"My rule was I wouldn't recruit a kid if he had grass in front of his house.
That's not my world. My world was a cracked sidewalk." —Al McGuire

Friday, November 01, 2019

Marquette Non-Conference Schedule Analysis

Earlier this week, Matt Norlander of CBS Sports released a ranking of the best and worst non-conference schedules. Marquette checked in tied for #14 in the country. While that definitely indicates a tough schedule, what I really like is how it is constructed. The best part of this schedule is not how hard it is, but rather how well it is set up to generate wins for Marquette against what I believe will be a harder schedule on paper than on the court.

Considering kenpom is one of the metrics factored into the NET and kenpom had the highest correlation to NET last year, I am going to compare the kenpom rankings over the years of previous schedules to determine where this one ranks. This is also the metric Norlander used for his article linked above. Further, I'm going to break down games by presumed Quadrants since that is where the Selection Committee will be looking come March.

One more note, I will strictly be using kenpom numbers for this. The reason is for a fair comparison. The alternative is to mix RPI and NET numbers, so to have one consistent metric, the one that came closest to mimicking NET was kenpom, and kenpom also goes back over Wojo's tenure for a full and fair comparison. I am adding a Quadrant 5 as well, to reflect the sub-300 home games that really drag on a team's strength of schedule. Here is how Marquette fared against these Quadrants by season in the past, as well as how many of each quadrant they play in 2019-20.

2015 2016 2017 2018 2019 Record Win Pctg 2020
Q1A 0-3 1-0 0-1 0-2 1-1 2-7 0.222 2
Q1B 0-0 0-1 2-1 1-0 3-1 6-3 0.667 3
Q2A 0-0 0-0 0-0 1-0 0-0 1-0 1.000 1
Q2B 3-0 2-0 0-1 0-1 0-0 5-2 0.714 0
Q3 1-0 0-1 1-0 2-0 1-0 5-1 0.833 0
Q4 1-1 4-0 3-0 3-0 4-0 15-1 0.938 5
Q5 3-0 4-0 3-0 2-0 2-0 14-0 1.000 1

So what does this tell us? First, Q1A games are very hard to win. That said, Marquette has flourished in the next three tiers, going 12-5 against Q1B and all of Q2. It's nice to see 4 of the projected 6 toughest games in those categories, insuring opportunities for quality wins. It's also worth noting Marquette is a combined 20-2 in Q3 and Q4 games. The 0.909 winning percentage in those two quadrants tells me that's where you want the bulk of the buy games to be. Q5 might be guaranteed wins, but I don't think you have to look further than the 2016 team to understand why you want to avoid those games. The Ellenson team had the most Q5 games and tied for the most Q4 games played and didn't go to the postseason despite one of only two Q1A wins Marquette has earned in non-con play under Wojo.

To get to the full 12-game schedule, I had to project opponents for the Orlando Invitational. Since we are using kenpom numbers, I projected the most likely opponents based on matchups, which led to neutral site games with USC and Maryland. Before we look at the Quadrant breakdowns, here's a reminder of what each Quadrant features:

Home Neutral Away
Q1A 1-15 1-25 1-40
Q1B 16-30 26-50 41-75
Q2A 31-55 51-75 76-100
Q2B 56-75 76-100 101-135
Q3 76-160 101-200 136-240
Q4 161-299 201+ 241+
Q5 300+

Let's break down the games in each Quadrant for 2020:

Quadrant 1A: Purdue (H), Maryland (N)
Marquette hasn't won a home Q1A game under Wojo, going 0-2. I like their odds to make Purdue the first, though I'm not convinced Purdue will still be Q1A come March. Maryland will be a tough test at any venue, though one of Marquette's two Q1A wins did come at a neutral site, last year against Louisville. Neither of these games are likely to hurt the resume come March.

Quadrant 1B: Wisconsin (A), USC (N), Kansas State (A)
The first thought I had was that having all of these away from home is tough, but then I noted that Marquette is 3-1 in Quadrant 1B games away from home, with the only loss at Indiana. The advantage to having these games away from home is that it gives a lot more cushion to where the teams need to be ranked. Wisconsin seems like to at worst remain top-75 while 86% of Big 12 teams have been top-75 in the past 5 years, so it seems likely that even a down Kansas State should be Q1B. USC is right on the cusp and could fall to Quadrant 2, but would be a solid game regardless, especially as it would mean a win over a Davidson team I believe will likely end up in this category. What I really love about these games is that if Marquette is legitimately good, they should be able to beat these teams on any court.

Quadrant 2A: Davidson (N)
I think this ends up being a Quadrant 1 team by the time the year is done. They have a favorable schedule and will be in the mix for their league title. Another reason why this is the most important game of the year is because of the two neutral site games above this one. With a win over Davidson, Marquette is projected to play two Quadrant 1 games. With a loss, they would be projected to play Quadrant 4 Fairfield and Quadrant 2A Texas A&M. Beating Davidson has a huge schedule impact.

Quadrant 4: Loyola-Maryland (H), Jacksonville (H), Grambling (H), North Dakota State (H), Central Arkansas (H)
Personally, I think the kenpom model is harsh on some of these teams. LMC and NDSU are both projected as Quadrant 3 teams by T-Rank and I feel have a good chance of ending up there. Grambling is another dark horse to reach Quadrant 3. The positive is that the majority of these teams project to the high side of Quadrant 4. Jacksonville could drop to Quadrant 5, but if they do, I expect they will be the only Q5 opponent we see.

Quadrant 5: Robert Morris (H)
I'll admit, this shocked me when I saw the kenpom numbers come out. T-Rank has RMU 78 spots higher in that ranking and second in their league. I'm not sure where the discrepancy is, but the Colonials return the bulk of a team that went 11-7 in their league last year while also adding two solid JUCO players. I feel confident saying this will be a Quadrant 4 team. Also, it's a good trend to see Marquette minimizing these games. Eleven from 2015-17, now just five projected from 2018-20. Well done by the schedule makers.

My final thoughts on the schedule are that there will be a little bit of a shakeup. I think Maryland should remain Q1A but expect that will be our only Q1A non-conference game. My expectation by the end of the season is that we have five games spread from Q1B, Q2A, and Q2B. I think that's the ideal schedule to bolster wins while also giving the Selection Committee some real quality to look at. On the buy end, I think it's more likely we end up with two Quadrant 3 and four Quadrant 4 games. That's a good way to spend money to get some decent wins while minimizing the number of true bottom-feeders.

No comments: