Inside Selection Sunday: Mapnalysis '15

October 26, 2015

By Geoff Kimmerly
Second Half editor

Consider that one hour Sunday night was the product of nine weeks of games on top of two more weeks of practice, hours of lifting weights, conditioning and camps and a total of six months of work at the MHSAA office, in addition to hours upon hours put in by athletic directors scheduling years in advance.

The anticipation for the release of another year's MHSAA Football Playoff pairings is matched by a giant exhale only after the results are broadcast on Fox Sports Detroit, streamed online and posted on locker room doors and Facebook walls all over our state.

And then the fun begins again. 

But the Selection Sunday Show is the favorite hour of the season for many fans across our state. And that’s why, for the fifth year, we’ll explain our most difficult decisions in this Mapnalysis 2015 breakdown of how we paired 272 teams that will play next month for championships across nine divisions.

For those familiar with our playoff selection process, or who have read this report in the past and don’t want a refresher on how we do what we do, skip the next section and go directly to the “Observations & Answers: 2015.” For the rest, what follows is an explanation of how we selected the playoff pairings during the morning hours Sunday, followed by how we made some of the toughest decisions and a few thoughts on the breakdown of the field. 

Ground Rules

Our past: The MHSAA 11-player playoff structure – with 256 teams in eight divisions, and six wins equaling an automatic berth (or five wins for teams playing eight or fewer games) – debuted in 1999. An 8-player tournament was added in 2011, resulting in nine champions total each season.

The first playoffs were conducted in 1975 with four champions. Four more football classes were added in 1990 for a total of eight champions each fall. Through 1998, only 128 teams made the postseason, based on their playoff point averages within regions (four for each class) that were drawn before the beginning of the season. The drawing of Districts and Regionals after the end of the regular season did not begin until the most recent playoff expansion.

In early years of the current process, lines were drawn by hand. Dots representing qualifying schools were pasted on maps, one map for each division, and those maps were then covered by plastic sheets. Districts and Regionals literally were drawn with dry-erase markers.

Our present: After a late Saturday night tracking scores, we file in as the sun rises Sunday morning for a final round of gathering results we may still need (which can include making a few early a.m. calls to athletic directors). Re-checking and triple-checking of enrollments, what schools played in co-ops and opted to play as a higher class start a week in advance, and more numbers are crunched Sunday morning as the fields are set.

This season, there were a record-low 216 automatic qualifiers by win total – with the final 40 at-large then selected, by playoff-point average, one from each class in order (A, B, C, D) until the field was filled. There were only four Class D additional qualifiers with 5-4 or 4-4 (playing eight games) records from which we could choose – so after those four we added 12 teams each from Class A, Class B and Class C.

Those 256 11-player teams are then split into eight equal divisions based on enrollment, and their locations are marked on digital maps that are projected on wall-size screens and then discussed by nearly half of the MHSAA staff plus a representative from the Michigan High School Football Coaches Association. Only the locations themselves are marked (by yellow dots) – not records, playoff point averages or names of the schools or towns. In fact, mentions of those are strictly prohibited. Records and playoff points are not part of the criteria. Matchups, rivalries, previous playoff pairings, etc. also DO NOT come into play.

The same process is followed for organizing the 8-player bracket, with the difference that the 16 teams are selected purely on playoff-point average.

Geography rules: This long has been rule number one for drawing MHSAA brackets in any sport. Travel distance and ease DO come into play. Jumping on a major highway clearly is easier than driving across county-wide back roads, and that’s taken into consideration. Also, remember there’s only one Mackinac Bridge and hence only one way to cross between peninsulas – and boats are not considered a possible form of transportation. When opponents from both peninsulas will be in the same District, distance to the bridge is far more important than as the bird flies.

Tradition doesn’t reign: Every group of 32 dots is a new group – these 32 teams have not been placed in a bracket together before. How maps have been drawn in the past isn’t considered – it’s hard to say a division has been drawn in a certain way traditionally when this set of 32 teams is making up a division for the first time.

Observations & Answers: 2015  

Go north with ease: For the most part, most Regional pairings were easier to draw than in past years. For the most part. Divisions 2 and 4 were not easy, specifically when it came to deciding which teams would be paired with an Upper Peninsula qualifier or a team from the Traverse City area. Keep in mind, distance on these maps isn’t based on how the bird flies, but how a bus would drive. In Division 2, that meant putting Muskegon, Muskegon Mona Shores and Grand Rapids Forest Hills Northern – near U.S. 31 and U.S. 131  with Traverse City Central instead of sending Midland and Midland Dow west, up I-75 and then east-to-west again on more non-highway roads. In Division 4, we considered pairing Alma and Saginaw Swan Valley with Escanaba instead of Big Rapids and Remus Chippewa Hills. That decision came down to Big Rapids being about 14 miles closer to Escanaba than Swan Valley (Whitehall would go with Big Rapids and Chippewa Hills regardless.).

