Print Page | Contact Us | Report Abuse | Sign In | Join
Blog Home All Blogs
Search all posts for:   


Top tags: NBA  Celtics  community  data analysis  Gatorade  GLeague  partnership  public education  sponsor 

Quantifying Community Outreach as Kids "Step Their Game Up" with the Celtics

Posted By Megan Morgan, Assn. of Sports Analytics Professionals, Monday, November 13, 2017

It’s a great picture, a larger-than-life professional athlete folded into a too-small chair among a group of mesmerized children. Whether the athletes show up at schools to encourage reading, healthy lifestyle habits, or with an even more specific message to deliver, being face-to-face with sports icons from their favorite teams are experiences most kids never forget. What else about these interactions sticks with the kids? How big of an impact can an athlete or group of athletes really have on a group of children?

The Boston Celtics, in conjunction with their Shamrock Foundation and the Pagliuca Family Foundation, implemented Step Your Game Up in partnership with Lawrence, MA school system. The program is incentive-based, designed to target at-risk middle school students who are in danger of failing. With both tangible and experiential rewards like logoed gear, tickets to Celtics games, and other team-related experiences, Step Your Game Up was designed to drive improvements in participants’ grades and attendance. Members of the Celtics’ basketball analytics team were also folded into the process in order to determine if there were quantifiable effects of the program in the students’ academic performance and attendance.  

At the 2017 MIT Sloan Sports Analytics Conference, the Celtics delivered a Competitive Advantage talk on Step Your Game Up and the positive effects such a program can have on public school education. The main takeaway from this presentation was that with a committed school system (to implement the program and accurately record students’ progress) and a fantastic funding partner, a small group of athletes can indeed have a measurable and significant positive impact on the lives of a large group of children.

The Step Your Game Up program was put together with classic DoE principles to ensure the ability to determine a relationship between the incentives and the students’ performance, and the magnitude of any impact. “At risk” students were identified through industry accepted guidelines; the students invited to participate, and their parents, had to opt-in and agree to the parameters of the program; intermediate data was recorded on the students’ academic performance (in mathematics and English language arts) and attendance; and conventional analysis of the data was applied.

At the end of the first grading period, the students most at risk for failing were identified and offered participation into the program. Their progress versus eligible at-risk students who did not opt-in was compared from the first grading period to the second, and again from the second grading period to the third. The results (seen in a chart from the MIT SSAC presentation) show that the participating students’ grades in ELA were 0.4 of a grade higher during period two and also 0.4 grade higher during the third grading period. In Math, the increases were smaller, 0.1 and 0.2, yet still significant for the participating students compared to the eligible students who were not part of Step Your Game Up. The impact on attendance was a positive, although not statistically significant difference between groups of students. In evaluating the data, the Celtics’ data scientists also looked at variables such as gender, race, and free/reduced lunch programs, and found that none of those variables showed statistically significant differences within the results.

At this writing, the program has been in place for less than three full cycles, so there are no metrics on long-term effects on the students’ grades, attendance, or other success parameters. However, as the Celtics’ team emphasized during their Sloan Conference presentation, because this was designed to target the most at-risk students (those getting D’s and F’s), the element of the program that requires the students to commit to something and follow through with it in a public/community setting likely also delivers a sense of confidence that will carry through well beyond the final marking period. In the short term, improving grades by a half a grade or so for hundreds of students is indeed a success, and ideally will allow each individual to reassess his or her potential as to what they can achieve in the classroom.

With Step Up Your Game Up as proof, along with the countless hospital visits, community center drop-ins, and volunteer hours athletes put in around the communities where they play, it is indeed possible to develop programs that inspire significant positive effects in the education and lives of large groups of students.

Tags:  Celtics  community  data analysis  NBA  public education 

Share |
PermalinkComments (0)

Gatorade and the NBA G League: Beyond Buyer and Seller

Posted By Megan Morgan, Assn. of Sports Analytics Professionals, Monday, November 6, 2017

Integrated: adj – combining or coordinating separate elements to provide a harmonious, interrelated whole

In sports, there are transactional sponsorships, where a buyer pays a seller for a package of assets without much more to the exchange (signage, program ads, etc.).  There are marketing partnerships where constituents benefit with an exchange of assets beyond the standard buy/sell (an “official hotel partner” gets a guaranteed number of room nights, the sports property gets discounted room rates, fans get contests and promotions, the broadcast partner gets an ad buy). Then there are relationships that transcend the sponsorship space into fully integrated partnerships, where the exchange of assets is beyond transactional and the ultimate value may not be calculated within the window of the contract.

The extended partnership announced in early 2017 between Gatorade and the NBA’s Developmental League is one of those truly, fully integrated partnerships, where indeed both parties expect a “combining of separate elements to provide for a (mutually beneficial) interrelated whole.”

The now NBA G League (formerly the NBA D League, once the NBDL) kicked off its 17th season at the beginning of November. The developmental league of the NBA consists of 26 teams, all tied to NBA teams – most owned by their parent clubs, has long been a proving ground for both the talented basketball players on the rosters as well as ideas and technology of interest to the NBA and its member teams.

