Shooting 59: Big Clubs or Big Data Analytics ?

Jeremy Stierwalt big clubs

By Kyle Swanson
Solution Director, SAP Analytics 
Optimal Solutions an NTT Data Company

Any fan, regardless of the sport, supports their team in hopes of catching one moment of greatness.  We go to games for the no-hitters, the alley-oops, and the hole-in-one moments where regular people earn Hero status in the blink of an eye.  However the moments that make it to SportsCenter Top 10 are rarely the norm and shed light on what countless hours of training can produce for a split second in time.  However, those moments are literally worth millions of dollars.  Diving into the complexity of achieving these moments of greatness is an interesting subject and has become a requirement among professional sports organizations.

The Setup

I find the role of analytics in sports particular interesting because I think about how the Men’s Volleyball Team I played for could have benefitted from this technology (yes that is my real hair!).  When focusing on volleyball there are numerous statistical metrics that are recorded for each player as well as their position.  For instance, I was a Middle Blocker so as a rule I never was a passer (aka Outside Hitter).  My responsibilities were hitting and blocking so the analysis on my success was based on limited information.  However the Outside Hitter, who also triples as a hitter and a blocker, will produce data for all three responsibilities.

Kyle Swanson: D1 Volleyball Player, Solution Director @ Optimal

Kyle Swanson: D1 Volleyball Player, Solution Director @ Optimal

Using this basic information, a Middle Blocker and an Outside Hitter cannot be rated on a similar performance scale.  Even if we try to simplify it to just the hitting statistic, Outside Hitters traditionally get twice as many hitting attempts as a Middle Blocker so relating the two is still not a cut and dry activity.  Lastly we didn’t even consider the performance of each player against a particular defense.  Some teams have strong blocking while others rely more heavily on their floor defense.  To get a good view of individual performance contribution to a victory you will have to go very broad and very deep into the data.

Another reason why I find analytics for sports so interesting is because there will always be a human element to consider.  Take sports betting for instance.  Gamblers will evaluate the teams, the match ups, injuries, etc.  They make bets on how someone like LeBron James, who at the time of this post has a broken nose, will be able to perform against the Knicks tonight.  Some things to consider:  Are the bettors looking at the history of his performances with a broken nose?  Are they evaluating what having a broken nose does to his endurance? Have his defensive matchup(s) been considered? Lastly, what are his opponents’ strengths/weaknesses and how do they relate to LeBron’s current condition?  That is a high level overview of the physical attributes for ONE PLAYER.  Assuming you rotate 10 players per game that is a serious set of analysis.

So what does it take to get through all the data and provide a roadmap for the perfect game?  It might not be possible.

The Spike

Take golfs coveted hole in one.  Technically speaking, the perfect golf score is 18.  Is that an attainable score by any human today? No.  Can we get a player close to the 60’s and 59’s that are currently so rare? I think we can.  Consider for a moment some of the constant factors that are addressed in a round of golf.

  • Total Yardage / Pin Position
  • Tee Time
  • Course Condition (Narrow Fairways, Deep Rough, etc.)
  • Equipment*
  • Weather Conditions and Patterns

Before the official start of each tournament players typically have an opportunity for practice rounds to test the current course conditions and formulate a game plan with their caddy and coach.  Each ball position, swing and result is a data point that can be captured and analyzed.  In addition you can relate the results of each action to the constants listed above.  The analysis is not quite as deep as a basketball scenario but to help build a foundation for how Big Data and Analytics can flourish in sports, let’s stay away from intricate situations.

Cue Analytics!

The ability for Big Data and Analytics to play a crucial role in the outcome of a particular shot, hole or round is a very real possibility.  Unlike my previous volleyball example, a golfer’s performance can be easily rated because they are contributing 100% of the activity.  Golfers also are not directly affected by physical impairments as it relates to their competition, as opposed to the Lebron James situation where the physical presence of an opponent can drastically change an outcome.  Yes golfers compete against one another for prizes, but let’s stick to the perspective that each golfer is competing against the course itself.  By analyzing performance against the constant, players will theoretically be able to collect enough data about their current situation and desired outcome to execute the perfect shot each time.

Additionally, players can begin to analyze their own performance based on non-constant factors. For instance the countless hours of training can net results geared towards swing tendencies by current weather conditions.  If it is raining a player may subconsciously change his/her swing pattern without realizing it but analytics can and will disseminate that information for them.  There is also an opportunity for players to evaluate how their competition has played certain holes.  A #10 golfer may want to review how the #9 golfer moved through the course each day (both currently and in past events) to identify strengths or weaknesses in course management.  Gaining a strategic edge over a close competitor could be the difference of a couple hundred thousand dollars in prize money.

Golf is a great example of how analytics can fuel a player’s performance and by looking at a simple statistic we can see glimpses of the analytics use cases.  The lowest round of golf ever recorded in PGA history is a 59 and there are only 6 players to ever do it.  They are:

  • Al Geiberger – 1977 Memphis Classic
  • Chip Beck – 1991 Las Vegas Invitational
  • David Duval – 1999 Bob Hope Chrysler Classic
  • Paul Goydos – 2010 John Deere Classic
  • Stuart Appleby – 2010 Greenbrier Classic
  • Jim Furyk – 2013 BMW Championship

Notice anything?  Since the first round of 59 was recorded 37 years ago, 50% of 59’s were recorded in the last 4 years.  From my perspective this is a big nod to the benefit of using Analytics in the sports world.

Being able to actively suppress weaknesses and capitalize on strengths for the best possible outcome is exactly what Analytics is designed to do for any industry.  Similar to a business transaction, each move an athlete makes is associated with a risk. The goal is to minimize risk and maximize the opportunity for short and long term gains.  When it comes to actually leveraging analytics for sports, the human mind will always be on the cutting edge of the next set of information that needs dissecting.  I don’t think technology will ever be able to tell us how LeBron can beat the Knicks defender with absolute certainty and as an athlete I am OK with that.  Sports would be boring if we could predict outcomes correctly every time.  Plus, we wouldn’t get to relive moments like these:

I dare you to not watch the whole thing.

*Advances in modern technology have given golfers the ability to attack golf courses like never before.  Despite the every changing technology, a golfers equipment during a round of golf is a constant because a player must choose the equipment he is using that day and is not allowed to change it once the round begins.

Leave a Reply