Lacrosse Analytics

[TNLAX]

This was originally published in the Sat Oct 28, 2023 edition of Expected Goals, my daily newsletter.

The analysis covers the 19 games Boston College played during the 2023 season against conference opponents and teams with similar LaxElo rankings. The team's record in these games was impressive, with 15 wins and 4 losses. Our goal with these analyses is not to identify what the offense did well (or not well) overall; it's to figure out which facets of the game were most important to their success. Places where, when they were above a certain threshold, they were much more likely to win.

Key to Boston College's victories was the performance of the midfield unit. In games where this unit scored 5 or more goals, the team went 13-1 and had an opponent-adjusted offensive efficiency of 39%. However, when the midfield unit scored less than 5 goals, the team's efficiency dropped to 19% and their record slipped to 2-3, indicating a significant correlation between midfield performance and overall success. We aren't talking about a ton of goals here, but it was important to the effectiveness of the offense overall that their midfielders were a credible scoring threat.

Another crucial factor was the team's overall shooting percentage. When it exceeded 38.7%, Boston College was, again, 13-1 with a 38% efficiency mark (the lone loss was against Northwestern). But when the shooting percentage fell below this threshold, their efficiency dropped to 22%. Despite the fact that the records are the same in this and the above finding, I put more stock in the midfield scoring aspect because the gap in efficiency was so much larger (20 percentage points) compared to the shooting percentage split.

Courtney Weeks' assist-to-turnover ratio also appeared to have a bearing on the team's performance. When her ratio was greater than 1.00 (so as many or more assists than turnovers), Boston College was 6-0 and scored on 36% of their possessions. But when her ratio was less than 1.00, the team's efficiency declined to 24% and they won only two games out of five.

In fact, the assist-to-turnover ratio of the entire midfield unit also seemed to be a determinant of success. When the ratio exceeded 0.17 (a low bar, I know), the team was 13-1 games and scored on 38% of their possessions, but 2-3 when they were below that mark. When BC could get anything out of the midfield in terms of creating assists, they were much much better.

In summary, the statistical analysis suggests that Boston College's performance hinges largely on the involvement of the midfield, the overall shooting percentage, and the assist-to-turnover ratio of Courtney Weeks. Useful nuggets of information if you are preparing a defensive game plan against the Eagles.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.

I enjoy reading your posts. Do you watch game film and make sure the box scores are accurate? Or do you just rely on the box scores?

I have attended and watched to many college lacrosse games and in a large percentage of those games the person who keeps the stats just doesn't do a real good job. (woman's lacrosse isn't college basketball or football). The only stat I think may be close to 100% accurate are goals. GB's, assists, CTO's, TO's shots on goal, saves etc. are a real crap shoot.

Thanks

[laxreference]

This was originally published in the Sat Oct 28, 2023 edition of Expected Goals, my daily newsletter.

The analysis covers the 19 games Boston College played during the 2023 season against conference opponents and teams with similar LaxElo rankings. The team's record in these games was impressive, with 15 wins and 4 losses. Our goal with these analyses is not to identify what the offense did well (or not well) overall; it's to figure out which facets of the game were most important to their success. Places where, when they were above a certain threshold, they were much more likely to win.

Key to Boston College's victories was the performance of the midfield unit. In games where this unit scored 5 or more goals, the team went 13-1 and had an opponent-adjusted offensive efficiency of 39%. However, when the midfield unit scored less than 5 goals, the team's efficiency dropped to 19% and their record slipped to 2-3, indicating a significant correlation between midfield performance and overall success. We aren't talking about a ton of goals here, but it was important to the effectiveness of the offense overall that their midfielders were a credible scoring threat.

Another crucial factor was the team's overall shooting percentage. When it exceeded 38.7%, Boston College was, again, 13-1 with a 38% efficiency mark (the lone loss was against Northwestern). But when the shooting percentage fell below this threshold, their efficiency dropped to 22%. Despite the fact that the records are the same in this and the above finding, I put more stock in the midfield scoring aspect because the gap in efficiency was so much larger (20 percentage points) compared to the shooting percentage split.

Courtney Weeks' assist-to-turnover ratio also appeared to have a bearing on the team's performance. When her ratio was greater than 1.00 (so as many or more assists than turnovers), Boston College was 6-0 and scored on 36% of their possessions. But when her ratio was less than 1.00, the team's efficiency declined to 24% and they won only two games out of five.

In fact, the assist-to-turnover ratio of the entire midfield unit also seemed to be a determinant of success. When the ratio exceeded 0.17 (a low bar, I know), the team was 13-1 games and scored on 38% of their possessions, but 2-3 when they were below that mark. When BC could get anything out of the midfield in terms of creating assists, they were much much better.

In summary, the statistical analysis suggests that Boston College's performance hinges largely on the involvement of the midfield, the overall shooting percentage, and the assist-to-turnover ratio of Courtney Weeks. Useful nuggets of information if you are preparing a defensive game plan against the Eagles.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.

I enjoy reading your posts. Do you watch game film and make sure the box scores are accurate? Or do you just rely on the box scores?

I have attended and watched to many college lacrosse games and in a large percentage of those games the person who keeps the stats just doesn't do a real good job. (woman's lacrosse isn't college basketball or football). The only stat I think may be close to 100% accurate are goals. GB's, assists, CTO's, TO's shots on goal, saves etc. are a real crap shoot.

Thanks

I love watching games, but just as a fan.

I rely on the box scores that are posted by the teams for my statistics and analysis. I am sure that you are right about there being inaccuracies in the data. Some teams are better than others, but they are all going to have errors over the course of the season.