The name change came with the execution of a seven-year extension of an existing deal the league signed with Gatorade, one of its inaugural marketing partners. Presenting the “Gatorade” League, is hardly a naming rights deal for the sports nutrition and performance company. In fact, Jeff Kearney, Global Head of Sports Marketing for Gatorade, might characterize that as the least important asset in this deal. 

Among the ways this mutually beneficial, integrated partnership delivers: The NBA and its clubs will get the full advantage of the Gatorade Sports Science Institute and its battery of nutritionists, trainers, and scientists to help players reach their goals. From nutrition to recovery to rest and more, everything will be focused on helping players optimize for peak performance. At the Stadium Sports Marketing, NBA G League President Malcolm Turner said, “Opt-in was never a problem, we are already a test bed for the league. Development means a lot of things.  This is an important resource for us that will help our players facilitate and accelerate the achievement of their goals.”

Gatorade will have at its disposal a cache of elite athletes with a known schedule of games, practices, and travel (often including grueling bus rides and unfavorable time tables) to test products and protocols. During the same panel as Turner, Kearney discussed Gatorade’s goals, “We don’t do naming rights deals. If the success is measured in media impressions, we failed. This will be measured on the impact we make by creating new products, the innovations we make… the key learnings that drive innovation.” This process is old hat for Gatorade, in fact, it is the very foundation for the brand itself. Having the opportunity to turn their relationship with the “second best basketball league in the world” into more than a standard sideline/product deal, Kearney said, “as a founding partner of D-league, it would have been easy to rubber stamp a renewal. How do you break through the clutter of sponsorship space and relive the story of 1965? With the NBA and partner leagues at the pinnacle, and basketball being fastest growing sport in the world and they are going to give us a league to play with? We are in!

So, when will the parent teams/league and fans taste the fruits of this partnership? Turner pointed out, while the league and Gatorade are playing the long game with this relationship, the on-court product will pay the earliest dividends. How the scientists help athletes to perform at their peak levels, reduce recovery time, and accelerate to their goals will be reflected in the G League play throughout the season. Further, fans can expect some original content chronicling the work of the G League athletes and the scientists at the GSSI as the 2017-18 season unfolds.

This content will only help elevate and expand brand awareness of the NBA G League.  If anything, this is inverting the process of renting space and brand association, as Turner said during the panel discussion, the league has had a lot of change happen very quickly and is proud to have a world class brand like Gatorade help tell the story and introduce G League basketball to even more fans.

For basketball fans watching G League games over the winter, in addition to players fueled by Gatorade Sports Science, they will see some other notable differences from standard NBA rules in this “test pad:”

·       Each team will be entitled to a “Reset Timeout” in the final two minutes of the fourth quarter and the final two minutes of any overtime period. These TO’s do not allow for a huddle, and do allow for all other elements of a timeout.

·       Shot Clock Reset: The 24-second clock will reset to 14 seconds after an offensive rebound or when the offense otherwise retains possession after the ball contacts the rim.

·       Instant replay will occur for these five circumstances: flagrant fouls; 2-point/3-point FG attempts or fouls; made basket at the end of period; foul at the end of a period; altercation between two or more players.

·       There is a Coach’s Challenge, one per game, to be used at any point during regulation or OT, on only: fouls called, goaltending/basket interference, or out-of-bounds calls.

·       The away-from-the-play foul rule (AKA anti- Hack-a-Shaq) occurs when illegal contact is made by the defense either deliberately away from the immediate area of offensive action, prior to the ball being released on a throw-in, or both. If this occurs at any point in the game, personal and team fouls are assessed, and one free through attempt will be awarded to any player in the game at the time the foul was committed.

·       Five team fouls per period (up from four) before free throws are awarded.

·       There is a whole difference TO structure: each team gets seven, all are “team timeouts,” not full or thirty-second, and they are implemented with restrictions, among them, limiting teams to two team outs in the last 3:00 of the fourth quarter.

·       One third of the teams will implement wearable technology during games this season.

In 2016-17, 17 NBA teams finished the season with seven or more players who had G League experience. It is only a matter of time until 30 NBA teams finish a season with an opportunity to incorporate G League proven GSSI products on the bench as well. 

Tags:  Gatorade  GLeague  NBA  partnership  sponsor 

Share |
PermalinkComments (0)

Data Deficit: No Strokes Gained for LPGA Tour

Posted By Kyle Stefan, Founder, 54 Analytics, Friday, August 18, 2017

Where are the Strokes Gained for women?

Team Europe’s Solheim Cup analytics usage shows larger problem with antiquated LPGA Tour stats

By Kyle Stefan

As the best professional women’s golfers from the United States and Europe converge this week on the Des Moines Golf and Country Club, one side has a quantifiable advantage in this 12-player team match-play Solheim Cup competition.

United States team captain Juli Inkster has prioritized player personality testing, telling the LPGA media on a July conference call that she is making her Solheim Cup decisions “75 percent on personalities last year and 25 percent on golf…maybe 50/50 this year. Everyone has taken a personality test…and we’ll map out some strategies.”