Four counties wide: Yes, on first glance it looks a little odd that Bay City John Glenn and Orchard Lake St. Mary’s would be in the same Division 3 District – but look at the map as a whole. Eight Division 3 teams are located south and mostly east of St. Mary’s, and John Glenn is the only team that could be considered in the northeastern quarter of the Lower Peninsula (Mount Pleasant and Gaylord are right down the middle and naturals to go with Sault Ste. Marie and Petoskey.). With the thumb playoff qualifiers of near-similar size in Division 4, there were no other options than to create this four-county trip between the Eaglets and Bobcats.

Semifinal selection: When all the Regional lines are drawn for a division, we also must figure out which Regionals will meet for Semifinals. This annually provides some challenges. Do we match east vs. west or north vs. south? If something seems iffy in four weeks when those games are played, again, keep in mind the entire map and entire Regionals that are matched up.

Scheduled strong: Because we’re looking only at dots on a map, we don’t see the matchups until everything is drawn – and in that way, we’re like everyone else. It wasn’t lost on us that two 9-0 teams will have road games this week or three 5-4 at-large qualifiers will be at home. For those 9-0 teams, it’s true: there’s nothing more they could’ve done on the field. But here’s why they will travel.

  • Clinton, in Division 6, hasn’t lost a regular-season game since 2011, and in fact plays in a strong Tri-County Conference made up of Class C schools and one Class D. Four of eight from the league made the playoffs, and the competition was so strong that Morenci and Petersburg-Summerfield will host games in Division 8. But Clinton’s opponent, Monroe St. Mary Catholic Central, while it did have one loss, fell to one-loss Class B Lansing Catholic – and the Falcons played eight Class B and one Class A school this season. Beating a Class B team is worth 16 points more than beating a Class C and 32 more than beating a Class D (with those points then divided over the number of games a team plays that season) – and all of that made the difference in St. Mary’s final eight-point playoff average edge over Clinton.
  • Saugatuck in Division 7 is a similar story. The Indians are one of three undefeated teams in their District, and did play a pair of Class B opponents. But by no fault of their own, especially in league play where the schedule is set, Saugatuck didn’t face a team this season that won more than five games. Hesperia played three Class B opponents including one that finished 6-3 – enough for the slim 1.8 points more in playoff average that earned the Panthers homefield advantage. Pewamo-Westphalia has the highest average in the District and didn’t play a Class B – the Pirates actually played three Class D teams. But they also beat four teams that made the playoffs including two that finished 8-1 – giving them a 2.7-point edge on Hesperia and 4.5 edge on Saugatuck.
  • The 5-4 teams that will host – Redford Thurston in Division 3 and Escanaba and Benton Harbor in Division 4, all earned their spots. Escanaba faced six playoff teams and beat three, Class B Benton Harbor played eight Class A teams, and Thurston played five playoff teams and a sixth that just missed an at-large bid.  

At the end of the day ... 

What you see is what our committee decided upon after multiple discussions among multiple groups that broke down every sensible possibility we could muster. There are certainly points open to argument – and we likely made those arguments as well.

Those who would like to see the playoff selection process changed are in favor of a larger strength-of-schedule component, and it’s interesting to see how strength of schedule inadvertently made a larger impact this season than in the recent past – especially given the examples above of undefeated teams going on the road and at-large teams hosting.

Why were there fewer automatic qualifiers than ever before? Here's one theory. There were 11 fewer teams in 11-player football this season than in 2014 (most moved to 8-player). An argument can be made that there were fewer wins to be gained against teams that last season might have struggled to field 11-player teams, shifting the balance to fewer automatic qualifiers and more parity with stronger teams facing each other to fill their schedules.

Meanwhile, the 8-player field grew by nine teams this fall and has its strongest ever, with seven teams that finished 5-4 missing the postseason after a team with a sub-.500 record got in just a year ago.