[njbill]

Well, it is very difficult to keep accurate running stats in a game. I keep some stats in most games I attend, though limited to goals (scorer), assists (sometimes), and draw controls (who got the DC). I don’t keep ground balls, saves, or turnovers. I have an immense amount of respect for those who keep stats during games.

Yes, there are inaccuracies. As far as I know, no one goes back and reviews games, frame by frame, to clean up any mistakes in the stats as they do, for example, in the NFL.

I love looking at box scores as much as the next person, but I never think the “lesser” stats are 100% accurate.

[TNLAX]

This was originally published in the Sat Oct 28, 2023 edition of Expected Goals, my daily newsletter.

The analysis covers the 19 games Boston College played during the 2023 season against conference opponents and teams with similar LaxElo rankings. The team's record in these games was impressive, with 15 wins and 4 losses. Our goal with these analyses is not to identify what the offense did well (or not well) overall; it's to figure out which facets of the game were most important to their success. Places where, when they were above a certain threshold, they were much more likely to win.

Key to Boston College's victories was the performance of the midfield unit. In games where this unit scored 5 or more goals, the team went 13-1 and had an opponent-adjusted offensive efficiency of 39%. However, when the midfield unit scored less than 5 goals, the team's efficiency dropped to 19% and their record slipped to 2-3, indicating a significant correlation between midfield performance and overall success. We aren't talking about a ton of goals here, but it was important to the effectiveness of the offense overall that their midfielders were a credible scoring threat.

Another crucial factor was the team's overall shooting percentage. When it exceeded 38.7%, Boston College was, again, 13-1 with a 38% efficiency mark (the lone loss was against Northwestern). But when the shooting percentage fell below this threshold, their efficiency dropped to 22%. Despite the fact that the records are the same in this and the above finding, I put more stock in the midfield scoring aspect because the gap in efficiency was so much larger (20 percentage points) compared to the shooting percentage split.

Courtney Weeks' assist-to-turnover ratio also appeared to have a bearing on the team's performance. When her ratio was greater than 1.00 (so as many or more assists than turnovers), Boston College was 6-0 and scored on 36% of their possessions. But when her ratio was less than 1.00, the team's efficiency declined to 24% and they won only two games out of five.

In fact, the assist-to-turnover ratio of the entire midfield unit also seemed to be a determinant of success. When the ratio exceeded 0.17 (a low bar, I know), the team was 13-1 games and scored on 38% of their possessions, but 2-3 when they were below that mark. When BC could get anything out of the midfield in terms of creating assists, they were much much better.

In summary, the statistical analysis suggests that Boston College's performance hinges largely on the involvement of the midfield, the overall shooting percentage, and the assist-to-turnover ratio of Courtney Weeks. Useful nuggets of information if you are preparing a defensive game plan against the Eagles.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.

I enjoy reading your posts. Do you watch game film and make sure the box scores are accurate? Or do you just rely on the box scores?

I have attended and watched to many college lacrosse games and in a large percentage of those games the person who keeps the stats just doesn't do a real good job. (woman's lacrosse isn't college basketball or football). The only stat I think may be close to 100% accurate are goals. GB's, assists, CTO's, TO's shots on goal, saves etc. are a real crap shoot.

Thanks

I love watching games, but just as a fan.

I rely on the box scores that are posted by the teams for my statistics and analysis. I am sure that you are right about there being inaccuracies in the data. Some teams are better than others, but they are all going to have errors over the course of the season.

Thanks and keep up the good research and sharing the analytics.

Happy Thanksgiving to all.

[laxreference]

This was originally published in the Thu Nov 2, 2023 edition of Expected Goals, my daily newsletter.

The analysis covers the Virginia offense's performance during the 2023 season, focusing on the 16 games played against conference opponents and teams with similar LaxElo rankings. The Cavaliers were 9-7 against this cohort of opponents.

Jaime Biskup's involvement emerged as a significant factor in producing that 9-7 record. When Biskup took 7 or more shots, Virginia had an unbeaten 5-0 record, scoring on 39% of their possessions. However, when Biskup took less than 7 shots, the team's record dropped to 4 wins and 7 losses with an efficiency of 33%. It's not that she had to put up a lot of points, but when she faded into the background, it coincided with days when the offense was much less effective.

The attack, on the other hand, was the driving force for Virginia. When the unit recorded 11 or more goals, Virginia went 7-0, scoring on 39% of their possessions. When the attack unit scored less than 11 goals, however, the team's record fell to 5 wins and 7 losses with an efficiency of 33%.

While they were at their best when the attack was scoring, it wasn't necessarily because they attack was creating that offense. The midfield unit's assist record proved to be another vital indicator. When the unit notched up 2 or more helpers, Virginia had a strong 7 - 1 record, scoring on 38% of their possessions. But when the midfield unit had one or zero assists, Virginia's record dipped to 2 wins and 6 losses with an efficiency of 32%. It paints a picture of an offensive system that was at its best when the midfielders were able to make the defense move, thereby opening up passing lanes to get the ball to the attackers for high-quality shots.

In summary, the performance of individual players like Jaime Biskup, combined with the collective performance of the attack and midfield units, played significant roles in determining whether Virginia won or lost their games.

If you want 5-10 minutes of team and player analysis in your inbox every morning, sign up for Expected Goals here.

[laxreference]

This was originally published in the Thu Nov 9, 2023 edition of Expected Goals, my daily newsletter.

The analysis here tries to uncover the true "keys-to-victory" for the Saint Joseph's offense, focusing specifically on data from the 16 games played against conference opponents and teams with similar LaxElo rankings. Saint Joseph's posted an overall record of 9 - 7 in these highly competitive matchups.

