How to Make a Conference Schedule With No Divisions

By: Mike Davis

In the last couple of weeks, the ACC has been very vocal in their desire to scrap the current system of conference divisions. Since the SEC, ACC, and Big Ten all expanded to 14 team divisions, many have proposed a solution:

“Why not set up the schedules so that every conference member a team doesn’t play in their current year, they must play them the next year?”

I have seen something along those lines posted countless times on various forums for the better part of two years now and the idea is rapidly gaining steam. Syracuse Athletic Director Daryl Gross proposed the idea with no fixed rivals to the ACC earlier in the year.  The website Southern Pigskin proposed the idea for the ACC/SEC in late January.

This is much easier said than done. Making this proposal a reality is a lot harder than it seems. There are a number of factors that need to be taken into account such as:

Keeping home/away consistent between programs - For example, making sure a matchup like Ohio State and Penn State continue to rotate evenly between Columbus and Happy Valley.

Keeping the amount of conference home games consistent - For eight game conference schedules, each team should have an equal amount of home/away. For a nine game conference schedule, some teams will have five home games whereas others will have only four. To keep the balance of a nine game conference schedule equal, all schools should be guaranteed nine conference home games every two years.

Keeping rivalry games – This should be pretty self explanatory. Matchups like Ohio State-Michigan or Auburn-Alabama not playing every season would cause quite a bit of backlash.

So far, I have yet to see a scheduling proposal that takes into account all three of these problems. This article is where I do just that, showing exactly how a “no divisions” scheduling scenario can play out.

Some Things to Note:

I will provide full schedules for both eight game and nine game schedules for the Big Ten, SEC, and ACC. I did not include the Big 12 because they already play a true round robin schedule.

I did not do a Pac-12 scenario because that conference has other scheduling factors to consider not seen in the other Power-Five conferences. For example, every Pac-12 member currently gets to play three California schools each year. This gives each non California member an equal “California presence” and allows all four California schools to play each other every year. The new members (Utah/Colorado) are placed in a division where most of their out of state alumni living in the Pac-12 footprint are located. This goes a long way to help them maximize the benefit of new Pac-12 affiliation in terms of recruitment, donations, and attracting prospective students. For the Pac-12, eliminating the divisions causes more problems than it solves.

A 14-team conference is quite literally the hardest conference size to schedule.  You can not do any sort of pod setup with this number. I also struggled to find a way to make a multi year round robin format because there were not enough teams. This project has taken me months and countless failed Excel sheets. I tried a number of different schedule styles and the only real option was an 8 game conference schedule with 3 fixed rivals.

Step #1 Assign Rivals

SP #20

Click on any picture in this article for a larger view.

Due to the size of these pictures, I could only include them in trimmed down sizes. Click on any of them for a full view. These are the three rivals I assigned per each team. If you want further information about why I chose these rivals, click here.

Step #2 Making the Schedule Sheet

SP #1

Click on any picture in this article for a larger view.

This picture is a basic rundown on how the schedule sheet works. On the far left, you have all the teams in bold. Their 13 fellow Big Ten teams all appear as you move to the right horizontally. The first team (Michigan) has the eight teams they play in 2015 in red and the five teams they don’t play in yellow.

I tried to make the orange/blue pods as fair as possible. I accomplished this by splitting the four strongest Big Ten programs (UM, PSU, OSU, & UNL*) and did my best to make other programs play only two of them each year rather than four of them one year and none of them the next year. I also split up schools from the same state. For example I tried to prevent schools from playing IU/Purdue or UM/MSU in the same year. This was particularly important with the ACC where there are huge recruiting differences between the northern and southern states.

I did things like this whenever possible for every conference when I made their schedule sheets.

*I refer to Nebraska as UNL as explained in the Explaining the Rivalries post.

Step #3 Determining a Rotation

SP #2

Click on any picture in this article for a larger view.

