Forum

xml	No replies possible in the archive

Author: OhioUefaFan
Date: 05-09-2005, 19:18

I wanted to get some opinion on the ranking system we use here in Ohio for high school football (gridiron). I believe this is allowed because it's a different method of ranking teams, and one that might bring about some interesting discussion about the difference between it and UEFA's system.

You can find the current rankings and previous ones at the site below.

www.joeeitel.com/hsfoot

Now for an explanation...

The system was created in 1971 and first implemented in 1972.

Schools are divided into 6 divisions based on school size. Official counts of male students are done every two years by the state, and the athletic association (OHSAA, our UEFA basically) divides the schools into the six divisions as equally as they can.

There are typically 120-130 schools in each division, and each division is split into four regions. By defeating a team, a school earn points based on what division the opponent belongs to.

Division I = 6.0 points
Division II = 5.5 points
Division III = 5.0 points
Division IV = 4.5 points
Division V = 4.0 points
Division VI = 3.5 points

There are no draws.

The system is totally based on how big the opponents are that you beat, and then how big are the opponents that they beat. Who you beat comprises your level 1 points. Then, every time one of your beaten opponents wins a game, you earn level 2 points.

The level 1 points earned are divided by the number of games played (typically 10) and the number of level 2 points are multiplied by 10 and divided by the total number of games your opponents played (typically 100). Add the two results together and that is the final ranking.

The top 8 teams in each region qualify to the state championship playoffs, with the top 4 given home field in the first round.

There's always been debate in my state over these rankings...I wanted a neutral perspective. If you have questions, I'll try to answer them, but I'm mostly looking for feedback.

Re: Ohio's coefficient system: what do you think?
Author: Lyonnais Date: 05-09-2005, 23:33	Spontaneously, I would say that it looks like complicated.

Re: Ohio's coefficient system: what do you think?
Author: spoonman Date: 06-09-2005, 10:58	Yes, that's exactly what I was thinking. It sounds so... err... American.

Author: kunkello
Date: 06-09-2005, 11:54
Edited by: kunkello
at: 06-09-2005, 11:55

Ohio, I am myself a huge fan of college sports in the US. I don't follow high school too much, not really being affiliated to any state or high school doesn't raise any emotion there.

With this background, I find your comparison highly interesting, and would think that actually the sort of "level 2" points would be a huge benefit for a lot of smaller teams and leagues in Europe that pull the occasional "upset" of big teams and get those victories weighted just as much as beating a team of a level our ranking systems would consider their own.

Take for example: a team from Belarus beating a team from Ireland (or vice versa) with both being neighbors in the country ranking receives the same points for this victory as if they would be beating a team from Belgium, Scotland or Greece (30 spots ahead of them in the country ranking).

In a 2-tiered ranking systems with victory points e.g. weighted by the country ranking points of the team you beat, those "smaller" teams would be boosting their statistical ranking much more and with a five year rolling ranking as we have it in Europe, a few of those "upsets" could boost and reward those teams significantly.

If I just had the time, I might be trying to create such a ranking, but maybe someone is willing to pick up the idea...

Re: Ohio's coefficient system: what do you think?
Author: ignjat63 Date: 06-09-2005, 12:22	The system is based on idea that a team gets extra points by winning against stronger teams. It is a good idea. A similar system is used in chess swiss tournaments. There if two players have the same number of points higher placed is a player whose opponents have a higher rating.The rest of it is technical. Not a bad system.

Re: Ohio's coefficient system: what do you think?
Author: Forza-AZ Date: 06-09-2005, 16:43	Something like that is also used in the FIFA world ranking I believe. You get more points by beating better teams.

Re: Ohio's coefficient system: what do you think?
Author: bert.kassies Date: 06-09-2005, 16:56	See maasranking for an advanced system thats based on similar rankings for USA College Football. It involves solving a system of (linear) equations. Real diehards also read Colly's ranking method But these methods serve different purposes then seeding only.

Author: OhioUefaFan
Date: 06-09-2005, 19:30

kunkello, who do you follow in college sports? Being from Ohio, I'm an Ohio State fan. Speaking of which, the atmosphere at Ohio Stadium Saturday night is going to be spectacular. 105,000 people make a lot of noise.

I tried a while back to use the UEFA system to rank teams in Ohio, but it was about 4 years ago and I didn't have a full understanding of it. I'd be interested in helping you set up an Ohio system for the countries and clubs. We're in the best place for the information.

The key thing to the Ohio system is classification. I'd think you could divide the 52 associations into four divisions of 13. I don't know the populations of each nation at the moment, but on the face of it that would be the most equal setup.

Thanks for the feedback so far. If you've checked the website (by the way, tell me how to set that up as a hyperlink) you'll notice we're only going into Week 3 of play. Things begins to get interesting, ranking wise, after week 5.

Author: ignjat63
Date: 07-09-2005, 06:51
Edited by: ignjat63
at: 07-09-2005, 06:53

OhioUefaFan

even if 52 UEFA members were classifies as you suggest I do not think Ohio system can work here.

The key element in Ohio system (in terms of what makes it interesting) are level2 points. Here in Europe we are used to league system where everybody plays against everybody else in home and away matches (in WC and EC qualifications). And in this case I think level2 points do not make much sense. It makes sense in competitions where a team does not play all of its opponents. For example, in chess round-robin tournaments the additional criteria I mentioned above is not applied.

However, in Euro cups I suppose a team can get more points if it played stronger opponents then some other club. As clubs are already ranked it should be easy to calculate. Just add opponents coeffs and divide with the number of matches. Something like that. Just an idea. We'll see by the number of posts to this topic if people find your idea interesting.

For setting hyperlinks go to the top of this page and click at CODES.

Author: OhioUefaFan
Date: 07-09-2005, 07:18

Right. I didn't intend to put this up as a way of ranking domestic leagues...the league system is the perfect way to do it as is. I only brought it up to be used in reference to UEFA club competitions and coeffs.

We do it this way because gridiron is a fall sport...teams only play 10 regular season games, usually 7 league games and 3 non-league games. All 10 count to the rankings, and the best 8 in each region make the division's playoff. Prior to this system, there were no playoffs, and the "state champion" was chosen by a poll of sportswriters (an idiotic American trend that still exists, unfortunately, to this day).

Author: tomschman
Date: 07-09-2005, 16:09

I really enjoyed reading your post about applying the Ohio High School formula to EUFA seedings. Having lived in Ohio and being a big high school fan, I am well aware of the Ohio formula. The formula as well ith any formula has its drawbacks. I remember one year when a 4 - 6 team made the playoffs in the lowest division while several 10 - 0 teams in the same division were left out. Other states have different ranking systems that also have flaws. I would, however, be interested in seeing the results of applying this system to UEFA. Please contact ne at tschman822@aol.com and maybe we can kick around some ideas without boring everyone else on the forum. We may be able to run a few scenarios and then post the results.