Key takeaways from the analysis show a strong correlation between the offensive performance of Alexa Capozzoli and the team's overall success. In games where Capozzoli took 2 or more shots, Saint Joseph's went 8 - 2, scoring on 38% of their possessions. However, in games where she took less than 2 shots, the team's record dipped to 1 - 3, with an efficiency of 22%. These shots didn't even need to go in necessarily (although that helped). The offense just worked better when she was involved.

You could tell a similar story about Emma O'Neill, but for her, it was a 5-shot threshold. The O'Neill stat is not as impactful though because they still won 4 games (out of 11) when she did not take 5 shots and the efficiency disparity was smaller than for Capozzoli. Still, it's clear that the offense was better when O'Neill was involved.

Looking at the offense as a whole, the assist-to-turnover ratio also proved critical in determining outcomes. When the ratio exceeded 0.70 (so 7 assists for every 10 turnovers), Saint Joseph's posted a record of 6 - 1 and scored on 42% of their possessions. A ratio less than 0.70 saw their record drop to 3 - 6, with an efficiency of only 26%.

Shooting percentage emerged as another important factor. In games where the team's shooting percentage was greater than 53.1%, Saint Joseph's went undefeated (4 - 0), scoring on 45% of their possessions. Conversely, a shooting percentage less than 53.1% resulted in a 5 - 7 record, with a scoring efficiency of 29%.

The performance of the attack unit was equally significant. In games where the attack unit scored 8 or more goals, Saint Joseph's maintained a perfect record (5 - 0), with a scoring efficiency of 43%. When the attack unit scored less than 8 goals, the record fell to 4 - 7, with an efficiency of 28%.

There isn't really a clear theme here related to any specific aspect of the offense. Saint Joseph's performance in the 2023 season appeared highly contingent on the offensive output of key players, shooting efficiency, the assist-to-turnover ratio, and the performance of the attack unit. Despite the lack of clear theme, the statistical thresholds uncovered here provide valuable insights into the team's strengths and potential vulnerabilities.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.

[Relax77]

Been reading some of the content on your site. You lose me at some of your rankings. Wanna break down the difference between lax Elo and regular RPI for us? Because I’m not really not feeling some of these very high and low Elos. So I’m gathering I’m seeing something wrong here. Just trying to broaden my horizons.

And I have read some prior posts and I agree. It’s not for everyone, but it’s made very easy to bypass the thread if you’re not interested.

[laxreference]

This was originally published in the Mon Nov 13, 2023 edition of Expected Goals, my daily newsletter.

Elizabeth Talluto's 2023 season with Fairfield was marked by an impressive consistency in her performance, which was underscored by her above-average individual player efficiency. Standing at the 81st percentile, her efficiency rating represented an uptick from her 77th percentile placement in 2022. The most striking aspect of Talluto's season, however, was her notable leap in shooting efficiency. She rose from the 46th percentile in 2022 to the 75th percentile in 2023, a testament to her refined technique and increased accuracy.

However, not all aspects of Talluto's performance improved in 2023. Her assist rate, for instance, saw a notable dip from the 77th percentile in 2022 to the 58th percentile in 2023. This suggests a shift in her role from a player who was fairly balanced between initiating and finishing to one who was more of a primary goal scorer.

As for Talluto's best stretch of the season, it took place between February 18 and March 4, encompassing games against Wagner, Albany, Hofstra, and Vermont. During this period, her individual efficiency rating jumped up to the 95th percentile. This high point was characterized by a top percentile shooting efficiency, a 69th percentile ball security, and a 78th percentile assist rate. In addition, Talluto had a significant share of the team's shots (19.3%) and assists (19.0%), suggesting that she was very heavily involved during this period.

On the flip side, Talluto's challenging stretch came between April 8 and April 19, featuring games against Siena, Mount St Mary's, Niagara, and Manhattan. Despite Fairfield's 3 - 1 record during this period, the same as her best stretch, Talluto's individual efficiency rating dipped to the 33rd percentile. Her shooting efficiency, ball security, and assist rate all dropped to the 27th, 28th, and 29th percentiles, respectively. Furthermore, her shares of the team's shots and assists declined to 14.3% and 2.9%, respectively.

The stark contrast between these two periods of Talluto's season highlights the critical role of shooting efficiency in her performance. When her shooting efficiency was high, she was a formidable force on the field, contributing significantly to Fairfield's successes. However, when her shooting efficiency dipped, her overall performance was adversely affected. Given the usage rate disparity, it seems likely that increased defensive attention could have been a root cause as well. The good news is that, given the 3-1 record in both periods, the Stags had other options that could step up and carry the flag.

In summary, Elizabeth Talluto's 2023 season with Fairfield was one of significant growth in her individual efficiency and shooting efficiency. Her best performances were characterized by high shooting efficiency and a substantial share in the team's shots and assists. Conversely, her challenging games were marked by a drop in these key statistics. Despite the ups and downs, Talluto consistently displayed a commendable level of skill and determination throughout the season. Her ability to contribute to the team's success, regardless of her individual performance, speaks volumes about her resilience and versatility as a player.

If you want 5-10 minutes of team and player analysis in your inbox every morning, sign up for Expected Goals here.

[laxreference]

Been reading some of the content on your site. You lose me at some of your rankings. Wanna break down the difference between lax Elo and regular RPI for us? Because I’m not really not feeling some of these very high and low Elos. So I’m gathering I’m seeing something wrong here. Just trying to broaden my horizons.

And I have read some prior posts and I agree. It’s not for everyone, but it’s made very easy to bypass the thread if you’re not interested.

Thanks for the question and sorry for the delayed response.

LaxElo is different than RPI in two main ways.