European captain Annika Sorenstam based her playing career on the opposite of that approach. “I’m very strategic,” Sorenstam said on the same call. “Some people think I’m very analytic. I use that a lot.” Her captaincy in Iowa will follow suit: Team Europe announced in February that it would utilize the services of The 15th Club, a UK-based performance analytics company that provides data-based insights to golfers.

Whether that edge is alone enough to produce a Solheim Cup win is to be determined. Danny Willet credited 15th Club insights as key to his breakthrough 2015 Masters win. The service also consulted for Europe’s triumph in the 2016 EurAsia Cup, along with the losing 2016 European Ryder Cup team (Darren Clarke captained both events). The European women enter the 2017 Solheim Cup as 2-1 underdogs and captain Sorenstam wants to identify and exploit even the slimmest of edges – by using the best available information on player performance to recommend pairings, strategy, course management, and in-round decisions – in hopes of lifting the trophy on Sunday.

The larger issue isn’t the adoption or application of golf analytics and data-based insights. Whether a feel-based or chemistry-based approach is antiquated, or an analytics-based approach is modern, is certainly up for debate. Captain Inkster is a Hall-of-Famer player in her own right and captained the winning Solheim Cup team in 2015. She’s free to manage as she pleases and her tactics may well win again this year.

But where is the LPGA Tour’s data so their players and coaches can access the most modern and correct information and make their own decisions?By not activating Strokes Gained collection, building a database, and providing the info to its players – and outsiders like the 15th Club to run deeper analytics and generate probabilistic insights – the LPGA Tour is doing its membership a grave disservice.

The PGA Tour developed ShotLink in 2004 to collect data on every shot played on tour: measuring the lie of the golf ball and its proximity to the hole. The system has been developed tremendously since its inception. Third parties like 15th Club have produced mathematical truths on everything from round strategy to long-term career success probabilities – all entirely related to the men’s game.

The LPGA Tour has yet to implement such a system. Why? Especially in 2017, and especially with the movement across business, tech, education, and life to big data and mathematically-driven decisions, the time to act is now.

Information is increasingly important to strategy and gaining competitive advantage. Yet, the counting stats tracked and promoted by the LPGA Tour – fairways hit, greens in regulation, sand saves, putts per round – are as antiquated as 1980s golf and can be misleading at best as actual indicators of golf performance.

Collecting Strokes Gained data is not cost- or tech-prohibitive anymore. It’s certainly not labor intensive. There are user-friendly smartphone apps that do a reasonably decent job of this for $9.95 per month!

It’s no secret that the LPGA Tour lags behind its male counterparts in terms of total purses. Excluding majors, World Golf Championships, and Playoff events, the PGA Tour played 34 tournaments for $210 million in prize money in 2016. Excluding majors, LPGA Tour players competed at 28 events for $46 million.

In theory, one immediate way to make inroads on the pay gap would be to improve the product through better play, lower scores, and longer careers from women’s players – generating more star power for the Tour to promote and monetize. A clear roadmap to this can be developed through efficiencies generated from comprehensive Strokes Gained data analytics.

Odds are, there are more than one player molded like Annika Sorenstam who would like to implement modern golf strategies based on proximity to the hole instead of simplistic counting stats. And when your earned income, career, and livelihood are based on beating the field each week, it would be extremely beneficial to know – precisely – how performance stacks up against said field on a week-to-week and season-long basis.

Implementing a shot-tracking system is entirely viable for the LPGA Tour to help itself and move into the 21st century on the statistics front. It is beyond time for the LPGA Tour to provide their players the opportunity to utilize data to achieve competitive advantage.

A European win this week would highlight the importance.

Kyle Stefan is the founder and owner-operator of 54ANALYTICS, a data-based sports consulting firm, and is a professional participant of ASAP.

This post has not been tagged.

Share |
PermalinkComments (0)

Atlanta United... Expansion Outlier

Posted By Megan Morgan, Assn. of Sports Analytics Professionals, Wednesday, August 16, 2017

At 10-7-5, with 35 points through week 22, Atlanta United is having a banner inaugural season in Major League Soccer. The expansion team is 4-0-2 in its last six games and would make the Eastern Conference playoffs as the sixth seed if the season ended today. Their goal differential of +13 ranks third in all of MLS, and they are ranked ninth league-wide in the Supporters’ Shield standings. Assuming this level of play continues through their final dozen games, Atlanta United is poised to have the best ever opening season by a true expansion team across North American professional sports leagues.

Dating back to MLB’s 1961 addition of the Angels and (now) Rangers, ushering in that league’s expansion era, not a single true expansion team in MLB, the NFL, the NBA, or the NHL has posted a winning record in its first full season. The closest any teams have come to .500 or better are the Carolina Panthers, who posted a very respectable 7-9 record in their first NFL season, and, coincidentally, the NHL’s Florida Panthers who nearly broke even at 33-34-17 in their maiden 1993-94 campaign.

