such as not all teams attending so not as many closely ranked teams across the divisional divide, upset results mixing things up, and increasing numbers of one-off tests and other tournaments, allowing cross-divide games to inform the system.

---
Brett Northey - Co-founder and Chief Editor of WFN

[ # ]
IC2011 - Divisions
Authored by: Cam Homes on Wednesday, April 21 2010 @ 02:56 pm ACST

Yep C might edge above B as you say, but, by reducing the number of 2 max games which we have had plenty in the previous IC's, points exchanges are smaller, might only be 0.2 or 0.5 instead of 2.00. B doesn't get any -2.00's so reducing the likelyhood of dropping below C. If C isn't getting any +2.00's even less likelyhood of climbing above B and because all the teams(half dozen or more) are close in rank then the odd wins that the B's get won't have their rise wiped out by those big 2 max losses.
Top 10 at present are all over 41 pts. not count Peace team the next bunch are below 39(2 teams) and the rest below 36pts. No 2 max games would not see the bottom ones rise above those above 41pts. not for at least a two or three games. I stiil reckon all the possible problems/apparent anomalies that appear to be in the system is because of the lack of numbers, teams and matches. So as Aaron says as more teams enter and we have more matches and we're not likely to use it until around or after the IC and we see how the IC pans out with numbers etc tweeking it is not a problem. Tweek too much and it might create bigger problems later when we have more teams and matches.
Cam

[ # ]