First, it incorporates data from prior seasons. Every team starts at 1500 (when they play their first game, not each year) and their rating goes up and down with each win or loss. The loser gives rating points to the winner after every contest. The amount of points they gain or lose after each game depends on whether they were expected to win and goal differential. A blow out by a heavy favorite probably would mean very few points transfer from the loser to the winner. A one-goal win by a heavy underdog could see them take a lot of points from the higher rated team. Where RPI is only about how your record looks in the given year, LaxElo is more of a program strength metric that acknowledges the fact that even before any games are played, in college sports, we have a pretty decent idea whether a team is going to be good or not. RPI is a measure of what you've done, LaxElo is an estimate of what you will do going forward.

Second, LaxElo is not meant to reward teams for past performance. It is calibrated to do one thing and one thing only: provide a reliable estimate of the chance that Team A would beat Team B. The LaxElo ratings can be converted into a win probability where the larger the gap, the higher the win probability. It's calibrated over all games to maximize the accuracy of the predictions. (That means I don't have an explanation as to why a particular team is where they are other than, those are the settings that produced the most accurate predictions across all games in DI WLAX.)

The way that the model is evaluated is like this. If there are 100 games where the favorite is given an 80% win probability by the model, then it's "accurate" if the underdog wins 20 of those 100 games. If there are 100 games where the favorite is given a 90% WP, then we want to see the underdog win 10 of them.

Some downsides to the model. It doesn't account for injuries or player movement. It's a simple model on purpose. I suspect that you could probably improve the predictions if you went through every team and adjusted their rating up or down based on graduations, injuries, transfers. That said, I'm pretty happy with the predictions, and I don't think it would be vastly inferior even if someone went through and make a tweak to every team's rating.

I wrote a longer article describing the model here if you want to read more.

[Relax77]

Been reading some of the content on your site. You lose me at some of your rankings. Wanna break down the difference between lax Elo and regular RPI for us? Because I’m not really not feeling some of these very high and low Elos. So I’m gathering I’m seeing something wrong here. Just trying to broaden my horizons.

And I have read some prior posts and I agree. It’s not for everyone, but it’s made very easy to bypass the thread if you’re not interested.

Thanks for the question and sorry for the delayed response.

LaxElo is different than RPI in two main ways.

First, it incorporates data from prior seasons. Every team starts at 1500 (when they play their first game, not each year) and their rating goes up and down with each win or loss. The loser gives rating points to the winner after every contest. The amount of points they gain or lose after each game depends on whether they were expected to win and goal differential. A blow out by a heavy favorite probably would mean very few points transfer from the loser to the winner. A one-goal win by a heavy underdog could see them take a lot of points from the higher rated team. Where RPI is only about how your record looks in the given year, LaxElo is more of a program strength metric that acknowledges the fact that even before any games are played, in college sports, we have a pretty decent idea whether a team is going to be good or not. RPI is a measure of what you've done, LaxElo is an estimate of what you will do going forward.

Second, LaxElo is not meant to reward teams for past performance. It is calibrated to do one thing and one thing only: provide a reliable estimate of the chance that Team A would beat Team B. The LaxElo ratings can be converted into a win probability where the larger the gap, the higher the win probability. It's calibrated over all games to maximize the accuracy of the predictions. (That means I don't have an explanation as to why a particular team is where they are other than, those are the settings that produced the most accurate predictions across all games in DI WLAX.)

The way that the model is evaluated is like this. If there are 100 games where the favorite is given an 80% win probability by the model, then it's "accurate" if the underdog wins 20 of those 100 games. If there are 100 games where the favorite is given a 90% WP, then we want to see the underdog win 10 of them.

Some downsides to the model. It doesn't account for injuries or player movement. It's a simple model on purpose. I suspect that you could probably improve the predictions if you went through every team and adjusted their rating up or down based on graduations, injuries, transfers. That said, I'm pretty happy with the predictions, and I don't think it would be vastly inferior even if someone went through and make a tweak to every team's rating.

I wrote a longer article describing the model here if you want to read more.

Thank you and I will take a look.

[laxreference]

This was originally published in the Sun Nov 12, 2023 edition of Expected Goals, my daily newsletter.

Isabelle Vitale's 2023 season with the USC Trojans was marked by consistency and impressive numbers in specific areas of her game. Despite an unremarkable 51st percentile in ball security, Vitale's statistics in other areas stand out, highlighting her significant contribution to the team's offense. One of the bright spots was Vitale's shooting efficiency, which was solid, coming in at the 72nd percentile. Even more impressive was Vitale's individual player efficiency, which finished in the 78th percentile, although this was a slight dip from her stellar 89th percentile in the 2022 season.

A key trend in Vitale's season was her increased involvement in all aspects of the team's offense, as evidenced by her 80th percentile usage rate and share of the team's shots. This was further evidenced by her exceptional assist rate, which was in the 93rd percentile, and her share of the team's assists in the 98th percentile, marking her as a crucial playmaker for the Trojans. The most notable factor of Vitale's season was her remarkable assist rate. Her consistent ability to set up her teammates for scoring opportunities was a major driver in her overall production and efficiency.

Analyzing Vitale's best and worst stretches of games gives more insight into what was the most important facet of her game. During her best stretch from February 19 to March 5, which included victories over Ohio State, San Diego State, High Point, and Villanova, her individual efficiency rating came in at the 91st percentile. Perhaps because she didn't have as heavy a role during this period, with a 13.6% share of the team's assists, her shooting efficiency was at a high 92nd percentile. Her ball security and assist rate were also strong, at the 71st and 74th percentiles respectively.

On the other hand, her worst stretch from March 3 to March 12 saw the Trojans go 3 - 1 against High Point, Villanova, Princeton, and Hofstra. Despite assuming a larger role with a 25.0% share of the team's assists, her individual efficiency rating dropped to the 43rd percentile. This period was marked by a significant dip in shooting efficiency and ball security, ranking at the 34th and 16th percentiles respectively, even though her assist rate remained high at the 92nd percentile. When she was at her best, her role was more-balanced and she didn't need to handle such a larger chunk of the initiating responsibility.

