Top 5 articles to read before the big game

Whether you’re a Chiefs or 49ers fan, simply a football fan or just here for the buffalo wings, prepare yourself for the Super Bowl by skimming the following five articles published by INFORMS (and a bonus podcast episode!). Show up with some interesting and fun football statistics and the halftime show won’t be the only thing to wow the people in your living room. [The following summaries were based on ChatGPT’s prompt of rewriting the abstracts so a high schooler* could understand the concepts, and the results were approved by the authors.]

1. Scoring a Touchdown with Variable Pricing: Evidence from a Quasi-Experiment in the NFL Ticket Markets

Even though sports teams can adjust ticket prices based on the popularity of a game, many teams have yet to fully adopt this strategy, known as variable pricing. This article examines the implications of teams switching to variable pricing, utilizing data from the National Football League.

Using a combination of statistical methods, we initially show that the implementation of variable pricing increases ticket sales by 1.59% per game. We then investigate the reasons behind this positive impact. We discover that lower prices for less popular games are well received by customers, and customers surprisingly don’t mind paying more for the most popular games.

We also examine how ticket resale prices change after the introduction of variable pricing. We observe that prices for popular games increase in the resale market, which explains why people are willing to pay higher prices for these games in the primary market.

Finally, we show that this pricing strategy is most effective in cities with lower income levels and higher income diversity, and having more price tiers for different types of games helps teams sell more tickets.

2. Does Losing Lead to Winning? An Empirical Analysis for Four Sports

Berger and Pope (2011) discovered that in basketball games, teams that are slightly behind have a better chance of winning. We wanted to see if this applies to other sports, such as Australian football, American football and rugby. Surprisingly, we didn’t find any evidence of this pattern in those sports.

When we looked back at basketball, we found that the trend Berger and Pope found only holds true for certain NBA games from the time they studied. It doesn’t seem to apply to NBA games outside of that time period, college basketball games or games from the Women’s National Basketball Association.

Even after combining data from all these sports, we didn’t find enough evidence to say that being slightly behind actually increases a team’s chances of winning. In fact, if there is any effect, it's likely very small.

3. Points Gained in Football: Using Markov Process-Based Value Functions to Assess Team Performance

This paper explores a new way to evaluate performance in professional American football. Instead of just looking at scores, we use a method called dynamic programming to figure out how valuable each play is over the game, even if a play doesn’t directly score points. In particular, we’ve been able to give a precise meaning to the value of a play – basically, how many points a team gains or loses compared to what’s normal. There is a similar idea in golf, strokes gained, which measures how much better (or worse) a golfer hits a shot compared to an average player in the same situation.

To prove our approach, we use data from more than 160,000 plays in the National Football League from 2013 to 2016. From this, we create a performance metric called points gained, which helps us determine how well a team is doing on any given play compared to what’s expected. We find interesting patterns, including how passing plays usually create the most value, while running plays often lead to losing value. Short passes are crucial for successful teams. We also look at how teams perform in different situations, such as different downs, quarters and field positions. 

To make it more relatable, we end with a case study of the 2019 Super Bowl and illustrate how different plays created value for the teams.

4. Converting NFL Point Spreads into Probabilities: A Case Study for Teaching Business Analytics

In this study, students explore how the predicted difference in points between two football teams (point spreads) relates to the probability of winning a game, using data from the National Football League (NFL). Even if students don't know much about football, they can still understand the methods and ideas used in this study.

Students begin by creating models using Excel, a program many students are familiar with, and will use functions such as PivotTables, trendlines and Solver. The case study is flexible and can be adjusted to fit different teaching styles and student needs and interests. It introduces the concept that by analyzing past point spreads and game outcomes, we can predict the likelihood of a team winning based on how many points they’re favored to win by.

The study guide provides clear instructions for both teachers and students, making it easy to follow along. Overall, this case study is a useful tool for learning

5. Super Bowl Ads

We looked at how ads during the Super Bowl affect sales of products like beer and soda. The Super Bowl is a huge deal for advertisers because many people watch it and pay attention to the ads. We used data on how many people watched the Super Bowl and how much beer and soda they bought before and after the game. Here’s what we found:

1. People tend to buy certain brands of beer and soda to drink while watching the game, presumably for Super Bowl parties. But whether a brand is going to advertise during the Super Bowl doesn’t really affect this.
2. However, after the Super Bowl, ads during the game seem to have a bigger impact on product sales when there are other sports events happening. This suggests that Super Bowl ads might make people associate the brands with sports.
3. We then looked at college basketball tournament viewership and found that brands advertised during the Super Bowl seem to do better when it comes to selling products during big sports events like these tournaments.

Basically, our research shows that advertising during the Super Bowl can make a difference in why people choose certain brands, especially when it comes to other sporting events.

6. Bonus podcast! The countdown to the Super Bowl is on! Let’s take a look at the role of analytics.

2023 Super Bowl flashback! At the time of this recording, NFL playoffs were well underway with the Super Bowl just a couple of weeks away. There were four teams in the playoffs – Bengals, Chiefs, Eagles and 49ers – with the next round coming up. Guest Walt DeGrange, director of analytics capabilities at CANA, discusses the continually evolving role of analytics in football.

And if artificial intelligence is your thing, you’ll want to pay attention to the Super Bowl ads as well.

*If you are a high schooler, may we suggest checking out operations research or analytics as a potential college major? We have some cool stories about how math is saving the world.