This picture is a breakdown of how the teams rotate. The yellow represents each program’s fixed rivals. In other words, the schools they play every year. The orange and blue represent the two pods that rotate back and forth. Schools will play either all teams in the blue pod or all  teams in the orange pod plus their fixed rivals. While I call them pods, they do not reflect any pod/divisional split in the conference.

SP #3

Click on any picture in this article for a larger view.

This is what the following year (2016) will look like.  You will notice the only difference in this picture from the last picture is the year. This is because the pattern of the blue/orange rotation goes blue, blue, orange, orange, and repeats. This means that a team plays the exact same schedule two years in a row.

SP #4

Click on any picture in this article for a larger view.

As you can see, the following year (2017) the teams will play the orange pod.

SP #5

Click on any picture in this article for a larger view.

And the year after (2018).

The reason why I set it up like this was to balance home/away between programs.  The reasoning behind using this tactic is quite complicated and hard to explain. The best way to state it is that if you don’t use it, you will have one of the following happen:

Using PSU & Michigan as an example.

- Penn State has to play at the Big House in Michigan in both 2017 and 2018

or

- Penn State plays at the Big House in Michigan in 2017, and Michigan visits Happy Valley in 2018. However, PSU gets only three Big Ten home games and five away games for 2018.

Neither situation is ideal and for a schedule to work, you cannot have either happen. The reason why this happens is that if you did this on an orange, blue, orange, blue rotation, then you have some teams who will fulfill their home/away requirements with some opponents in two years other in three years. That is where things start to get screwed up and causes scheduling imbalances. The fix is to have every program fulfill their home/away requirements with every opponent at the same time. So having a “repeat year” allows for the exact same schedule to occur, only the home/aways are the opposite of the previous year thus eliminating the problem.

This concept is not new. The Big Ten used a method similar to this for the exact same reason back when they were an 11 team conference.

Step #4 Determining Home/Away

Now that the rotation has already been determined, the next step is to decide home/away. I accomplished this by filling in the home matchups dark green and the away matchups in light green.

SP #6

Click on any picture in this article for a larger view.

As you can see every team has four home games and four away games.

SP #7

Click on any picture in this article for a larger view.

In this picture, you will observe that the only differences from the last picture (besides the year) is that all the dark greens have been turned to light greens and vice versa.

The above examples show a completed Big Ten schedule for the 2015 and 2016 seasons would look like. I have provided a 2015 below for the SEC and ACC.

SEC 2015 fix

Click on any picture in this article for a larger view.

SEC

SP #9

Click on any picture in this article for a larger view.

ACC

One important thing to note that this is simply a spreadsheet that shows who each team plays and the location of the game. This does not mean these teams must play in this specific order. They can move games around however they please.

For a full Excel spreadsheet of an eight game schedule on Google Docs, all three conferences are provided below:

Big Ten

SEC

ACC

One thing you will notice is every team has exactly four home games and four away games each.  You will also notice that it takes only four years before the cycle repeats.

Step #5 Adding 9th Conference Game

The Big 12 and Pac-12 both play nine game conference schedules. The Big Ten also uses a nine game conference schedule and the ACC is making a serious attempt to do the same. So of course I am not going to present to you with only an eight game schedule.

The way I set up a nine game schedule was to keep the eight game schedule plan that I have already shown and add a 9th game. It is basically an 8+1 model using all the rules of the eight game model with an added 9th game. The team that a program plays as part of their extra game is taken from the “Does Not Play” column.

SP #10

Click on any picture in this article for a larger view.

So this is what the schedule looks like so far. As you can see, it looks just like an eight game schedule with an added 9th game. You can see that the orange/blue pod setup is the same and I have placed the 9th game (I call it the slot game) under game #1 in the far left in bright green.  If the Color concept is confusing anyone, sorry for having so many different shades of green.

SP #11

Click on any picture in this article for a larger view.