Not only would posting a winning record set a new gold standard for the regular season, Atlanta United is sitting at an 89% probability to make the MLS playoffs according to FiveThiryEight Club Soccer Predictions. Taking the pitch in the postseason would set a challenging precedent for all future expansion franchises – not to mention future United teams. None of the 37 teams in this review played beyond the regular season in their first year of competition, and only three teams made it to the postseason in year two: the 1996 Panthers and Jaguars of the NFL and the Diamondbacks played October baseball in 1999. The range was 2-19 years to break that postseason barrier for the teams in the study. Playoff rounds were added as the leagues grew, and in the NBA and NHL more than half the field traditionally play in the postseason. The mean is 7.35 years to make an initial appearance, and the mode is 5 years (seven teams).

With expanded playoffs in all sports, making the postseason, doesn’t necessarily translate to playing for championships. Of the 37 true expansion teams added to the big four leagues in the modern era, 24 have played in the last game of the season and only 13 have won a title. Three teams have run the table, hoisting the championship trophy in their first postseason appearance: the 1969 (Miracle) Mets, the 1997 Marlins, and the 2000 Ravens. Atlanta United probably wouldn’t mind following the Mets’ and Marlins’ lead by notching their first winning record, first postseason appearance, and first league championship in the same season. As of this writing, Atlanta United have 12 regular season games left to play, eight at home. They have a 6-2-1 record at home so far this season at Bobby Dodd Stadium. The remaining home games will be in the brand-new Mercedes Benz Stadium, a venue, if proves as advantageous as the temporary home, will ensure the most successful debut season of an expansion team in North American modern professional league history.

Expansion Teams' Performance Data: NFL, NBA, MLB, NHL

[* Current Name

^ Note that when the NBA awarded Charlotte an expansion franchise in 2004, it also awarded it the old Bobcats/Hornets records from 1998-2002 when the original Charlotte franchise moved to New Orleans, thus effectively making the Pelicans the “expansion” team retroactive to 2002, with records only dating back to the 02-03 season. That is not fair for this comparison because it is the Charlotte team roster that was created from scratch and went 18-64 in 2004-5, while the veteran New Orleans team posted a 47-35 record in 2002, the year it was deemed an expansion team on paper. This chart expresses data the opposite of how official NBA stats would present it.

# 1981 was a strike shortened season and MLB modified postseason qualification such that a first-half champion and second-half champion in each division (east/west, AL/NL) played a five-game series to advance to the LCS, doubling the number of playoff teams to 8 for one season only.]


 Background on Expansion Teams:

The teams in this comparison had to build their inaugural season rosters from scratch, as opposed to coming from another league with personnel who had already been playing competitively together at a relatively high level. Each professional league has their own restrictions for how expansion teams sign players to their rosters, be it through the amateur draft, an “expansion draft” whereby the new team is allowed to select players from a list of players from the other teams in the league not otherwise protected by those teams according league regulations, and of course free agency.


In the NFL, there are only two franchises, the Chicago Bears and the Arizona Cardinals, which are founding members of original league formed in 1920, the American Professional Football Association.  Since there was a significant amount of instability with relocations and mergers during the early years, it makes sense for this exercise to consider expansion teams joining the league after the 1970 merger of the NFL and AFL: Seahawks (1976), Buccaneers (1976), Panthers (1995), Jaguars (1995), Ravens (1996), Texans (2002).


Similarly, with the NBA, there was a lot of expansion and contraction with the league consisting of anywhere from 8-18 teams from 1946-76.  In 1976, the NBA treated the addition of four teams formerly with the ABA as expansion teams rather than a merger, however this significantly stabilized the league, which has grown to 30 teams.  Teams added since 1980 will be considered: Mavericks (1980), Heat (1988), Hornets^ (1988 NO), Timberwolves (1989), Magic (1989), Grizzlies (1995), Raptors (1995), and Hornets^ (2004 NC).


MLB has arguably the cleanest line of demarcation for expansion in 1961 when the then Washington Senators (Texas Rangers) and Los Angeles Angels were the first additions to the league since the turn of the century. Also in this comparison: Astros (1962), Mets (1962), Royals (1969), Nationals (1969), Padres (1969), Brewers (1969), Mariners (1977), Blue Jays (1977), Rockies (1993), Marlins (1993), Diamondbacks (1998), and Rays (1998).


The NHL will play with 31 teams in 2017-18 when the Las Vegas Golden Knights take the ice, the first new team in the league since Y2K. Between 1967, when the league had doubled from the “original six” to 12 teams, and 1979, when the NHL settled in at 21 teams, there were 10 additions, one contraction, and a lot of relocations.  In 1991, the modern era of expansion began, with the teams considered here: Sharks (1991), Senators (1992), Lightning (1992), Ducks (1993), Panthers (1993), Predators (1998), Jets (1999), Blue Jackets (2000), and Wild (2000).

This post has not been tagged.

