It's Not So Bad Being Trendy

I was killing time on the Internet machine the other day -- which is almost as fun as killing time by taking a meat cleaver to a wall clock and trying to sell people on your performance art -- and I came across a fancy document from the NCAA: a short, one page piece charting statistical trends in college lacrosse since 2002. Now, as the trends are almost all based on per-game measures -- which is as bright as power sliding through water on stage while smashing an electric guitar into the puddle -- there are a few things in the report that raised my interest: (1) how are these trends relating to tempo-free analyses?; and (2) what kind of impact could the new rules have on these trends? Specifically, I'm concerned with three statistical categories in the report: turnovers per game, extra-man opportunities per game, and save percentage. Let's attack these in order because linear thinking is totally awesome!

TREND: Turnovers per game has declined over the last three seasons.
HYPOTHESIS: As pace of play has decreased, so too has sloppy play.
QUESTION: How will the modified shot clock and new stringing rules impact this?

Over the last four seasons, the average Division I lacrosse game -- based on 60 minutes of play -- has lost about two possessions per year. In 2009, an average game saw about 71 possessions per contest; in 2012, that number had moved to a woeful 65.71 figure. Not only that, but the slowest teams in the country have progressively gotten even slower, seeing Villanova as the nation's slowest team in 2009 at only 60.66 possessions per 60 minutes of play and Hofstra surpassing that mark in 2012 with a mind-numbing 58.01 value.

The result of this decline in tempo? Sloppiness, apparently, is being controlled (I think). I haven't tracked turnovers in a tempo-free enviornment through the years so I'm not exactly sure if this trend comports on a possession-basis, but there's enough complimentary evidence here to say that the decrease in pace through the years has -- in some fashion -- impacted the number of turnovers per game (and the decrease similarly trends with the decrease in possessions per game). In an atmosphere where more possessions are desired under the new rules -- both in restarts and especially through the modified shot clock -- turnovers may see a comeback in 2013; combined with new stringing rules that may impact hold and control, allowing a defender to actually strip a ball from an offensive player, turnovers may find a revival as well. We'll see how this plays out.

TREND: Extra-man opportunities have declined fairly substantially over the years.
HYPOTHESIS: I have no clue.
QUESTION: How will unsettled situations -- due to new substitution rules -- and pockets that are perceived to be looser due to stringing rules impact this?

This trend was weird when I first saw it. With all the points of emphasis that referees have had over the years, I thought the trend would actually be going in the opposite direction in terms of extra-man opportunities per game. The tempo-free analysis actually shows a different aspect to this all: While extra-man opportunities per game have dropped, teams are actually playing more offensive possessions with the extra attacker. Weird, right? In 2009, the average team played about 9.95 percent of their offensive possessions in an extra-man posture; that number rose to 10.49 percent in 2012 and saw 11.07 percent and 11.34 percent marks in 2010 and 2011, respectively. So, while there are less extra-man opportunities per game, there are actually more extra-man opportunities per possession. So, yeah; the game isn't exactly getting any cleaner out there.

Here's why there's a disconnect between the per-game metrics and the possession-based metrics: The per-game metrics do not consider the decrease in number of possessions played per game. Accordingly, there is naturally going to be a drop on the per-game side because there are fewer opportunities to commit infractions that allow for the extra attacker. A possession-based approach doesn't care about that; it only cares what happens when a team has the ball, and that's not impacted by tempo. The focus here should be on the possession-based trend, not the per-game one.

As for how the new rules will impact this, I don't have a good answer. It's too unclear at the moment. Anyone that tells you different is a liar.

TREND: Save percentage is going in the tank.
HYPOTHESIS: Teams are asking goalies to do more to end defensive possessions.
QUESTION: [I have no idea what question to ask]

This is weird, and I think it has to do with how coaches are approaching their defensive postures, but I'm not exactly sure. Here's the thing: Save percentage has dropped precipitously over the years -- from around 56 percent in the early 2000's to a woeful 51.3 percent in 2012. When looking at goalie activity on a possession-basis, though, saves per defensive possession has slowly risen since 2009 -- from about 31 percent of defensive possessions ending with a save to around 32 percent in 2012. The movement isn't drastic, but when viewed in the context of save percentage (which has continously dropped), I have to ask this question: Are coaches asking goalies to do more than they have in the past?

I can't answer that question, mostly because I can't ask the number of coaches necessary to pull secondary data together to provide context. It's a weird relationship happening here, and I can't say that it correlates in any meaningful way to pace or talent or otherwise. It's just weird, and I think that it comes down to defensive approach.