Here I have filled in the empty game #1 using the first column of the “Does Not Play” section. I will use this column for the next four years before moving on to the 2nd column of the “Does Not Play” list. I have done this because it takes the other eight games four years to fully repeat. This means it takes a nine game conference schedule 24 years to go through a complete cycle before repeating.

SP #12

Click on any picture in this article for a larger view.

Example of what the schedule will look like in after 5 years (2019) when the 2nd cycle starts. I used a shade of blue to identify the “slot game” on the right under “Does Not Play.”

SP #13

Click on any picture in this article for a larger view.

Example of what the schedule will look like in after 17 years (2031) when the 5th and final cycle starts.

SP #14

Click on any picture in this article for a larger view.

Example of what the schedule will look like after 21 years (2035) when the cycle fully repeats and goes back to the 1st cycle. The schedule is exactly the same as 2015.

Step #6 Adding Home/Away for a 9th Conference Game

SP #15

Click on any picture in this article for a larger view.

This is how I assigned the home/away. Some teams have five Big Ten home games while others have only four home games.

SP #16

Click on any picture in this article for a larger view.

As you can see the home/aways have been switched, meaning all the teams that had only four conference home games last year have five conference home games this year and vice versa.

The problem is that it is not this simple every year. In some years, you will have teams play consecutive seasons of having 4 conference home games. It is an unavoidable problem.

For the Big Ten and SEC, I could have set it up where it happens only once every 20 years for 7 of the members in each conference. However, in the interest of fairness, I didn’t want to do that so I attempted to make it happen only once for every school every 20 years. That was too problematic so I took the easy way made it happen 2x times for every school every 20 years. It might be possible to make it happen just once a year per school, however these schedules are incredibly time consuming and I did not want to spend the time to redo those schedules with the hope of accomplishing that.

With the ACC, I was successful in making the consecutive four home games a year event happen just once every 20 years. Part of this was because I configured the ACC schedule last and learned from my mistakes with the Big Ten/SEC schedules. This allowed me to build the ACC in a way that was much easier to work with. Another reason was because the ACC fixed rivalries had a different setup using a semi pod system that also made working with the schedule easier. If you are confused about what I mean by the ACC rivalry setup, I explained it in my “Explaining the Rivalries” post.

SP #17

Click on any picture in this article for a larger view.

This is a graph I designed to display which years Big Ten teams have 4 or 5 conference home games. The graph shows how often the consecutive 4 conference home games event occurs for each Big Ten member and is represented in yellow. The same graph also shows that every 20 years (not counting 2015), each member has the same amount (10) of white and orange boxes. This means that the number of home/away games each team has is balanced throughout the full cycle. The graph can also be used as an example to show in which years the 5-cycles run through.

SP #18

Click on any picture in this article for a larger view.

Here is the same graph for the SEC.

ACC fix

Click on any picture in this article for a larger view.

As well as the ACC.

I have provided longer versions of this graph all the way to the year 2054:

Big Ten

SEC

ACC

And of course, who would I be if I didn’t provide full excel sheets for 9 conference games? Each spreadsheet shows a full cycle (20 years).

Big Ten

SEC

ACC

How do you determine who plays in the conference championship game?

The two teams with the highest BCS ranking will be paired up. Since OSU-UM will be paired up in the final week before the conference championship game, one of them has to lose and thus drop in the BCS rankings. This  minimizes the potential for back-to-back rematch games between any two opponents.

Concluding Thoughts:

At first, I was strongly against giving each team three fixed rivals. But I had to do it because doing it made the eight game schedule work so smoothly. After awhile, the concept grew on me as several schools have three major conference rivals and it meant more rivalries could be saved.

While this scheduling style is a major upgrade over the current divisions setup, some of the problems the divisions setup creates are still present in this system. Teams still miss other teams for two straight years, which allows the potential for “conference dodging” between the strongest teams to exist.

