11 Epilogue

We have covered a lot of ground in the foundations of business strategy for a sports organization, with a focus on working inside a club. The throughline has been using analytics to build strategies that grow ticket sales and revenue and that make business operations run better — and, just as much, on how to think about those problems. The book came in two parts. Chapters 1–4 build the foundation needed to analyze a problem; chapters 5–10 apply that foundation to specific problems and demonstrate techniques for solving them.

1. Chapter 1 introduced how to think about and use data, which is fundamental to business strategy.
2. Chapter 2 introduced the book’s data sets and the R language used throughout. Some programming knowledge is still essential to the work.
3. Chapter 3 covered exploring data — building and interpreting the foundational graphs, and summarizing before modeling.
4. Chapter 4 covered framing projects and some basic project management.
5. Chapter 5 introduced customer segmentation through several methods, and how to use the results.
6. Chapter 6 covered pricing and forecasting.
7. Chapter 7 extended segmentation into lead scoring.
8. Chapter 8 covered promotions and some brand-management ideas.
9. Chapter 9 introduced formal research methods, including hypothesis testing and sampling.
10. Chapter 10 covered operations — simulation and queuing for venue problems.

That is a lot of material, and we have only scratched the surface. At the start of the book we set four goals; it is worth asking whether we met them:

1. To give you an analytically grounded toolbox you can build on
2. To teach you how a sports team’s business works
3. To apply that knowledge to real problems to achieve desired outcomes
4. To build a reference manual of solutions to common problems

Ultimately this all coalesces into strategic thinking. No one is born understanding strategy; it is built on experience. Analytics is tightly coupled to it — it touches every function of a business and helps answer the questions a strategist poses.

11.1 Your analytics toolbox

We worked through a lot of code and a number of concepts. At a high level, you now have tools for:

• Manipulating and summarizing data with code
• Regression
• Implementing machine-learning algorithms
• Formal research methods
• Simulation and its applications
• Building graphics
• Framing problems and projects

We also left a great deal out, deliberately: neural networks and modern AI, more exotic forms of regression, Bayesian methods, and other ensemble techniques. Once you can think a problem through and get your data in the right shape, applying a specific algorithm is usually the easy part. There are hundreds of algorithms, but in our context they mostly do one of three things:

1. Predict a numeric value
2. Predict a class
3. Reveal structure

We skipped tools like convolutional neural networks for computer vision and speech recognition because, in a club, specialists and vendors will do that work better and more efficiently than we can. You might deploy facial recognition for suite entry, natural language processing for service, or a digital twin for spatial analysis of the park — but you will buy those capabilities, not build them. What you need is enough fluency to know when a tool fits and when it does not, so that no one can wow your executives with “AI” that is the wrong tool for the job. Now you know what that means.

11.2 The business of sports

We also spent time on how sports differs from other industries. Outcomes are often outside your control, and you cannot simply transplant business-school techniques because the business is constrained in unusual ways — geographic territories, agency agreements, labor rules, and more. Many leagues are, at heart, trade organizations, and much of our work is a constrained optimization. Some seasons are up and some are down; the job is to navigate them and to think about problems in advance and in the abstract.

Fans are drawn to sports for many reasons, and that should shape your thinking. Clubs face public-relations exposure on many fronts and can be appropriated by politics at the worst moments. The brand is critical, and you are not always in control of it. Protecting it through public relations and community involvement — defensively and offensively — is a real strategic problem we did not cover but that you should understand. How those choices affect sponsorship and ticket sales is rarely knowable in advance. You do the best you can.

11.3 What we did not cover

A number of enabling functions live in IT or at the executive level and fell outside our scope:

• Asset valuation in corporate sponsorship
• Dashboards and other business-intelligence work
• Building ETL pipelines

So did several higher-order strategic problems, often involving legal rights or the fundamentals of a customer or asset class:

• Media rights
• Digital rights
• Political and legal issues
• Growing the top of the funnel (reaching Gen Z and Gen Alpha)
• Business development and growth

11.4 How to learn more

Google it. Seriously. There are countless free resources for learning R or Python — free courses and entire books — and the same is true for machine-learning techniques. One warning: techniques change quickly. Random forests and support vector machines may fade as neural networks get easier to deploy; tools like TensorFlow (tensorflow 2020) accelerated that shift years ago.

If you are a student, take classes and network. Contact someone at a club and ask to work on a school project with them. If you already work in the industry, broaden your horizons — do not get locked into a few narrow projects. The wider your experience, the better you will be at everything you do.

