Welcome to Beyond the Basics!

My name is Zack Capozzi, and I run LacrosseReference.com, which focuses on developing and sharing new statistics and models for the sport.

The folks at USA Lacrosse Magazine offered me a chance to share some of my observations in a weekly column, and I jumped at the chance. Come back every Tuesday to go beyond the box score in both men’s and women’s lacrosse.

Four weeks into the season seems like a good time to open a nice can of worms. Who is the “best” team in Division I men’s lacrosse?

I’ve done a top 20 ranking this year for the first time, and it’s clear that people have … opinions. It’s also been interesting to see how many different frameworks people use to think about who is best. Oftentimes, it’s the eye test. Sometimes, it’s mentally comparing two teams’ best wins. Rarely is it based on numbers, but then sometimes it is.

So today, let’s use some numbers to look at five teams with a claim on the title of “best.”

1951

With the win over Notre Dame, Maryland maintained its hold on the top of the LaxElo ratings with a rating of 1951. LaxElo is my implementation of an algorithm that rates individuals or teams based on head-to-head contests (it was originally developed by Arpad Elo for chess). The essence of the model is that the winner takes rating points from the loser, with the amount varied based on the relative strengths of each team and the margin of victory. The numbers don’t mean much in a vacuum, but when you see a list of all teams, it gives you a good sense of who’s strongest.

Compared to some ranking models, Elo is explicitly meant to be predictive. In fact, you can convert an Elo rating directly into a win probability. That also means that compared to some ranking methods (i.e. media polls), Elo tends to be a much more forward looking way to rank teams. It doesn’t react too much to recent results because overreacting to recent results produces less accurate predictions.

Another thing to note about Elo ratings is that they incorporate past seasons into the ranking. Each team is adjusted part of the way back toward the mean (a 1500 Elo rating is average) every offseason, but in terms of the rank order, where you ended the previous year is where you start the next season. But lest you think Maryland’s lofty ranking is just a holdover from last year, the Terps are also first in the Strength-of-Record table, too. I use a combination of LaxElo and SOR in my weekly rankings because I think you need that mix of past history and a focus on this season’s results to construct a fair ranking.

The last fun thing about Elo ratings is that you can use them to tell the story of a program’s evolution. Ohio State is one of my favorite LaxElo charts since it shows the rise and fall and rise again.

.737

What about everyone’s least favorite ranking method? RPI (Ratings Percentage Index) is first among equals with respect to the rankings the committee uses in May to select the NCAA tournament field. And as of today, your No. 1 team in RPI is Cornell (RPI rating of .737). Despite being just one of seven undefeated teams in Division I men’s lacrosse, the Big Red grab the top RPI spot because their opponents have a collective winning percentage of .708.

And that’s the important fact about RPI; it’s entirely based on winning percentages. RPI doesn’t care what the score of a game was. It just cares who won. The formula for RPI is based on a weighting of three numbers: your winning percentage (factor one), the collective winning percentage of your opponents (factor two) and the collective winning percentage of your opponents’ opponents (factor three). Those three numbers are weighted 25 percent, 50 percent and 25 percent respectively.

Here’s my issue with RPI being such a big part of the committee’s process. Because 75 percent of the ranking is based on how well a team’s opponents have done, the strength of your conference is a huge factor in your RPI. Too much of a team’s final RPI ranking is out of their control, and because RPI is such an important factor in the selection process, I don’t think that’s fair.

20.3%

Defense wins championships, right? I don’t have time to debate this with you right now. Maybe it does, maybe it doesn’t. But it certainly helps to have a good one. And if you want to know who the best defense is right now, you’ve got to head to Boston.

The BU Terriers have allowed goals on 20.3 percent of the possessions the defense has faced. You can quibble with the offenses they’ve played (Merrimack, Bryant, UMass and Colgate), but that’s why we adjust for opponent strength. The best defense (in raw terms) is Michigan, but after you adjust for the slate of opposing offenses, they drop to No. 7.

For Coach Ryan Polley’s bunch, it’s not one thing that you point to. Whenever I look at an efficiency rating, I try to dissect it. Are they turning teams over a lot? Are they not allowing teams to get shots on goal? Are they just saving shots at a high rate? For BU, the answer is: yes.

So far, they’ve allowed the sixth-lowest shooting percentage (21.8 percent). They’ve turned teams over at the fourth-highest rate (50.3 percent). And Matt Garber has a 64-percent save percentage. Do everything well. Good way to become the No. 1 defense in the country.

37.0%

I have loved watching the Ivy League this year. Turns out that the institutions, the facilities and the coaches have been able to weather the loss of several talented stars and a season of no games. In the absence of new information, it’s sometimes best to just assume that things are the same as what they were before. And so it has been with Princeton, who has the top rated offense, with a 37-percent efficiency rate against a very strong group of defenses.

As with BU, it’s important that the opponent-adjustments get made here as well. The Tigers are the fifth-best offense if you stop at raw efficiency. After applying the adjustment, their efficiency jumps four percentage points and plants them tenuously on the throne (just decimal points ahead of Maryland).

And as with BU, this is a case of doing everything very well. On an adjusted basis, they are No. 1 in shooting percentage, No. 5 in turnover rate, No. 4 in shot-on-goal rate and No. 1 in the extra-man. And all this without Michael Sowers. It seems like the Ivies have replaced those talented players with more talented players. In terms of individual player efficiency, which is measured on a 0-100 scale, the Tigers have been driven by Chris Brown (93), Alex Slusher (95), Sam English (89) and Jake Stevens (79).

64.5%

Georgetown was edged out by BU for the top defensive spot, but the Hoyas goalkeepers have earned them a claim on the best defense list. It’s up to the defensive system to prevent opponents from getting good shots, but then you need a keeper on the back end to stop the ones that do come his way. And on that measure, Owen McElroy has put the Hoyas into the top spot with his 64.5-percent save percentage. When you look at the raw save percentages, Georgetown is already first nationally with a few teams close on its heels, but after you account for the opponents faced, they stand alone.

Now as much as it is possible to look at defensive efficiency separately from save percentages, they are still intertwined. The best defenses force opponents into bad shots. I have a model that attempts to quantify the quality of the shots taken by looking at how often similar shots have gone in. And that is where McElroy does get a boost from the Hoyas’ defensive system. So far, the Hoyas have allowed an average expected shooting percentage of just 24.5 percent, which is the ninth-lowest mark in all of Division I.

Expected shooting percentage is a noisy metric this early in the year, so for now, it’s more of a trivia point than a hard-and-fast statistical fact, but it goes to show that McElroy’s brilliance has been the cherry on top of what is already an excellent defense.

LACROSSE STATS RESOURCES

My goal with this column is to introduce fans to a new way to enjoy lacrosse. “Expand your fandom” is the mantra. I want you to walk away thinking about the players and stories presented here in a new light. But I also understand that some of these concepts can take some time to sink in. And part of the reason for this column is, after all, to educate.

To help this process along, I have several resources that have helped hundreds of lacrosse fans and coaches to internalize these new statistical concepts. The first is a Stats Glossary that explains each of my statistical concepts in more detail than I could fit here. The second is a Stats 101 resource, which provides context for each of my statistics. What is a good number? Who’s the current leader? That’s all there.

And last, I would love to hear from you. If you have questions or comments about the stats, feel free to reach out.