For Beer Lovers: Analyzing the Great American Beer Festival’s Results
Lindsey Schubert
11/15/2020
Introduction
Craft breweries have been all of the rage in the United States in the past 10 or so years. In 2019, pre-COVID, there were over 8,000 craft breweries open in the U.S. Per the Brewers Association, a craft brewery is small (less than 6 million barrels of beer produced annually) and independent (less than 1/4 of the brewery is owned by an “alcoholic business industry member).
The goal of this analysis is to analyze the results from the Great American Beer Festival (GABF). This festival, a long-running competition since 1983 is a beer competition. The GABF’s goal is “to identify the three beers that best represent each beer-style category as described and adopted by the Great American Beer Festival.” Breweries are awarded gold, silver, or bronze - seen as “the most coveted in the industry and heralded by the winning brewers in their national advertising.”
The results from this specific festival were chosen for analysis as a result of the event being held in such high regard. This will be a case study of beer trends in the United States.
Methods
I downloaded the data set from the 10/20/2020 TidyTuesday event package.
The packages used in this tidytext, ggplot2, rvest tidyverse, dplyr, ggthemes, tidyverse and tidytuesdayR.
State and City Analysis - All Time
I wanted to see what states had the most awarded beer in the entire history of the competition to see if there was a correlation between certain U.S. regions and medal-winning beer.
States
beer_data %>%
group_by(state) %>%
count(sort=TRUE) %>%
filter(n > 120) %>%
ggplot(aes(reorder(state, n),n)) + geom_col(fill="#c99c0e") + coord_flip() + labs(title ="Most Awarded States", x = "State", y = "Amount of Medals") + theme_light()
California has won a staggering 957 medals, Colorado has won 653 medals and Oregon has won 327 medals. Here is a interactive map that allows you to see the medal count in each state. This was created using Tableau.
Cities
beer_data %>%
group_by(city) %>%
count(sort=TRUE) %>%
filter(n > 50) %>%
ggplot(aes(reorder(city, n),n)) + geom_col(fill="#509aff") + coord_flip() + labs(title ="Most Awarded Cities", x = "City", y = "Amount of Medals") + theme_light()
I filtered out cities who contained breweries that have won over 50 medals. Surprisingly, a California city ranks 5th. Denver has the most medals in any U.S. city, with Portland and Milwaukee close behind.
Here is another interactive graph that shows the cities with over 50 medals. The graph is shown with a terrain layer as I wanted to see if beers brewed in mountain ranges had more awards.
Cities with 50 Medals or More
This map shows that most of these beers were produced in breweries along the coastline and in mountain ranges. This map was created using Tableau.
Mountains are crucial in the freshwater world, as they contain some of the United States’ freshest water.
Water quality plays a large part in beer production, as it can make or break a good batch. Water “affects beer in 3 ways”: wort (the liquid extracted from mash) flavor, bitterness of the beer and if contamination is present, the beer can taste funky. Brewers want to be in places with fresh water, so it makes sense that highly awarded cities are near mountains. Colorado, especially Denver and Boulder, are great examples.
Types of Beer - Classic Irish-Style Dry Stout Case Study
Next, the most-awarded types of beer were analyzed. Beer types that won over 40 medals were graphed. The “Classic Irish-Style Dry Stout”, such as Starr Hill Brewing’s Dark Starr Stout, a two-time gold medal winning Dry Stout, has won over 60 medals. Guinness Beer is a famous example of an Irish-Style Dry Stout.
beer_data %>%
group_by(category) %>%
count(sort=TRUE) %>%
filter(n > 40) %>%
ggplot(aes(reorder(category, n),n)) + geom_col(fill="#4b2600") + coord_flip() + labs(title ="Most Awarded Type of Beers", x = "Beer Type", y = "Amount of Medals") + theme_light()
I wanted to see where the most-medaled beer was being brewed to see if there was any correlation with the states with the most medals.
This type of beer is only brewed in 19 states. California is home to many breweries who have won 23 medals for their Classic Irish-Style Dry Stouts.
Gold Medals
The highest award that a beer can receive is the gold medal. Per the GABF’s website, a gold-medal winner is "A world-class beer that accurately exemplifies the specified style, displaying the proper balance of taste, aroma and appearance.
The larger dataset was filtered out to just include beers who have won gold medals.
beer_data %>%
filter(beer_data$medal %in% "Gold") -> GoldMEdalsI filtered out just the category of beers to see which type has won the most. Robust Porter beers have won 23 medals, Imperial Stout beers have won 22 medals and Classic Irish-Style Dry Stout, the winner of the most cumulative (bronze, silver and gold) medals, received 21 medals.
Category
GoldMEdals %>%
group_by(category) %>%
count(sort=TRUE)## # A tibble: 454 x 2
## # Groups: category [454]
## category n
## <chr> <int>
## 1 Robust Porter 23
## 2 Imperial Stout 22
## 3 Classic Irish-Style Dry Stout 21
## 4 German-Style Pilsener 21
## 5 American-Style Pale Ale 19
## 6 American-Style Amber Lager 18
## 7 American-Style Brown Ale 18
## 8 Bock 18
## 9 German-Style Wheat Ale 18
## 10 American-Style India Pale Ale 17
## # … with 444 more rowsFollowing similar methods, I wanted to look into the states who have the most gold medal winning beers. This map was created using Tableau.
Unsurprisingly, California, Colorado and Oregon all have largest amount of gold medals.
Cities
Also, I wanted to analyze the cities with the most gold medal beers to see if they correlated with the states.