Given how some matchups shook out this fall, the next argument by those seeking change likely will center on seeding entire Regionals instead of just Districts. But keep this in mind as well: if Regionals were seeded with this year's groupings, it would create possibilities of first-week trips like Battle Creek to Traverse City and Cedar Springs to Sault Ste. Marie. We're fairly sure most coaches and players would dread such journeys for a first-round game. 

It's a lot to digest, and the scrutinizing will surely continue long after these playoffs are done as we all work to conduct the best tournament possible. 

But at the end of the day – and the end of these next five weeks – to be the best, teams will need to beat the best no matter the matchups. And we’ve got plenty to look forward to starting this weekend and all the way through the 11-player Division 3 Final on Nov. 28 at Ford Field.

The MHSAA Football Playoffs are sponsored by the Michigan National Guard.

PHOTO: The Division 4 map was among the most difficult to draw during this year's selection process.

Inside Selection Sunday: Mapnalysis '13

October 28, 2013

By Geoff Kimmerly
Second Half editor

At the end of the day – Sunday, in this case – the 2013 MHSAA football playoff field was determined based on a set of numbers we began working with long before the first kickoff of this season.

So to kick off our discussion of how some of the 2013 playoff-selection decisions were made, here are a few numbers that might boggle the mind – or at least surprise:

  • A total of 3,111 high school results were used in determining this season’s field – 2,978 for 11-player and 133 for 8-player games.


  • We inputted and then followed the schedules for 623 MHSAA teams.


  • We also inputted and followed weekly the schedules for 50 teams from surrounding states and Ontario that played at least one game against one of our MHSAA schools.


  • We worked through complicated maneuverings made necessary by seven teams playing a mix of 11 and 8-player games, plus two more teams that played a mix of varsity and junior varsity opponents.


  • And by Sunday morning we ended up with a few more numeric rarities: only 225 automatic qualifiers, the fewest since the current playoff system was introduced in 1999, and also an uneven number of at-large bids from our four classes because only six Class D teams reached the number of victories needed to be considered. (This was balanced by taking more at-large qualifiers from Classes A, B and C.)

And that was just the start of one of our most exciting days of the school year.  

Following are more details. First, I explain some of the history of the MHSAA playoffs – I’ve lifted this in part from our 2012 report, so skip ahead if you’ve already got that down. Next, I touch on five themes that emerged as we built the brackets for this season’s tournament.

The process

Our past: The MHSAA playoff structure – with 256 teams in eight divisions, and six wins equaling an automatic berth (or five wins for teams playing eight or fewer games) – debuted in 1999, as mentioned above. An 8-player tournament was added in 2011, resulting in nine champions total when November is done.

That’s a long way from our start. The first playoffs were conducted in 1975 with four champions. Four more football classes were added in 1990 for a total of eight champions each fall. Through 1998, only 128 teams made the postseason, based on their playoff point averages within regions (four for each class) that were drawn before the beginning of the season. The drawing of Districts and Regions after the end of the regular season did not begin until the most recent playoff expansion.

In early years of the current process (or until the middle of the last decade), lines were drawn by hand. Dots representing qualifying schools were pasted on maps, one map for each division, and those maps were then covered by plastic sheets. Districts and Regionals literally were drawn with dry-erase markers.

Our present: After a late Saturday night tracking scores, we file in as the sun rises Sunday morning for a final round of gathering results we may still need (which can include making a few early a.m. calls to athletic directors). Then comes re-checking and triple-checking of enrollments, what schools played in co-ops, some records and more before the numbers are crunched and the fields are set.

Those 256 11-player teams are then split into eight equal divisions based on enrollment, and their locations are marked on digital maps that are projected on wall-size screens and then discussed by nearly half of the MHSAA staff plus a representative from the Michigan High School Football Coaches Association. Only the locations themselves are marked (by yellow dots) – not records, playoff point averages or names of the schools or towns. In fact, mentions of those are strictly prohibited. Records and playoff points are not part of the criteria. Matchups, rivalries, previous playoff pairings, etc. also DO NOT come into play. The same process is followed for organizing the 8-player bracket.

Observations and answers: 2013

This doesn’t happen overnight: Preparation for selecting the MHSAA playoff field begins long before the first kickoff of fall, much less the first practice. We load schedules for all 600-plus varsity teams during the summer, and many schedules remain fluid right up until the first Friday of the season – and this fall, a few weren’t settled until Week 2 or 3.

This summer as in some past we also worked through schools closing (Inkster, Saginaw Buena Vista, Detroit Northwestern, Flint Northern), and others deciding in mid-July and early August they would not field teams because of a lack of players.