There are also countless books on every corner of analytics. The field is broad enough that you will naturally gravitate toward parts of it — visualization, databases, modeling and statistics. Find what you enjoy, buy a used book on it, and try to reproduce what you read. R, Python, TensorFlow, and much else are free; download them and start on some data. R and Python even ship with thousands of data sets. In R you can list them for a package:

data(package = "ggplot2")

Or see everything available across installed packages:

data(package = .packages(all.available = TRUE))

Find some interesting data and use it. Many packages also have excellent documentation and full vignettes. With a little time and focus, you will be surprised how much you learn.

11.5 What the future holds

Many of the examples in this book are well-understood, solved problems inside and outside sports. We only scratched pricing, for instance; demand-based dynamic pricing has been running in industry for over forty years. It is commoditized — and that is a good thing. Early in the analytics revolution the analyst might have been seen as an unwelcome presence in the boardroom. Now analytics is a necessity, and its influence will keep growing. The fluency to lead an analytics team is becoming a prerequisite for leadership, and working in the field is the best way to earn it.

Over time, more of these functions will be absorbed into CRM platforms, and the skills will keep commoditizing as students flock to analytics degree programs. That is neither good nor bad. I think of it like computer engineering: you do not need a million elite chip designers — you need a small number who are genuinely excellent, supported by many more who apply the tools. Analytics is the same. You do not need an army; you need a few people who are very good at it.

Analytics will also keep reshaping the sports workforce. Chatbots are already absorbing routine sales and service work, and the pattern generalizes: the more formulaic the task, the more exposed it is to automation. That is not unique to sports, and it raises hard questions about work that run well beyond this book.

Media will keep changing fast. Gen Z and Gen Alpha — born after 2000 — will continue to reshape how content is consumed, which affects sports both through media and, more critically, in person. Nobody has the formula yet, but the literature tends to say things like this (Jeff Fromm 2018):

“Start by humanizing your brand. This means giving your brand a personality that consumers can engage with.”

The rules of communication have shifted. To the youngest consumers, interacting with old hardware can seem absurd:

“Pivotals learned to swipe before they could speak. Attempting to swipe the unswipeable — like TV screens or the pages of a magazine — they assumed the image in front of them was broken.”

I will admit I have tried to swipe a magazine page myself. Reaching young people is a major strategic problem, and at the club level it has to be treated as an investment — which is hard, because investment-style thinking does not come naturally to marketing budgets. We have to do it anyway.

Venue operations will keep changing too. New venues are being built deliberately for more convenient, more immersive experiences. I have worried for years that a terrorist attack on a large sporting event would transform ticketing and ingress the way the 2001 attacks permanently changed air travel ²⁹. And the rapid spread of legal gaming and gambling will reshape the in-venue and media experience in ways we are only beginning to see.

Finally, everything is incremental — nothing in this book is truly new. Neural networks have existed for decades. Reviewing the neural-networks chapter of a college textbook from 2001, Decision Support Systems and Intelligent Systems (Efraim Turban 2001), almost nothing is fundamentally different; incremental gains in hardware, software, and mathematics simply pushed the technology further. Its closing section, “The Future of Management Support Systems,” made predictions worth revisiting. On collaboration:

“Groupware technologies for collaboration and communication will become easier to use, more powerful, and less expensive. They will make electronic group support a viable initiative even in small organizations.”

Thank you, COVID-19 — Zoom, Slack, and Teams are now ubiquitous. On language:

“The use of voice technologies and natural language processing will further facilitate the usage of MSS.”

Amazon Alexa, Siri, and the rest have made that true, powered by the same neural networks. And one more:

“Frontline decision support technologies that mostly support CRM will become an integral part of IT in most medium-sized and large corporations.”

This is the most accurate of the three. Salesforce became dominant and keeps expanding by acquisition — Tableau in 2019, Slack in 2020 — and CRM platforms increasingly use NLP for conversational intelligence. These shifts take time: nearly twenty years on, the work is still moving fast.

We are not in chaos, but we face plenty of disruptive and chaotic problems. Littlefinger, in Game of Thrones, called chaos a ladder, and he was right — uncertainty creates opportunity, and the problems clubs face are more complex than ever. If your core business is not efficient, you never get to work on the more rewarding problems. I hope this book gave you a solid sense of how to carry out the basic analytical work that serves the strategy of managers across your organization. Now go find some data.