In conclusion, Isabelle Vitale's 2023 season was characterized by strong shooting efficiency and a remarkable assist rate, making her a vital component of the USC Trojans' offense. Her performance differences in her best and worst stretches underscore the importance of maintaining high shooting efficiency and ball security in maximizing her individual efficiency rating. Despite minor fluctuations, Vitale's consistent performance played a significant role in the team's success.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.

[laxreference]

This was originally published in the Sat Nov 18, 2023 edition of Expected Goals, my daily newsletter.

The Cavaliers' offense maintained a high level of efficiency, with a slight improvement from 39.7% in 2022 to 40.3% in 2023, keeping their national ranking steady at 4th. The significant boost came from their reduced offensive turnover rate, which dropped notably from 20.6% to 16.8%, vaulting them from 16th to 4th nationally. This reduction in turnovers was a key factor in sustaining their offensive efficiency, ensuring they capitalized on their possessions more effectively.

On the other side of the field, the defense saw a decline, with their efficiency slipping from 27.8% to 29.5%, causing their national ranking to drop from 46th to 63rd. This was compounded by the team's ability to force turnovers, which decreased from a rate of 29.0% to 26.6%, also resulting in a significant fall in their national ranking from 38th to 67th. The decline in defensive performance, particularly in their ability to disrupt the opponent's offense, became a challenge throughout the season.

In terms of possession, the Cavaliers experienced a slight decrease, with their per-game possession margin falling from +3.9 to +3.0 and their ranking dropping from 17th to 33rd nationally. Their draw-control win rate also witnessed a decline, from an impressive 63.2% to 58.2%, leading to a descent in national ranking from 9th to 19th. Extra possessions do not typically lead to victories directly, but they do take the pressure off the offense and defense, which tends to improve efficiency. They were still above-average in terms of possession-margin, but it wasn't the strength it was in 2022.

Despite these challenges, the Cavaliers' offensive proficiency, particularly their ability to keep turnovers low, was a highlight of their successful season. The offense's ability to consistently convert possessions into goals was a driving force behind their rise in the LaxElo rating, moving up 7 places to 23rd nationally. This offensive stability, even as other areas experienced declines, underscores their overall successful campaign.

If you want 5-10 minutes of team and player analysis in your inbox every morning, sign up for Expected Goals here.

[laxreference]

This was originally published in the Thu Nov 16, 2023 edition of Expected Goals, my daily newsletter.

Reilly Casey's 2023 season with the North Carolina Tar Heels saw growth and a much larger impact on the field. Standing out as the most notable factor this season was the improvement in her ball security, rocketing from the 15th percentile to the robust 81st percentile. She didn't have a huge role in the offense last year, but when she did get a chance, ball security was an issue, so the fact that she was able to play with poise under pressure this year was huge. If her role grew as it did and her turnover rate didn't improve, it would have been a real drag on the overall offense.

Casey experienced a particularly impressive stretch of performances from February 11th to February 24th, with North Carolina securing victories against James Madison, Liberty, Florida, and Virginia Tech. During this spell, she notched an individual efficiency rating in the 89th percentile. She played a dominant role as a distributor during this period, reflected in the fact that she recorded nearly 40% of the team's assists. Additionally, she exhibited a very solid 81st percentile shooting efficiency, a 92nd percentile in ball security, and a 96th percentile in assist rate. Everything was clicking.

In contrast, Casey's most challenging stretch ran from April 20th to April 30th, including contests against Duke, Clemson, Syracuse, and Boston College. Despite an admirable 3-1 record, her individual efficiency dipped to the 33rd percentile. Her role was smaller too, as indicated by a reduced 16.7% of the team's assists. This phase saw a drop in her skill-based metrics, with shooting efficiency at the 11th percentile, ball security at the 66th percentile, and assist rate at the 66th percentile, which suggests that maintaining high performance consistently across larger roles and tougher opposition is a demanding task. It is interesting that the biggest disparity here was her shooting efficiency. When she struggled, she wasn't able to create a lot of in terms of assists, and she wasn't able to fall back on creating value by going to goal as a finisher.

Throughout the 2023 season, despite a shift in her role from the previous year, Casey demonstrated a resilient and impactful presence on the field. Her 88th percentile usage rate and 82nd percentile share of the team's shots underscored the increased reliance on her contributions both as a passer and a scorer. While her shooting efficiency and assist rate saw declines from the previous year's peaks, her overall performance remained above-average, speaking volumes about her adaptability in taking on greater responsibilities. Her 94th percentile share of the team's assists and 87th percentile assist rate further solidified Casey's status as a vital playmaker for the Tar Heels.

Casey's season paints the picture of a player who embraced her expanded role and thrived, even under the spotlight of increased attention from opponents. The fluctuations in her efficiency metrics can be contextualized by her larger role, rather than a decline in skill. Her season stands as a testament to an athlete who not only rose to the challenge but also emerged as a key contributor to her team's endeavors on the lacrosse field.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.

[laxreference]

This was originally published in the Thu Nov 23, 2023 edition of Expected Goals, my daily newsletter.

Maddie Yoder's 2023 season with the Saint Joseph's Hawks was an evolution in her on-field performance, particularly in terms of ball security and shooting efficiency. Yoder elevated her ball security from the 36th to the 63rd percentile and her shooting efficiency from the 32nd to a respectable 51st percentile. These enhancements in her skill set were pivotal in her overall contribution to the team, even though her assist rate experienced a decline from the 48th to the 17th percentile. Yoder finished with an above-average individual player efficiency, which, despite a slight decline from her exceptional 92nd percentile ranking in 2022, remained solid at the 75th percentile.