GoldMEdals %>%
group_by(city) %>%
count(sort=TRUE)## # A tibble: 475 x 2
## # Groups: city [475]
## city n
## <chr> <int>
## 1 Portland 47
## 2 Denver 45
## 3 Seattle 35
## 4 San Diego 33
## 5 Milwaukee 32
## 6 San Antonio 28
## 7 Fort Collins 27
## 8 Chicago 25
## 9 Austin 24
## 10 Bend 24
## # … with 465 more rows2020
The last part of the dataset that I wanted to analyze was the trends in 2020.
Methods
beer_data %>%
filter(beer_data$year %in% "2020") -> twentytwentyStates
twentytwenty %>%
group_by(state) %>%
count(sort=TRUE) %>%
filter(n > 10) %>%
ggplot(aes(reorder(state, n),n)) + geom_col(fill="#c99c0e") + coord_flip() + labs(title ="Most Awarded States in 2020", x = "State", y = "Amount of Medals") + theme_light()
Cities
twentytwenty %>%
group_by(city) %>%
count(sort=TRUE) %>%
filter(n > 5) %>%
ggplot(aes(reorder(city, n),n)) + geom_col(fill="#509aff") + coord_flip() + labs(title ="Most Awarded Cities in 2020", x = "City", y = "Amount of Medals") + theme_light()
1990
To do a quick comparison - I did the same methods and looked at data from 1990, 20 years ago. 1990 was a big year for beer. According to a 1991 New York Times article, 1990 saw a 2% increase in beer shipments - 193.2 million barrels.
beer_data %>%
filter(beer_data$year %in% "1990") -> nineteenninetyStates
nineteenninety %>%
group_by(state) %>%
count(sort=TRUE) %>%
filter(n > 5) %>%
ggplot(aes(reorder(state, n),n)) + geom_col(fill="#c99c0e") + coord_flip() + labs(title ="Most Awarded States in 1990", x = "State", y = "Amount of Medals") + theme_light()
California still held its own in 1990, with 20 medals. Surprisingly enough, Wisconsin and Washington held second and third place in 1990. Wisconsin is home to Miller Brewing Company, who produces Miller Lite. This brewery saw its peak in 1990 - “shipping 19 million barrels.”
Cities
nineteenninety %>%
group_by(city) %>%
count(sort=TRUE) %>%
filter(n > 3) %>%
ggplot(aes(reorder(city, n),n)) + geom_col(fill="#509aff") + coord_flip() + labs(title ="Most Awarded Cities in 1990", x = "State", y = "Amount of Medals") + theme_light()
Seattle, San Antonio and Milwaukee were the most-awarded cities in 1990. Milwaukee is home to the Miller Brewing Company, who had a great year in 1990.
Professional Reviews vs User Reviews - A Case Study
The Beer Mapping Project Website that uses Google Maps API with “custom developed tools” to show breweries across the United States. There is a running, comprehensive list of Google reviews that is compiled in order to make “The Beer Mapping Project’s highest scored locations”. These results are all based on consumer preference and reviews. While this website is by no means a comprehensive list of public opinion of breweries, I thought it would be interesting to compare some of the data to see if any trends were apparent.
Methods
I converted the table from the website into a table.
beermapping <- read_html("https://beermapping.com/highestscores/")
tables <- html_nodes(beermapping, "table")
beermapping_tables <- beermapping %>%
html_nodes("table") %>%
.[1] %>%
html_table(fill = TRUE)
as.data.frame(beermapping_tables) -> beermaps
beermaps %>% write.table('beermaps.csv')These rankings are based off of Google Review scores, on a 5-point scale. The Beermapping website states, “Reviews give context to a location’s value or worth. In the end a review is simply a personal opinion, but when several reviews are placed beside each other, they might be able to give a broader view on that location’s value as a destination on the maps.”
Most Reviewed Cities
beermaps %>%
group_by(City..State) %>%
count(sort=TRUE)## # A tibble: 75 x 2
## # Groups: City..State [75]
## City..State n
## <chr> <int>
## 1 Portland, OR 8
## 2 Chicago, IL 4
## 3 Baltimore, MD 3
## 4 Boulder, CO 3
## 5 Denver, CO 3
## 6 Grand Rapids, MI 3
## 7 Austin, TX 2
## 8 Cleveland, OH 2
## 9 Falls Church, VA 2
## 10 San Diego, CA 2
## # … with 65 more rowsInterestingly enough, breweries in Portland and Chicago had the most reviews. Portland seems to be a prominent city, along with its state of Oregon, in this analysis.
Highest Ranked Breweries
beermaps %>%
group_by(Total.Score) %>%
group_by(City..State) %>%
count(Total.Score, sort=TRUE) %>%
filter(Total.Score > 96) %>%
knitr::kable()| City..State | Total.Score | n |
|---|---|---|
| Aurora, CO | 97.14 | 1 |
| Bellaire, MI | 96.67 | 1 |
| Blanco, TX | 96.25 | 1 |
| Deming, WA | 96.11 | 1 |
| Philadelphia, PA | 96.67 | 1 |
| Placentia, CA | 96.33 | 1 |
| Roseland, VA | 97.00 | 1 |
| San Marcos, CA | 97.00 | 1 |
| Washington, DC | 97.67 | 1 |
I filtered out breweries with a score higher than 96. The top brewery in this dataset is in Washington, DC - a city not mentioned in the top lists in the GABF dataset. A Colorado brewery comes in second and a California one comes in third.
Conclusion
If you are looking for a good beer, head to the West Coast. California, Oregon and Colorado have been awarded time and time again for the beer that their breweries are producing.
This graph depicts every brewery who has won a medal in since the beginning of the data collection in 1987. It is separated by medal type and decade. This is helpful if you’re ever in a new brewery and are curious about what beers are award-winning!