As a PGA Professional for 16 years and a nearly three-decade veteran of the golf industry, I have dedicated my career to understanding and enhancing the game through coaching, equipment selection, and course management. Throughout this time, I’ve witnessed remarkable advancements, but none have been as transformative as the rise of artificial intelligence (AI) and machine learning. These technologies are changing how we play the game and revolutionizing the entire golfing experience.

At the recent PGA Merchandise Show in Orlando, I had the opportunity to talk with Brent Neville, the CEO of ShotSense Golf and the creator of a remarkable and what I feel will be a game-changing app for golfers.

Brent is a high-tech industry leader with 15 years of experience in engineering, geospatial analysis, and AI-driven analytics. He is also the real deal: a Notre Dame grad with a BS in Mechanical Engineering and an MBA in Finance and Marketing from Northwestern’s Kellogg School of Management. He and his team at ShotSense Golf are revolutionizing on-course decision-making with their free iOS app, which delivers personalized shot strategies for every shot on any course.

Brent and I have had lengthy discussions about AI and LLMs (large language models), how they can help society in general, and their applications in golf. Below is some of what Brent shared with me.

1. AI and Training Data

Artificial Intelligence, particularly large language models (LLMs), depends on the data it learns from. These models work by predicting the next word in a sequence based on patterns they recognize. For example, if a model often sees “cat” followed by “whisker,” it will likely predict “whisker” after “cat” in the future. This approach is effective for predictable data, like the English language or consistent human behaviors, since these patterns tend to repeat.

For instance, a machine learning model can predict dentist appointments by analyzing historical data, such as patient turnout on different days of the week. Google Maps uses similar concepts by examining large amounts of location and traffic data to forecast conditions. However, predictions can become tricky when individual choices deviate from common patterns, like someone opting for a different route.

In sports like golf, the complexity increases significantly. Each golf shot is influenced by many factors, such as course layout, weather conditions, and a player’s unique situation on that particular day. Unlike language, where patterns are more stable, predicting a golf shot is much more challenging because every shot is its own distinct scenario. This is why the ShotSense Golf method focuses on optimization rather than straightforward machine learning—there are just too many changing factors to rely solely on past data for accurate predictions.

2. Brody and Strokes Gained

A great example of how analytical methods have evolved is seen in the progression of statistical models in sports. The book “Moneyball,” released around 2003, changed how baseball teams evaluate players by introducing innovative metrics like WAR (wins above replacement). A decade later, “Astroball” built on this by using advanced computing and a better understanding of data correlations to provide even deeper insights.

In golf, Mark Broadie’s book Every Shot Counts introduced the strokes gained concept. This concept uses average data sets to evaluate performance from different distances, such as saying it takes about 2.82 strokes to reach the green from 100 yards on the fairway. However, this approach misses key factors, such as wind direction or the difficulty of hitting a small green at a challenging course like TPC Sawgrass.

Broad averages were initially necessary because computing limitations prevented detailed analyses of every situation. Now, technology can handle millions of calculations quickly, allowing for more specific and context-driven evaluations. ShotSense Golf exemplifies this by using real-time data to analyze various shot scenarios, considering nuances that traditional strokes gained metrics might overlook—like whether being behind a tree gives a better shot opportunity than being in the rough.

Brent has experience working at an AI/ML startup and Microsoft. He highlights that the biggest challenge in AI is delivering real value. While large language models (LLMs) showcase significant productivity gains, applying them to specific industries, like golf, can be complex.

Simply adding AI to a problem is often ineffective. Defining the use case and determining if AI is genuinely needed is vital. Currently, there’s a trend of applying LLMs to everything, from chatbots for weather updates to rules of the game. However, the key is identifying valuable applications for AI. For instance, LLMs have proven helpful in summarizing golf notes.

In golf, AI can be categorized into two main areas: hitting the ball and strategy.

Hitting the Ball:

• Club Development: AI helps optimize club design using simulations, a practice that’s been around for over a decade.
• Swing Analysis: Many players are interested, but this requires extensive labeled data from successful swings, which is hard to gather.
• Launch Monitors: Use machine learning to analyze data and predict performance.
• Fitness Routines: AI can customize fitness plans by combining ML and LLM approaches.

Strategy:

• Rangefinders: Limited AI, mainly predictive models based on shot data.
• GPS Apps: Primarily for data display, offering little AI advantage.
• Course Management Tools: Often outdated, lacking AI integration.
• Hardware Tracking: Focuses on data collection, with basic analytics provided in apps.

AI is not a one-size-fits-all solution but can drive value in targeted applications. ShotSense Golf, for example, uses machine learning and optimization, similar to how Google Maps functions—efficiency improved through data-driven updates.

Reader, please support me by checking out my three weekly columns on RG.org. On Mondays, I present “The Starter,” which recaps the week gone by in golf. On Thursdays, I give you “Fairway Focus,” which previews the week to come in golf. And on Saturdays, I dish up the “Weekend Fore-cast,” a look at what’s on tap for the upcoming weekend.

Editor’s note: “My Take” will be an ongoing weekly series where Brendon shares his thoughts and opinions on various aspects of the game and industry. These are Brendon’s opinions and do not necessarily reflect those of GolfWRX, its staff, and its affiliates.