Share |
PermalinkComments (0)

Five Times Equipment Rules Changed and it Mattered… or Didn’t

Posted By Megan Morgan, Assn. of Sports Analytics Professionals, Wednesday, July 19, 2017
Updated: Wednesday, July 19, 2017

With the recent debate about “is it juiced or isn’t it?,” surrounding the five ounce, not quite three-inch orb, always with 108 stitches, hurled 60’ 6” by MLB pitchers, we looked at some of the times leagues and organizing bodies made changes to the equipment standards in their sports and the results those changes had on scoring and performance.

1.       1981: NBA installs breakaway rims in all league arenas.  This was a pace-of-game and player safety issue that came into play with Darryl Dawkins first shattered backboard in Kansas City on November 13, 1979.  The new backboards, which were already in use in some NCAA arenas, did have a discernable effect on scoring or shooting percentage.  However, when the NBA further updated the rims in 2009 to the 180-degree breakaway hinges, there was at least anecdotal evidence from the players that the ball did react differently when impacting the new construction.  Over the course of the entire 2009-10 season, the players effectively adjusted to the new rims and Shaq-sized dunk forces were held at bay.

2.       2005: NHL downsizes goalie equipment (leg pads were limited to 11”, catching glove circumference was reduced from 48” to 45”, and blockers were streamlined to 15” from 16”, and form-fitting jerseys were required), which contributed to an increase of about one goal per game in the 2005-6 (6.16 GPG) season as compared to the 2003-4 season (5.14).  Of course, the equipment change came in conjunction with rule changes such as eliminating the two-line pass and addition of a no-handling zone for goalies behind the net following the lockout of 2004-5, as well as an increase in the number of power plays, as noted midseason by the New York Times.  Stay tuned for the full-season data on reduction to the size of goalie pants that were introduced midseason 2016-17 and the brand-new chest protectors.

3.       2006: FIFA introduces the +Teamgeist ball for the World Cup, improves ball handling and consistency, with no correlation to scoring. With 14 panels glued together (reduced from the traditional 32 hand stitched panels).  With far fewer and now flatter seams, the new ball was markedly more consistent coming off the foot.  Bonding the water-resistant panels to one another instead of stitching them also reduced the water absorption during a standard match, reducing weight gain to less than 0.1% of the original weight, negligible with respect to FIFA’s maximum allowable 10% weight gain.  The impact on aerodynamics, however, had mixed reviews.  Smooth spheres travel more erratically than textured spheres because they create more turbulence/drag to disrupt the flight (that’s why golf balls have dimples).  In this case the effects of the increased drag were more pronounced at higher speeds – longer, harder kicks.  The Brazuca ball, introduced for the 2014 World Cup in Brazil, solved for each of these issues.  It has only six panels with bonded seams, and the entire surface of the ball is textured with tiny bumps.  Fewer seams mean fewer erratic kicks; waterproof material and seam bonding ensures consistent ball weight throughout a match; a uniformly textured surface means smother laminar airflow for improved aerodynamics at any speed/distance; and the symmetrical panel orientation means minimal dispersion in flight path based on orientation of the ball at impact.

4.       2008-09: Speedo LZR racer polyurethane swimsuits were worn in 23 of 25 world record setting swims in the 2008 Beijing Olympics and in 43 record breaking swims at the 2009 World Aquatics Championships.  The reports and test results on the added buoyancy and lowered drag created by these neoprene (non-textile, non-porous) racing suits had caused the international governing body FINA to announce a ban to take effect in 2010.  Although the announcement was made in 2008, the suits were legal for competition until June 1, 2010.  The impact the suits had on the efficiency of elite swimmers was astounding, as the records were broken in 08-09 by margins two to five times of those posted over the previous two decades.

5.       2011: NCAA changed the test standard and limits on bat liveliness (from BESR to BBCOR), reducing batted ball speed (BBS) from composite bats to be equivalent to wood bats, resulting in a drop in homerun production by 25% for top DI sluggers.  A combination of an imbalance of increased offense and concerns for infielder safety due to high batted ball speeds led to this change.  Due to the hollow construction of aluminum and composite bats, the resulting trampoline effect, and the shifted MOI, balls were exiting the bats at speeds too high for infielders to have adequate reaction time, and with a force that could cause severe injury. It was also contributing to offensive production outpacing pitching and defense: In 2007, Division I teams were hitting homeruns at a rate of 0.68 per game; the rate increased to 0.84 the following year and was 0.96 in 2009; and from 2005-10, the top 50 homerun hitters averaged 19.4 a year ( archived stats).  After the change limiting BBCOR, the top 50 homerun hitters went yard an average of 15.5 times a season from 2011-17, one of several stats indicating the goal of reining in the BBS was accomplished (

(6.)       20??: If the NFL does indeed create a standard for the sticky gloves favored by virtually every receiver and defensive back in the game, will it draw the line with the current products or will there be some rollback?  There are no regulations dictating the performance factors of the glove, such as the level of tackiness, materials used, or thickness of the gloves – the colors, logos, etc., are all governed by the NFL’s uniform policy (Article 4, Section H). When a decision is made on the standard construction and performance specifications for gloves worn by non-interior linemen, we will learn a lot about how the league values highlight-reel catches and interceptions.