Looking at Yoder's usage within the team's framework, there was a notable increase in her responsibilities. Her usage rate doubled from 4.5% in 2022 to 9.7% in 2023, and this was driven by her shooting volume, which also saw a significant rise from 9.2% to 17.1% as a share of the team's shots. Such increases underscore Yoder's importance to the Hawks' offensive efforts and her willingness to take the initiative in games.

The contrast between Yoder's best and worst stretches of the season offers a deeper insight into her performance. Her most productive period came between April 22 and May 5, against UMass, VCU, St. Bonaventure, and Richmond, where she achieved an individual efficiency rating in the 91st percentile. During these games, her shooting efficiency was int the 56th percentile, but most strikingly, her ball security was exceptional, ranking in the 97th percentile. Her assist rate also hovered around the median at the 48th percentile.

Conversely, Yoder faced challenges during her least effective stretch from February 15 to March 1, featuring matchups against Lehigh, Villanova, UConn, and Towson. In this interval, her individual efficiency plummeted to the 24th percentile. This dip was driven by a decrease in shooting efficiency to the 33rd percentile and ball security to the 51st percentile, and her assist rate was at a low 15th percentile. The stark disparity in ball security between her best and worst performances is evident and was the most salient factor affecting her game. But across the board, judging by these two stretches, her effectiveness grew as the season went along.

Despite Yoder's personal ups and downs, the team's performance remained consistent, as the Hawks recorded a 1 - 3 outcome in both her best and worst stretches. This suggests that while Yoder's individual contributions certainly had an impact, the team's fortunes did not hinge solely on her performance. The Hawks' struggles were reflective of broader dynamics within the team that extended beyond the scope of any single player's influence.

In summation, Maddie Yoder's 2023 season was marked by significant personal growth in key areas of her game. Her increased involvement in the team's offensive system, highlighted by a higher usage rate and more shots, was a testament to her developing skills and the coaches' confidence in her abilities. She was less involved as an initiator this season, but Yoder's progress, particularly in ball security and her finishing ability, made up for it.

If you want 5-10 minutes of team and player analysis in your inbox every morning, sign up for Expected Goals here.

[laxreference]

This was originally published in the Wed Dec 6, 2023 edition of Expected Goals, my daily newsletter.

In 2023, The Notre Dame Fighting Irish defense was excellent, ranking among the top in the nation in several key defensive statistics. Their raw overall defensive efficiency, calculated without adjusting for the strength of the opposing offenses, placed them in the 84th percentile. After adjusting for the strength of the opposing offenses, they rose to the 92nd percentile, a much more accurate assessment of their effectiveness. They finished the year as the 10th ranked defensive unit in the country. Their defense was particularly outstanding in limiting the number of shots an opponent was able to take per-possession, ranking in the 96th percentile nationally. They finished with an opponent-adjusted shots allowed per possession of 0.51, the sixth-best mark in the country.

However, the season was not without its ups and downs. The team's best four-game stretch came between April 28 and May 18, during which they recorded wins over Mercer and Florida and losses against Boston College (twice). Their adjusted defensive efficiency during this period was an impressive 15%, which falls in the 99th percentile. On the other hand, their worst four-game stretch, between April 14 and April 26, saw them go 4-0 against Pittsburgh, Robert Morris, Louisville, and Virginia, but their adjusted defensive efficiency fell to 28%, placing them in the 59th percentile. The key difference between these stretches was the shooting percentage they allowed, with a 4% difference between the best and worst stretches. This suggests that the ability to limit the opposing team's shooting accuracy was an especially important factor in their defensive performance.

The importance of limiting shots was also evident in their performance when we narrow in on the 15 games they played against conference peers and teams with similar LaxElo rankings. When the opposing offense took more than 0.66 shots per possession, Notre Dame's record was 1-5, with opponents scoring on 39% of their possessions. However, when they were able to keep the opposing offense below this threshold, their record improved to 8-1, with an opponent efficiency of just 19%. This highlights the critical role that shot prevention played in the team's overall defensive success.

Examining the team's defensive performance depending on the length of the opposing possessions also provides some interesting insights. Notre Dame's defense was particularly effective in limiting transition and very early offense, with their best performance coming in possessions that lasted less than 20 seconds, where they allowed goals on only 5.9% of the possessions. This is 7.5 percentage points better than the average defense. However, as the possessions got longer, their defensive efficiency dropped, suggesting that opposing offenses were more successful in finding gaps in their defense. On possessions lasting greater than 60 seconds, they were basically a league-average defense. Taken all together, this makes me feel like turnovers were a big part of their approach. You don't allow 0.51 shots/possession unless at least half of the possessions you face end in a turnover. And teams that were able to make it to the latter stages of the shot-clock had much better success, which makes me think that it was patient offenses with good ball security that were the Achilles' heel for this team.

All in all, the Fighting Irish defense in 2023 displayed both strength and areas of potential improvement. Their ability to limit shots and shooting percentage were key factors in their success, but maintaining their defensive performance in longer possessions could be a focal point going forward. Overall, their solid defensive rankings and performance against similarly ranked opponents position them as a formidable force in the defensive table.

If you want 5-10 minutes of team and player analysis in your inbox every morning, sign up for Expected Goals here.

[laxreference]

This was originally published in the Mon Dec 11, 2023 edition of Expected Goals, my daily newsletter.

The most significant decline from the previous season was seen in the Panthers' offensive efficiency, which plummeted from 30.4% (49th nationally) to a paltry 24.5% (101st nationally). This downturn was heavily influenced by their shooting percentage, which also experienced a steep drop from an impressive 46.9% (25th nationally) in 2022 to 38.9% (93rd nationally) in 2023. The considerable decline in shooting accuracy had a direct impact on their ability to convert possessions into goals, indicating that the offense struggled to find the back of the net consistently throughout the season.

