It was a classic case of making sure you read beyond the headlines.
I was loosely monitoring the Massachusetts Interscholastic Athletic Association's Tournament Management Committee meeting a few weeks ago when I caught this statement: "Using margin of victory in some form is the most accurate way to seed a statewide tournament."
Nonsense, I quickly concluded, firing off a few Tweets on the matter. What of the Northeastern Conference champion Danvers High football team that went 7-0 in 2019 but memorably secured three straight wins by only one point? A comparably lousy Peabody squad got its two wins by a combined 86 points that season. Surely margin of victory doesn't tell us the better team between those two.
Come to find out, the new ratings formula being discussed at this meeting doesn't rely solely on margin of victory. With a helpful nudge from retired St. John's Prep athletic director (and guru on all things MIAA) Jim O'Leary, I took a much deeper look at the potential metric.
The truth is it's a really good formula.
For as long as most anyone remember, high school state tournaments in Massachusetts have been seeded by straight winning percentage. Though there are some exceptions (boys and girls lacrosse, football), the vast majority of teams are assigned their playoff matchups by their regular season record and nothing else.
Especially in the smaller divisions, this always leads to false upsets. How many 17-3 baseball teams that earn a high seed are bounced by some 10-10 outfit from a battle-tested conference like the Merrimack Valley or Dual County? On paper, a 12th seed toppling a No. 5 reads like March Madness, but when you follow the Eastern Mass. sports scene you know it's often no upset at all.
It's also unfair to good teams. A league champion with a great record that winds up seeded No. 4 might be better off at No. 5 or No. 6 if the aforementioned really good team from a tough league is sitting there at No. 13.
There aren't many corners of the Massachusetts high school sports community that would argue power seeding isn't the way to go. A power formula should compensate for strength of schedule and create a better path for the very best teams to meet each other later in the postseason.
The best way to calculate a power rating? That's a bit more complicated.
Football, for instance, uses its own unique formula that considers wins and then awards points for each of your opponents' wins; the bottom line on the football rating is that the more a team that you defeat wins its other games, the better your rating will be. I like that formula a lot; there haven't been many complaints about football seeding since the true playoffs began in 2013.
There are plenty of other options. College basketball sometimes relies on RPI, a formula that factors winning percentage with opponents' winning percentage. College hockey fans know all about complicated KRACH and Pairwise systems. Long before Bill James' math took over baseball, statistician Jeff Sagarin was devising formulas to rate college football.
Massachusetts decided to use the ratings calculated by MaxPreps as its standard for all sports when it decided to implement power seeding last January. While MaxPreps is far better than straight winning percentage, there were still concerns: the formula doesn't account for ties (harming soccer and hockey) and its proprietary, secretive nature would potentially leave coaches in the dark as to what they could do to earn a higher seed.
That brings us up to speed with the tournament committee meeting of last month. The MIAA's tournament folks were exploring a new, in-house formula that could potentially replace MaxPreps. By leaning on margin of victory, the formula can account for ties by setting that margin at zero. It can also set a known margin cap, which would dissuade teams from running up the score.
The first reaction many people have to using margin of victory is that rewarding huge scores in high school sports is wrong. The margins set on the MIAA's sample brackets, though, are very reasonable. Basketball is capped at 10 points, softball and baseball at 10 runs and lacrosse at 10 goals. Hockey, field hockey and soccer are capped at three goals.
That means whether you win by 10 or by 25, you get the same value; there's no bonus for winning a soccer match 9-1 rather than 4-1.
It's very much in line with the MIAA's stated goal of having a formula that is transparent, simple and accurate. I perused the sample brackets the MIAA posted on their website and they look pretty solid.
Absent the global pandemic, the MIAA would've begun using MaxPreps to seed its traditional tournaments this past fall. We were slated to begin the statewide tournament in the fall of 2021, and certainly any power formula is going to make comparing Hamilton-Wenham to Northbridge a lot more manageable.
It would've been nice to see how the MaxPreps' seeding played out this year. To really evaluate and critique its formula, I'd like to have seen it in action. Still, you've got to commend the MIAA for continuing to tweak things and work to address its members concerns and make the playoff experience better.
The new formula is still in its infant stages. No one knows when it'll possible for the MIAA to hold state tournaments again; it could be this April, this June, not until November or even later than that. These days, most athletes would agree to be seeded by winning percentage, MaxPreps, a new formula or by drawing straws as long as they could play.
The tournament folks at the MIAA have a more forward looking vision. They're not sure how the next state tournament will be seeded, but they're still putting in work to make sure its the best and most fair formula for the Bay State's student-athletes.
In any formula, that's an endeavor that grades out to 100.