This post has not been tagged.

Share |
PermalinkComments (0)

What's Next? From VORP to the Goose Egg to...

Posted By Megan Morgan, Assn. of Sports Analytics Professionals, Tuesday, May 23, 2017
Updated: Tuesday, June 13, 2017

"What does that even mean?"  That question can be as complex as some advanced stats themselves, in that one could be asking: how is it counted/measured, what makes up the stat, what is a good (or bad) value, or what is that supposed to tell me about player X?  An equally valid question being asked since the dawn of complex statistics in sports is, so what?

It is that last question that has given rise to the debate(s) surrounding advanced statistics which have permeated virtually every team sport played beyond the neighborhood sandlot.  There are plenty of basic stats – accumulated stats like points scored and homeruns hit, and rate stats like free throw percentage and batting average – that may be interesting and impressive (or horrifying) on their own, but not necessarily informative when it comes to the big picture of an individual’s overall contribution to the team’s record.  And then there are the “advanced” stats that either combine two or more basic stats into a more revelatory and/or informative form of data or analyze performance across a complex set of criteria.

Nate Silver recently published a piece on which he presented an improved statistic for measuring the value of a relief pitcher, particularly a closer.  His premise is that the save is a wildly overrated stat that does little more than count a pitcher’s number of accumulated saves, which I happen to agree are based on a rather arbitrary set of situational conditions.  One of the best part of this article, other than perhaps fueling the fire within Goose Gossage, is that it is a fantastic How To guide for developing an advanced statistic.  Consider this the 21st century stat geek version of Schoolhouse Rock’s How a Bill Becomes a Law.  For anyone interested in turning his/her hypothesis into the next VORP, Silver provides a fine process blueprint.

Silver walks through all the rationale in developing the Goose Egg.  He meticulously breaks down the situations in which pitchers are rewarded or penalized in the algorithm, provides the context around why each element was included in the calculation (or was not), all the while mathematically relating the performance to the outcome of the game – which is, at the end of the day, the ultimate relevant team stat.  Silver then recalculates the career performances of some of the game’s best relievers using the Goose Egg criteria.  He then validates the premise that this stat is more germane than the save by comparing it to the already accepted Win Probability Added statistic.  (Spoiler alert: it is, significantly.)

Silver’s intention was not to demonstrate how an advanced statistic is born, I am serving it up as that.  Once he has the Goose Egg sorted, he digs even deeper, further refining the statistic to “correct for” ballpark and league.  In doing so, he layers on the concept of WAR (wins above replacement).  This concept is one of my all-time favorites in the world of performance analysis, and has been since the 90’s when Keith Woolner first introduced us to VORP. 

Value Over Replacement Player at its introduction was a concept about which many people were skeptical could be accurately calculated….there were so many variables! In fact, the first keynote speaker at the first MIT Sloan Sports Analytics Conference (2007) was then Toronto Blue Jays GM J.P. Ricciardi, who said during his remarks that he supposed he should start paying closer attention to VORP – if he could figure out what it was telling him.  At the end of Ricciardi’s remarks, Keith Woolner himself was called on to start the Q&A session and offered to walk Ricciardi though the stat.  For those who want to brush up on it, Baseball Prospectus published an overview of VORP by Derek Jacques in 2007.

The reason VORP, or WAR, or whatever the specific version of this stat is in a given sport or for a given position, is so valuable is because of just that…done correctly, the calculation factors in all of the components of scoring for one’s team, adjusts for position and venue where applicable, and allows for offense and defense to be included for sports where players play both ways.  So, thank you, Keith Woolner, for winding up this particular statistical top, may it spin on and on.

Another revolution in the performance analysis landscape is the grading system used by Pro Football Focus, whose grades appear in NFL on NBC broadcasts and on countless fantasy football platforms.  Understanding the philosophy behind the grading system is as critical as understating the algorithm itself.  In a nutshell, PFF grades each player on how well he executed his portion of the play, without regard to the actual outcome of the play.  That is, if a play broke down because of factors out of a given player’s control, his score on that play is not impacted.  With eleven players on a side, each with a speicif assignment, it’s probable that on any given play quite a few players will execute their roles satisfactorily and their team will come away with an unsatisfactory result.  For example, if a quarterback stands in the pocket, executes a good pass under pressure, and the receiver drops the pass, the quarterback is not penalized for that incompletion in the PFF grading system, nor are the linemen.  Even though this process requires a subjective review of each snap, the system itself has become so robust, with enough built-in checks and balances, that PFF introduced in-game scoring during the 2016 NFL season.

As Nate Silver proved with Goose Egg, and the integration of PFF grades into mainstream platforms has indicated, there is a ton of white space in the world of advanced statistics.  There is probably as much white space as any in the realm of global football (soccer, to me).  The reasons are many why the world’s most popular sport lags behind so many others when it comes to dissecting it by the numbers.  To name a few, the rate of scoring is about the lowest of mainstream team sports, the types of shots on goal are widely disparate (a header from 10 feet, a ball booted from 40 feet, all else), and there are limited “basic” stats upon which to build advanced metrics.