Defensively, the team also took a step back, with their defensive efficiency rising to 31.0% (75th nationally), a regression from the previous season's 25.5% (30th nationally). This was partly due to their inability to force turnovers at the same rate, with the defensive turnover rate dropping from 31.0% (24th nationally) to 25.6% (74th nationally). The second-half improvement in turnover rate, from 23.9% to 25.8%, suggests some adjustment and resilience, but it wasn't enough to offset the overall negative trend.

In terms of possession, the Panthers saw their per-game possession margin worsen to -3.9 (102nd nationally), compared to -1.5 (79th nationally) the previous season. This was exacerbated by the drop in draw-control win rate, which declined from 50.2% (56th nationally) to 45.2% (85th nationally). Although there was a marked improvement in draw-control win rate during victories, rising to 55.0%, compared to a lower 42.1% in losses, it highlights that possession was a crucial factor in their overall performance.

Ultimately, the offensive unit's struggles were emblematic of the Panthers' difficult season. The sharp decline in shooting percentage and efficiency is indicative of an offense that was not only less effective at creating scoring opportunities but also less proficient at capitalizing on them when they did arise. This lack of offensive firepower severely limited the team's competitiveness and contributed significantly to their fall in the national rankings.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.

[laxreference]

This was originally published in the Sun Dec 24, 2023 edition of Expected Goals, my daily newsletter.

In the 2023 season, Albany's performance against conference opponents and teams with similar LaxElo rankings revealed critical statistical thresholds pivotal to their success. Within this 16-game subset, Albany finished with a 12-4 record, highlighting the team's capability to excel IF certain statistical thresholds were reached.

Analyzing the Albany offense, the standout trend was the influence of Sarah Falk's goal-scoring. When Falk netted 3 or more goals, Albany boasted an impeccable 11-0 record, and they scored on 34% of their possessions in those games. This starkly contrasted with the 1-4 record and 31% scoring efficiency when her goal count fell below this threshold. Her performance thus emerged as a significant barometer for the team's offensive fortunes.

Shonly Wallace's general level of involvement also correlated with positive outcomes for Albany. In games where Wallace took 6 or more shots, the team went 6-1, with a scoring efficiency of 35%. In contrast, when her shot count was below 6, the record dipped to 4-3, and possession scoring dropped slightly to 30%. This is not saying that she had to score, just that having her involved in the offense was correlated with good outcomes.

Allie Maloney's assist-to-turnover ratio further delineated wins from losses. A ratio greater than 1.00 (so as many or more assists than turnovers) saw Albany secure 3 wins out of 4 games, scoring on an impressive 40% of possessions. Below this ratio, the win-loss split was 4-2 with a reduced efficiency of 29%, underscoring the impact of Maloney's decision-making on the field. As with the Wallace fact above, the record splits here are not that large, so these are not as critical as the Falk nugget. That said, the fact that Maloney and Wallace show up here and no one else does singles those two out as key contributors.

The midfield unit's productivity was another important bellwether. When the midfield scored 10 or more goals, Albany's record was 7-1, alongside a 38% scoring efficiency. Falling short of this mark, the team's record was a less convincing 5-3, with efficiency dropping to 28%, reflecting the importance of midfield production in the team's overall offense.

Lastly, avoiding empty possessions was a key factor in games where Albany's offense found its rhythm. Taking more than 0.54 shots per possession correlated with a 10-2 record and a scoring efficiency of 36%. Below this threshold, the record evened out to 2-2, with efficiency tapering to 27%, highlighting the benefit of a more assertive and opportunity-rich offensive approach. 0.54 is a pretty low value, so this is less about second chances and more about avoiding possessions where they weren't able to get off any shots.

In summation, Albany's 2023 season was characterized by specific individual contributions that often swayed the outcomes of their games. The statistical thresholds, particularly those related to Sarah Falk's goal-scoring and Shonly Wallace's general level of involvement, were significant indicators of the team's offensive success. The performances of Allie Maloney and the midfield unit, alongside the overall shot volume, similarly aligned with the team's fortunes. These patterns suggest a reliance on key players' productivity and the importance of a patient offense that avoids empty possessions.

If you want 5-10 minutes of team and player analysis in your inbox every morning, sign up for Expected Goals here.

[laxreference]

This was originally published in the Sun Dec 31, 2023 edition of Expected Goals, my daily newsletter.

As we dissect the Duke Blue Devils' roster for the upcoming 2024 season, it's clear that the team faces a challenging road ahead, particularly in the draw control department. The staff will face the daunting challenge of replacing a significant chunk of draw control production after returning just 16% of their experience in this crucial area.

The departure of Maddie Jenner is huge; her stat line from last year, which included 187 draw control wins and a draw control-related EGA in the 99th percentile, leave big shoes to fill. The spotlight now turns to incoming transfer Margaret Anne Warner, whose performance at Davidson placed her in the 83rd percentile for draw control-related production. Warner's addition could be the salve for the Blue Devils' draw control woes, but it's clear that the unit, which ranked in the 95th percentile nationally last season, is in a rebuilding phase. This significant loss of draw control expertise could be a hurdle for Duke, potentially impacting their ability to maintain possession and control the tempo of the game.

Turning to the roster's overall experience, Duke presents a mix of veteran savvy and fresh potential. The median rotation player has appeared in 46 career games, and with 10 players boasting at least 30 career games, the Blue Devils are not short on seasoned campaigners. This depth of experience, with 13 seniors and grad students, might be the stabilizing factor for a team integrating 14 less experienced players. This breadth of in-game knowledge can be a substantial asset, offering consistency and a foundation for the younger talent to emerge and shape the team's future trajectory.