Another reason I was against the divisions setup is because teams go too many years without playing at certain locations. An example of this is Syracuse not playing Georgia Tech in Atlanta until 2027. This system prevents teams from skipping games at another team’s stadium for more than four years, which is still quite a long time.

An eight or nine game schedule without divisions is entirely possible as I have shown and can be implemented without causing any major problems. Now if only Delany, Slive, or Swofford read this and decide to use it.

Follow me on Twitter: @sportspolitico or email me: sportspolitico@gmail.com

18 responses to “How to Make a Conference Schedule With No Divisions

  1. I have done these 6 ways to Sunday.. I think 3 rivals is key and a decent number. I might rather 4 but whatever. The one thing missing is that they realtive streangth should be somewhat even. Look at FSU and Louisville. You have FSU Playing GT, Miami, and Clemson every year while Louisville plays Cuse, BC and Pitt. That is way out of line.

    • What makes 3 rivals key is that it allows for a 3-10 setup. You can then split 10 evenly for 2 groups of 5 that can be played in rotating years. If you do 2 or 4 rivals then you get a 2-11 or 4-9 and splitting them becomes a problem.

  2. BTW, very well done. I would change one thing though. While there might not be divisions the first factor for making a conference champ game has to be conference record. After that you can choose BCS (or whatever it will be called) ranking. Clemson is playing USCe and UGA OOC this year while NCSt plays no one. Clemson could have a better ACC record but be ranked lower then another team if they lose those two games.. Start with ACC record then ranking.

    • Thanks. You are right and that is a very good idea. I was on the fence about using conf. records to determine a CCG but ultimately decided against it for two reasons.

      1. The reason for this is because the fixed rivals cause a SOS imbalance. With BCS you have a lil bit of an SOS input but not much. And I wanted the CCG to have a the very least an SOS aspect to it. It is rare for the top 2 in conference standings to not be the top 2 in the BCS in the week before the CCG. While this is a scenario that could happen, I don’t expect it to happen on a frequent basis.

      2. The BCS reflects not just a conference schedule but an entire season. Because we now have a playoff and wining a CCG could be turned into an automatic bid one day I did not want a situation where a team loses every ooc game, makes a run in conference play and gets into the playoffs with three losses. The BCS allows Clemson to get a SOS advantage so if they had the exact same record as NC State they would most likely pass NC State.

  3. “The two teams with the highest BCS ranking will be paired up.”

    There isn’t going to be a BCS ranking anymore.

      • It “can,” but the issue is will it? The only reason the BCS rankings existed were to determine the National Championship game. Now that it doesn’t, is the ranking going to be maintained? By whom?

        Putting that aside, what’s the purpose of using the BCS rankings to determine your conference championship game, when the BCS ranking itself has no value?

      • The BCS is operates like any private business. There are plenty of people who have a personal interest in keeping it around. The BCS ranking came from individuals who developed their own mathematical formulas and then combined them with the polls to determine a teams BCS. It i something that is simply too valuable to just role over and die now that a playoff is here.

        The most likely scenario is the BCS takes on the role similar to that of RPI in the NCAA tournament. The CFB playoffs will have a selection committee and they very well could rely on the BCS to help them chose a team.

        The reason why I say BCS is because it was the standard for the past 15 years. If you try to come up with a new ranking system from scratch it will most likely lead to some sort of bickering among the schools about how much weight to put into SOS, OOC, etc.

  4. The Playoff Committee has already announced that they aren’t going to use a ranking system to determine the selections. Therefore, it’s pointless to pin your conference’s champion to a ranking that has absolutely no bearing on the success of your conference’s teams. What is the logic in determining your conference champion by an obsolete formula?

    • There is a big difference between the way the BCS used rankings to determine a NCG vs. the way the NCAA basketball committee uses rankings to determine seeds. The CFB playoffs has been clear that they won’t go down the “only top four get in” route, but that does not mean that they will use ranking systems to judge which teams get a bid in the same manner RPI is used for basketball.