For example, it was popular for a while in some soccer circles to adopt PDO from ice hockey, which had gained wide acceptance and was further modified to accommodate power play and other game-specific situations.  (PDO is 10 times the sum of a team’s shooting percentage plus its save percentage, or 10(Sh%+Sv%).)  For soccer, however, it had significant criticisms and never really caught on.  Among the criticisms was that the equation suggests that on the soccer pitch, offensive performance and defensive performance are related, and since this has not been proven to be the case, those are essentially two unrelated numbers combined in an attempt to draw a conclusion.  It is not an entirely useless stat in soccer, as Richard Whittall pointed out in 2014 for 21st Club.  However, “in the same way you can do horrific things with a nail gun (happy Halloween!), so too can you wreak havoc with a simplistic, one-size-fits-all approach to PDO.”  (Incidentally, Hull City was relegated after the 2014-15 season.)

I, for one, am looking forward to the inevitable onslaught of soccer analytics that are sure to permeate our statistical lexicon any day now that will, much like VORP once did, account for multiple performance variables such as distance from goal, angle of attack, shot/attempt type (head/foot), fast break, time of game, etc.  With the wearable technology teams are utilizing and the precision data that is being collected from the pitch, the biggest question will be whether the teams will be willing to go public with their revelations.

This post has not been tagged.

Share |
PermalinkComments (0)

Who Owns the Data?

Posted By Megan Morgan, Assn. of Sports Analytics Professionals, Wednesday, March 22, 2017
Updated: Wednesday, March 22, 2017

As the 11th Annual MIT Sloan Sports Analytics Conference was taking place in Boston the first weekend in March, bringing together thought leaders and innovators from across all disciplines on and off the field in virtually every sport, several other meetings and conference were taking place around the world and one question pinged on the radar on all of them: “Who owns the data?”

In the sports landscape, the question is complicated in that it is applicable to multiple domains and constituencies.  Generally any data being collected about an individual that can be tied to the individual is with consent; everyone has clicked on “terms and conditions” and “privacy policies” for their online transactions, and team sports athletes have collective bargaining agreements and contracts to protect them.  It is the downstream use of the data where the alarm bells may start to ring.

At the Sloan Conference, as Nate Silver was interviewing Adam Silver, this very subject came up.  Nate asked Adam if players would one day be using wearable technology in games, and the expected follow-up questions such medical privacy, CBA issues, what data makes it to the public domain (broadcast relevance), were addressed.  The NBA Commissioner said, “yes, we will get to a point where players will use wearables in games… because of all of your follow-up questions, we determined that collective bargaining, with just a few owners and a few players and some lawyers around a table, was not conducive to coming up with the right framework.  We said we both have an interest in doing this; let’s empanel experts – physicians, technologists, etc. – let’s come up with a standard so that it’s not just any wearable…so that we have belief in the data. We are going to move there in conjunction with the Players’ Association, with a protocol that protects everyone.”  It was clear that in the foreseeable future NBA players will be using wearable technology in games, and it was also clear this is too convoluted an issue to solve in the window of time the league and the union expected to come to terms on everything else on the table during the collective bargaining process.

On the same weekend, there was a technology conference in India at which an Internet of Things panel discussed in depth who owns the data generated by IoT devices; everything from the data generated by smart automobiles to smart household appliances to municipal technology (red light cameras, toll trackers, facial recognition software) was discussed.  Regardless of who winds up owning it, knowing what to do with the data, how to mine it and use the insights to inform the next round of decisions is the real value, as Dan Rosenbaum noted from Mobile World Congress in Barcelona here.

A significant distinction relevant to wearable technology used across the sports industry is this: the device data versus the physiological data.  It is reasonable to expect the technology partner to want to retain co-ownership of the device data, such as battery life, range, accuracy, and such to be able to maintain a robust product development process.  However, the specific physiological data of the wearer is not necessarily relevant to the technology company’s R&D process, and the wearer could expect some privacy.  The collector of the physiological data, however, be it the athlete’s team or equipment sponsor, may also be a right’s holder to that data, and that is where ownership could get murky.  For team sports, as Adam Silver alluded to above, any further dissemination would fall into some version of collective bargaining.  For athletes in individual sports, or working with another party, that is not the case.  The athlete is on his or her own to negotiate the narrowness of the audience to whom data can be distributed. 

As for marketing data, the public has come a long way in being comfortable with standard practices associated with digital transactions.  To a certain degree, individuals implicitly agree to surrender some of their privacy simply by engaging digitally (it says so in those Terms and Policies everyone’s certainly taken the hours to read through); they’ve agreed to be “tracked” online, or to have their personal information recorded and perhaps even sold to a third party, or to have their “friends” lists identified.  If, like above with the manufacture of the wearable device, the holder of this data is using it to optimize a user experience or improve a product, that seems to be a welcome part of the deal.  In the sports industry, creating better fan experiences, optimizing communications and direct marketing materials, and special offers and contest alerts, are all examples of beneficial outcomes of strategic data analysis.  However, where those alarms will go off is if the holder of the data is selling it to the highest bidder with no regard for the ultimate impact on the individual.