Despite the experienced roster, the Blue Devils must confront the loss of Anna Callahan, a key offensive initiator. Callahan accounted for 9% of the team's assists and held an individual efficiency rating in the 93rd percentile, highlighting her effectiveness in creating opportunities for her teammates. Although her absence doesn't leave an insurmountable void, it's a noteworthy departure that requires Duke to find a player capable of stepping into the role of a playmaker with the same efficiency and vision.

Overall, the roster composition suggests that Duke might face a transitional season, especially in draw controls, where they must compensate for the loss of all-world Maddie Jenner. Yet, the seasoned core of players provides a layer of optimism, as their experience will be relied on. The Blue Devils' success will hinge on how effectively they can leverage their experience to mentor emerging talents and fill the gaps left by key players like Anna Callahan. The coming season will test Duke's adaptability and depth, ultimately revealing whether this blend of old and new can sustain, or even elevate, the team's performance from the previous year.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.

[laxreference]

This was originally published in the Thu Dec 28, 2023 edition of Expected Goals, my daily newsletter.

The Vermont Catamounts' 2023 defensive unit performed admirably, finishing the season as the 24th ranked defensive team in the country. After adjusting for the strength of opposing offenses, they were in the 81st percentile. They faced a set of below-average offenses, so if that ranking seems too low. that's why.

In their best 4-game stretch, which spanned from March 12 to April 1, the Catamounts' adjusted defensive efficiency was an impressive 20% - a figure that would have put them in the 92nd percentile if it were sustained over a full season. This stretch included wins over Siena and Bryant, but losses against Albany and Binghamton. Conversely, their worst 4-game stretch, between April 8 and April 29, saw their defensive efficiency drop to 31%, a figure that would have put them in the 39th percentile if sustained over a full season. Their record was actually better during their worst defensive period, putting up a 3-1 mark with a single loss to Rutgers. The disparity between these stretches can be attributed to the number of shots allowed per possession. During their most effective defensive period, they were allowing just 0.52 shots/possession. That's a lot of empty possessions and very few second chance attempts. During their worst period, they allowed a much more substantial 0.66 shots per possession. That's how a pretty consistent opponent shooting percentage can lead to very different actual efficiency levels.

A closer look at Vermont's performance against conference peers and teams with similar LaxElo rankings reveals a key statistical threshold that significantly influenced their wins and losses: the opposing offense's shooting percentage. When the opposition managed to shoot above 39.5%, Vermont's record was a disappointing 2-7, with opponents scoring on 29% of their possessions. In contrast, when the Catamounts were able to limit the opposing offense to a shooting percentage below that threshold, they were undefeated at 4-0, allowing an opponent efficiency of just 17%. This highlights the importance of the defense's ability to prevent good shots and the goalie's role in stopping them.

I think the main takeaway here is more about shooting percentage than shots/possession. Their splits were much starker when looking at how their opponent shooting percentage changed. They were actually better in longer possessions than shorter ones, which is not what you'd expect if the second chances were really critical factors for the defense.

In summary, the Vermont Catamounts' 2023 defensive unit showed flashes of brilliance, with a few hiccups along the way. Their ability to limit the effectiveness of their opponents' shooting was a key determinant of their success. While there is room for improvement, the Catamounts have a solid foundation to build upon for future seasons.

If you want 5-10 minutes of team and player analysis in your inbox every morning, sign up for Expected Goals here.

[laxreference]

This was originally published in the Mon Dec 25, 2023 edition of Expected Goals, my daily newsletter.

In the 2023 season, Princeton's data against conference opponents and similarly ranked teams revealed a set of key statistical thresholds that often determined whether each outing was a win or a loss. Of the 12 games within this sample, the team finished with a record of 7 wins and 5 losses. Let's dive in and see what their keys were.

A major factor in Princeton's victories was the performance of their midfield unit, which had a significant influence on the team's offensive success. In the 8 games where the midfield unit scored 8 or more goals, Princeton recorded a 6-2 record and an impressive 41% scoring efficiency. However, when the midfield unit failed to meet this goal threshold, the team's record dropped to 1-3, with a decrease in scoring efficiency to 24%. That's a 17 percentage point gap in effectiveness.

Similarly, the performance of the attack unit, played a pivotal role in Princeton's success, signaling the need for balance. The team went 6-1 and a scoring efficiency of 38% when the attack unit scored 6 or more goals. On the other hand, a less productive attack unit was associated with a 1-4 record and a 31% scoring efficiency. Taken together, the key point here is balance. Yes, the midfield was the most important unit, given the larger efficiency split, but they needed a well-rounded set of contributions to be at their best.

The contributions of key players also emerged as significant bellwethers for the team's success. For example, when McKenzie Blake scored 3 or more goals, Princeton's record was 6-2, with a 40% scoring efficiency. However, when Blake's goal production was limited, the team's record diminished to 1-2, with a scoring efficiency of 23%.

Oddly enough, Princeton's record was perfect at 3-0 in the three games where Grace Tauckus had 2 or more turnovers. In these games, the team scored on 37% of their possessions. In contrast, when Tauckus had fewer than 2 turnovers, the team's record was a less impressive 4-5, with a scoring efficiency of 35%. Aggressive Tauckus was the most effective Tauckus, even if it came with more turnovers.

Overall, Princeton's season was defined by a strong correlation between certain statistical thresholds and their victories. A balanced offense, as well as the individual contributions of key players, proved to be telling indicators of the team's success.

Get smarter about college lacrosse in 5 minutes per day. Sign up for Expected Goals today.