      While the BCS was controversial I see it as one of several valuable tools out there the the playoff committee has at their disposal.

      • Well there is your first problem. It doesn’t matter if you personally like the BCS. The committee has already said it’s not going to have some sort of formula or ranking.

        I’d like to address something you said in your previous answer. You mentioned that the BCS has been the standard for the past 15 years. Well, the BCS was only the standard because it determined the national championship game. It wasn’t the standard “just because.” Being ranked #1 or #2 meant you went to the title game automatically, so it was important where you were ranked. That’s not the case anymore. It doesn’t matter where you are ranked in the BCS or whatever ranking, because those rankings don’t determine who gets into the title game anymore. The BCS rankings ceased to have any function once the playoffs were installed.

      • The reason why I wrote this article is because I was tired of other bloggers simply posting this idea then assuming it will work perfectly without getting into all the details and working out the problems that go along with it. To determine the CCG bids I didn’t want to say “oh they could just use a mathematical formula” because that would make the article incomplete and hypocritical of myself. I wanted to give a blueprint of how to do this from start to finish and making a mathematical formula wasn’t something I wanted to do.

        When I say the BCS has been the standard for the past 15 years I am referring to beyond its 1 vs 2 selection matchup. The BCS had contract stipulations such as:
        -If Notre Dame achieves a top 8 ranking they get an automatic BCS berth.
        -If a non AQ school receives a top 16 ranking they get an automatic BCS berth.

        Using the BCS to determine game bids is not something new. I used the BCS as an example because it is so widely known and understood. Personally, determining the CCG by BCS is not a required condition for this to be a slam dunk idea for the conference’s to use. If they want to use the AP…..fine. If they want to make their own formula….fine, If they want to go based off conference record….fine.

  5. “A 14-team conference is quite literally the hardest conference size to schedule. You can not do any sort of pod setup with this number.”

    Sure you can. In fact your ACC setup should translate nicely, as follows:
    Pod A: Virginia Tech, NC State, Wake Forest
    Pod B: Virginia, North Carolina, Duke
    Pod C: Louisville, Syracuse
    Pod D: Boston College, Pitt
    Pod E: Clemson, Florida State
    Pod F: Georgia Tech, Miami

    Pod rotation is as follows:
    Year 1
    Atlantic Division: Pods A, C, and E
    Coastal Division: Pods B, D, and F
    Year 2
    Atlantic: A, D, E
    Coastal: B, C, F
    Year 3
    Atlantic: A, D, F
    Coastal: B, C, E
    Year 4
    Atlantic: A, C, F
    Coastal: B, D, E

    Play every school in your division each year, plus two in the opposite pod (A is opposite B, C opposite D, E opposite F). If there are only two schools in the opposite pod, then you will play both every year. If there are three, then play your rival every year and each of the others twice in four years (once home, once away). A nice eight game conference schedule.

    Over four years each school will have played three other schools every year, and each of the other ten schools twice (once home and once away), which solves the original problem without having to eliminate divisions. Should be the exact same games as your solution, though not necessarily in the same years.

    You could use this approach for the Big Ten and SEC also, but you would need to change your rivals.

    And of course, division champions play in the conference championship game, so you don’t open another can of worms.

    This pod rotation should also work for 16 schools with a nine game conference schedule, 18 schools/10 games, 20 schools/11 games, and even 12 schools/7 games.

  6. I lost all respect for everything in this article when he listed duke and unc as rivals #3 with each other.

    • The #3 slot is reserved for the biggest rivalry game. It does not mean they are the third most important rival, rather the 3rd fixed rivalry game played in the season.

  7. I realize this is from several months ago but it’s a great article. Explains a pretty difficult subject as cleanly and quickly as possible — plus it’s actually a viable solution for a problem faced by the super-sized leagues. Would be interested to see how C-USA could utilize this format.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s