While these are just a few examples of the types of data collected on an average day in sports, they represent only the very tip of the data iceberg.  To date, industry feedback has indicated that maintaining secure data warehousing, implementing robust privacy policies, and using the data for mutually beneficial outcomes are among the best practices an organization can implement.  When working with third parties, whether hardware or software vendors or marketing partners, the obvious answer to “who owns the data?” and what each party is at liberty to do with is, read the contract.

This post has not been tagged.

Share |
PermalinkComments (0)

When Sailors Collectively Bargain

Posted By Megan Morgan, Wednesday, March 1, 2017

In January the defenders and challengers of the oldest trophy in international sporting competition, the Americas Cup, gathered to announce rules framework that would govern not the upcoming race, but the following two competitions, to take place in 2019 and 2021. That THIS competition, of all the sports across the globe, just locked in its rules in any fashion, let alone for multiple challenges before the current Cup has been contested is a complete departure from 165 years of tradition – one that will serve the sport very well.

There may be no single competition which throughout its history, and especially over the past three decades, has benefited from the gamut of technology, adeptly exploited innovations, and applied advanced analytics across the spectrum as well as the Americas Cup teams have.  From the earliest races, boat designers would manipulate the “maximum waterline” regulation by creating hull shapes that would meet the requirement when at rest, but have much longer waterlines when heeled over in the wind.  Such boat design wars took a quantum leap forward in 1987 when Dennis Conner’s team from the San Diego Yacht Club famously entered a catamaran into the competition, as multihulls were not expressly prohibited by the cup trust deed.

To grossly oversimply, the way the Americas Cup works is that the reigning champion sets the rules for the next cup challenge – including the location of the race, the specs of the boats, the dates of the race(s), the terms of the competition, etc.  There is also only one official challenger to the current cup holder, so the terms must be set for the field of challengers to face off and determine who will go forth to race for the cup.  Imagine if, after the 2016 Super Bowl, the Denver Broncos declared that the 2016 regular season would be played seven-on-seven until the playoffs and then it would be back to 11-man teams?

Over the past quarter century Americas Cup boat builders have tapped into every material science advantage to make the boats lighter, faster, and stronger; they have tested every conceivable coating; they have looked at endless designs for sails and now rigid wings.  With the addition of hydrofoils, when the hulls are out of the water, drag is reduced so much that the boats can actually sail faster than the wind that is powering them.  This is not to say that technological advancement has been smooth sailing.  At almost every turn, the innovator has inevitably found himself in court.  The first synthetic hull was challenged, the first multihull was challenged, and the shape and size of virtually every part of the boat was protested by someone at some point over the past 160 years.

As for sailing the boat well, the analysts and tacticians on an Americas Cup team are no different than the analytics departments in any other team sport.  What they analyze is.  They gather all the environmental data for the race course: wind patterns, tide, temperature, etc., and combine all of that with the properties of the boat to generate an attack strategy for the course.  During the 2013 Americas Cup, Oracle Team USA, on the brink of elimination, down 8-1 to Team New Zealand, mounted the greatest comeback in sports history to win eight straight races and win the cup 9-8 in San Francisco Bay.  Part of the reason Team USA was in that hole is that there was a glitch in one of their modelling programs, not allowing for an attack strategy that charted a longer but faster course.  Stu Woo for the Wall Street Journal chronicles the comeback in detail here.

The eleven members of the boat’s crew are as well conditioned and trained as any group of world-class athletes.  Their playing field is a seven ton, 72-foot-long, 46-foot-wide, 131-foot-tall, carbon-fiber and trampoline-like net…that moves, in three dimensions, on water.  Oh, and the eleven people on the team are responsible for moving the boat, and moving it precisely.  Further, the migration from monohulls to catamarans means that grinders aren’t working hard for a minute or two and moving over a couple of feet until the next tack.  Now they cross the 45-foot trampoline while it is moving at speeds up to 40 knots.  They are “putting out high wattage for up to 40 minutes,” according to Paul Cayard, a veteran AC helmsman.  The conditioning needed for an Americas Cup race was compared to “playing a half of rugby union without the contact and running,” in this New York Times examination of what it takes to train to sail Americas Cup.

Among the biggest beneficiaries to the announcement that the Americas Cup will be sailed under a known set of guidelines for the next two additional races are the “small market” challengers.  Under the old system, the holder of the cup in setting the parameters for the next race could (and has) prescribe a boat design that is essentially cost prohibitive under the timeline for anyone but the wealthiest of sponsors to even entertain commissioning.  There is no doubt that the challenger series for both 2019 and 2021 will see additional teams enter for the right to officially challenge for the oldest international sporting trophy in the world, the Americas Cup.  This is better for competitive sailors, better for fans, and better for sponsors and markets who host the challenger series races.


Watch this space later this summer for facts and news from AC35 as Oracle Team USA defends their title on the Great Sound of Bermuda.

This post has not been tagged.

Share |