Way Too Early MAAC RPM Projections

Rosters are starting to settle (mostly) down and summer league is starting up. Now seems like a great time to start looking at what next season could look like in the MAAC.

To do that I’m generating¬†projections using a statistic called returning-possession minutes. The formula for RPM is simple. You just add up the percentage of minutes played times the usage for each player. By combining playing time and on-court usage you get a real sense of how much production a team returns.

I’ve then taken that information and built a rather naive linear regression projection model that takes into account a team’s performance the past two seasons and how much talent it has coming back. The RPM part of the calculation nearly doubles the effectiveness of the overall model (which is a big reason I’m using it),¬†but it’s certainly not perfect.

Let’s first take a look at how a similar model did last season.

Projected Finish. Team – Actual Finish

  1. Siena – 8
  2. Iona – 1
  3. Manhattan – 3
  4. Saint Peter’s – 7
  5. Monmouth – 4
  6. Quinnipiac – 6
  7. Marist – 11
  8. Fairfield – 10
  9. Niagara – 9
  10. Rider – 2
  11. Canisius – 5

Whoops! The model had some rough times last season. It missed big on Siena and Rider in particular. In the case of the Saints, RPM bought into a team with a lot of talent returning that was expected to be near the top of the MAAC race, but Brett Bisping played six games and the defense massively regressed. In the case of Rider the projections had no idea that Matt Lopez and Teddy Okereafor were coming. Both of those projections go to show some of the flaws of RPM. Key players will get hurt and schematic changes can reveal big flaws. The addition of a transfer or two can make a huge difference as well. Still, we’re going to give this another shot. First, here is how much each team has returning according to RPM.

  1. Fairfield – 78%
  2. Iona – 72%
  3. Monmouth – 67%
  4. Rider – 63%
  5. Niagara – 59%
  6. Siena – 52%
  7. Marist – 52%
  8. Canisius – 49%
  9. Manhattan – 44%
  10. Saint Peter’s – 42%
  11. Quinnipiac – 25%
What will Quinnipiac do without Ousmane Drame next season? (Photo Credit - Matt Eisenberg)
What will Quinnipiac do without Ousmane Drame next season? (Photo Credit – Matt Eisenberg)

The first thing that pops off the page is just how little Tom Moore has returning for his third season in the MAAC. The two big losses are Zaid Hearst and Ousmane Drame, but the Bobcats also lost Evan Conti and Justin Harris. Moore will basically be building from scratch this season. The other thing that is quite impressive, is just how much regular season league champion Iona returns. The Gaels do lose David Laury, but the remainder of the core that went 17-3 in league play is back, including A.J. English. That’s why Tim Cluess’ team is the one to beat yet again.

Projected Standings Using RPM (Team – Projected KenPom Pythag):

  1. Iona – 0.6116
  2. Rider – 0.4745
  3. Monmouth – 0.4484
  4. Canisius – 0.4454
  5. Manhattan – 0.4454
  6. Fairfield – 0.4138
  7. Siena – 0.3496
  8. Saint Peter’s – 0.3418
  9. Niagara – 0.3307
  10. Marist – 0.3006
  11. Quinnipiac – 0.3002

These projections would make Iona about a Top 125 team according to KenPom. Really though, the MAAC looks like three tiers: 1) Iona, 2) A group of potential contenders, 3) Teams that are still in or entering a rebuilding cycle. A team like Siena could be helped out by getting some players back from injury. Fairfield is adding an experienced player in former Stony Brook forward Scott King, which could help. Aaron Rountree is transferring to Iona, which promises to help the Gaels up front. So these projections certainly won’t be perfect. But they’re a good start as we await the beginning of college basketball season in a little less than five months.

4 thoughts on “Way Too Early MAAC RPM Projections

  1. How the heck is Niagara 5th? They are a disaster.

    My early MAAC predictions –

    1. Iona
    2. Monmouth
    3. Siena
    4. Rider
    5. Manhattan
    6. Canisius
    7. Marist
    8. St. Peter’s
    9. Fairfield
    10. Quinnipiac
    11. Niagara

    Like

    1. Niagara is 5th in RPM, but much lower in the actual projection if you go a little bit further down the post. A good example of why RPM doesn’t work in a vacuum and needs additional context to be particularly useful.

      Like

Leave a Reply to Johnnyrockets Cancel 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 )

Connecting to %s