Sometimes we have to take odd paths to find scores for these games. The last 11-player score to be added to our data this regular season came in as a result of tweeting the sports anchor of a Wheeling, W.Va., television station. Our last 8-player score came in via email from a Wisconsin athletic director at 10:30 Saturday night. Thankfully, we get plenty of assistance from some of our friends in the field, who keep an eye on the data and alert us when something appears missing or incorrect.

Win and advance: This season’s list of 5-4 teams includes a number of heavy hitters that did not receive at-large bids – East Grand Rapids, Utica Eisenhower, Orchard Lake St. Mary’s and Flint Powers Catholic to name a few. All were solid teams and played strong competition. All missing the playoffs likely raised some eyebrows.

But we have to take a look at this from a statewide view. There admittedly can be some argument about what schools qualified for the 226-256 spots in the field – but the important part is that 225 qualified because they all met the minimum win requirement. A playoff is simply that – it decides a champion based on teams winning. For some it’s harder to pile wins, of course, because they play in tough leagues. But the winners of those leagues are in the field – and surely will credit that tough road with getting them prepared to now play the state’s best.

Geography rules: This long has been rule number one for drawing MHSAA brackets in any sport, and is a repeat as well for those who have read this report the last two Octobers. Travel distance and ease DO come into play. Jumping on a major highway clearly is easier than driving across county-wide back roads, and that’s taken into consideration.

Also, remember there’s only one Mackinac Bridge and hence only one way to cross between peninsulas – and boats are not considered a possible form of transportation. When opponents from both peninsulas will be in the same District, distance to the bridge is far more important than as the crow flies.

The best example of this comes this season in Division 5. Grayling clearly is east of both Kingsley and Kalkaska – but also sits on I-75, while those two do not. So while those more western teams are geographically closer to Houghton, Menominee and Kingsford from the Upper Peninsula, we instead paired the three U.P. teams with Grayling because being on a main highway made for a shorter trip. The trip to Grayling for any of those U.P. teams would be 36 miles shorter to Grayling than Kingsley and 13 miles shorter to Grayling than Kalkaska.

Sometimes it’s where the points aren’t: Sure, it would be best-case scenario to have perfect sets of eight dots split into four quadrants from Calumet to Bedford. But generally that doesn’t occur. “Dots determine the map” is a common phrase heard here during this selection process, but that works the other way as well. If there are no qualifiers in a division from a specific area of the state – see Division 1, with none south of Holland or west of the greater Lansing area – there’s no choice but to create the unusual Regional Final possibility of Traverse City West vs. Brighton. Brighton is simply closer to the west side of the state than our other options.

Border to border vs. coast to coast: Should Regions be grouped north to south or east to west? There isn't a right or wrong answer – it just depends on that set of dots.

Whenever we have Upper Peninsula teams in a division, they’ll be grouped with those from the northernmost points of the Lower Peninsula for a District. The next northernmost schools will be grouped into a District, and together those eight will form a Region.

But the tough decision comes with the other six Districts. Look at this season’s Division 5 map: Six Districts are grouped south of U.S. 10 with three near or west of U.S. 127 and three east of that highway, which runs through the center of the Lower Peninsula. We grouped the two southwestern Districts into a Region and the two southeastern Districts into a Region – leaving a final Region that stretches from Muskegon on Lake Michigan to Almont, about 35 miles west of Lake Huron.

That’s a haul. But it’s also the best of our possible compromises. We could’ve instead paired regions that would’ve stretched from Hopkins to Monroe – only 19 fewer miles in distance than Muskegon Oakridge to Almont, but a scenario that could’ve created travel increases for a number of additional teams. Another option included a possible trip from Detroit University Prep to Freeland, which also would take more than two hours.  

Bottom line – it’s been written here before – we pour all we have into this process, asking questions often more than once until we come up with a consensus. We do appreciate the arguments that arise once brackets are released to the public: The discussions are proof of how much players, coaches and fans care – and often show us new ways we can look at a system that’s now 15 years old.

But we must remember that the good news is the tournament is still set up to reward nine champions over the next five weeks, and five schools – Auburn Hills Oakland Christian, Coldwater, Detroit Allen, Eaton Rapids and Muskegon Mona Shores – will be competing for those titles for the first time.

It’s not so much how the tournament starts as how it ends. And we’re preparing for nine more memorable conclusions.

PHOTO: Each collection of grouped dots is a District on this season's Division 3 playoff map.