Beyond the Basics: The Importance of Pacing

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.

The shot clock has done many things for college lacrosse. It eliminated stall-ball and made comebacks more possible. It limited the advantage of a dominant faceoff guy and increased the number of possessions per game.

For our purposes, it also provided some structure around which to analyze team offense. With no shot clock, if a team didn’t have a lot of goals early in a possession, it could be because they were winning games and had an incentive not to take early shots. It could also be because they just were a team that needed to probe a defense until a good shot could be generated. With the shot clock, we can get closer to a better understanding of the structural characteristics of an offensive (or defensive) system.

This isn’t going to be all about the shot clock, though. It is more a deep-dive into pacing. We’ll talk about how pacing can be used to differentiate between offensive styles and what that means for scouting opponents.

SHOT CLOCK SHOOTING 101

First, let’s build some foundational pace-based knowledge. Going back to 2019 in Division I men’s lacrosse, here are the shooting percentages by shot clock bucket (assuming no shot clock reset):

60-80s left: 30.0%
40-60s left: 29.7%
20-40s left: 28.4%
< 20s left: 27.9%

Clearly, the later in the shot clock, the worse the shooting percentage. And that’s not hard to reconcile with what we know about the game. If you have a transition opportunity and you get a shot off, that should, by definition, be a higher percentage shot. And as the shot clock gets low, you are going to take what you can get. Players that shoot low percentage shots early in a possession won’t get a chance to take many more shots. The trend makes sense.

We can also look at the number of shots taken in a possession and how that relates to the likelihood of scoring on that possession. For example, since 2019, there have been 193,072 offensive possessions. Of those, 31.2 percent saw no shots taken (think turnovers); 46.6 percent of all possessions saw a single shot taken. That means that just about 22 percent of all possessions saw more than one shot taken.

And there is no advantage to having more shots in a possession. The percentage of all possessions with goals scored actually goes down the more shots are taken:

One shot: 43.8%
Two shots: 41.2%
Three shots: 40.7%

Again, this makes sense; early shots have the highest shooting percentage. If you aren’t shooting early, your chances of scoring on any subsequent shot goes down.

PACING: IT’S A CHOICE

So far, we’ve addressed what I think of as a sort of general rule of shot-clock lacrosse. The later you get into a possession, the less likely you are to score. If you had a better look early, you’d have taken it.

But so far, we’ve talked about shots in a bit of a vacuum. There is also the somewhat more interesting question of how a team’s aggressiveness early affects their overall efficiency. If you are a team that is taking more of your shots early in the shot clock, one of two things is true. Either you have poor shot selection or you have an offense that is good at creating quality chances. And poor shot selection tends to get resolved by benching the offending player.

For example, when the first shot of a possession occurred within the first 20 seconds of the shot clock (regardless of whether it went in), the probability that the possession will result in a goal is 45.6 percent. These are the possession-based efficiencies depending on when the first shot occurred.

First shot came with 60-80s left: the possession ends in a goal 45.6% of the time
40-60s left: 42.6%
20-40s left: 41.5%
< 20s left: 38.5%

And the interesting thing to note here is that the gap in efficiency is much larger than the gap in shooting percentage.

This demonstrates that early shots are associated with good offenses and good offenses have higher efficiencies. But it’s a bit of a compounding effect as well. You can be good on two dimensions: how well you generate good offensive looks and how quickly you are able to get into your offense. A team that doesn’t generate great offensive looks from their base system may still rate as a good offense if they are quick to get into their offensive sets. It gives them more time to generate a good shot. And a great offense may not look so great if they take a long time to get it into the offensive zone.

ASSESSING/SCOUTING SPECIFIC TEAMS

I’m a stats guy, so I think the general trends are interesting. To me, stats and modeling are about trying to understand the underlying mechanics of the game. But as fans (or coaches) we also want to know: what about my team (or next opponent)? And while the trends above are general, they obscure some pretty stark differences from team to team.

For example, we can look at what each team’s shooting percentage is in the first 20 seconds of the shot clock. To illustrate that these are stylistic differences, I’ve limited this particular sample to the top 20 offenses. Among that group of elite-ish units, these are the top five in terms of early shooting:

Jacksonville: 48.6% (111 shots)
Maryland: 47.8% (90 shots)
Michigan: 40.3% (176 shots)
High Point: 39.8% (83 shots)
Rutgers: 39.5% (152 shots)

Conversely, and remember, these are still top-20 offenses, here are the worst early shooting percentages:

Virginia: 31.7% (123 shots)
Princeton: 31.7% (101 shots)
Hobart: 30.7% (75 shots)
Denver: 29.1% (103 shots)
Richmond: 22.3% (112 shots)

It’s not as if, to be a good offense, you must also be a good early shooting team. Richmond gives us a great example of how this sort of information can be useful in understanding an offense (or defense). The Spiders’ offense is much improved this season. Ryan Lanchbury is having a fantastic season and Dalton Young has filled the hole left when Richie Connell transferred to Denver. But this is not a normal offense.

As we have discussed, the normal trend is that the longer into the shot clock, the lower the shooting percentage. Richmond is the opposite. They struggle in early shot clock situations (a 21.6% shooting percentage in the first 20 seconds), but as the possession gets longer, their percentages improve. In the last 20 seconds of the shot-clock, Coach Chemotti’s team is shooting a whopping 41.9% (the average across Division I Men’s lacrosse is 26.6%).

The other interesting thing is that the offense actually takes more shots in that first 20 seconds than the average team. 33.2% of their shots are in the first 20 seconds, compared to 27.1% for Division I overall. This offense has been good, and it’s hard to recommend any sort of stylistic change based on a chart, but it makes you wonder what their offense could be if they were a bit more patient in early shot-clock situations.

THE MISSING LINK

Because we know the general trends with respect to which shots are more likely to go in, we can actually build a model to “predict” how likely a given shot is to go in. A late shot-clock opportunity at even strength when you are down one goal is less likely to go in than a man-up shot that comes early in the penalty. It is not the case in every instance, but over the course of a season, the noise filters out, and the model becomes a useful tool.

It’s useful because it allows us to calculate an “expected shooting percentage” for every team. In my mind, expected shooting percentage gives us an answer to the question: how well does your offensive system generate good scoring opportunities? This is interesting because it allows us to separate shooting percentage into its component parts. First the offense has to generate quality looks and then the shooters need to have the skill to bury them. Every team has a different level of effectiveness on those two dimensions.

These are the ten best offenses in terms of expected shooting percentage:

Now the missing link here is tracking data. My expected shooting model is noisy because it doesn’t factor in where a shot came from. A last-second shot could be an extremely high percentage shot if it comes on the doorstep, but since I only have timestamped play-by-play data, my model would give it a low probability. Adding tracking data would bridge this gap and make the expected shooting percentage model really useful (and with far fewer shots needed).

But that is still over the horizon. One day lacrosse will have player tracking and it will be glorious.

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.