Click to enlarge — it’s worth it
The remarkable chart shown above tracks each NFL team’s jersey color over every game for the past 21 seasons. For each team shown, each horizontal row has 16 cells, representing that team’s 16 games in a given season, and each vertical column has 21 cells, representing every season from 1999 through 2019. So the top-left cell is the team’s first game in 1999, and the bottom-right cell is the team’s final game of last season.
It’s an engrossing and very satisfying example of data visualization. As you can see at the bottom, it was designed by a guy named Anthony Reinhard, who created it by using game-by-game jersey info from the mighty Gridiron Uniform Database. I’ll have more to say about Reinhard in a bit, but first let’s explore this wonderful graphic he’s created. Here are a bunch of things that jumped out at me:
1. If you had asked me which team had the fewest colored-jersey games over the past 21 seasons, I would have said, “Duh, the Cowboys.” But the correct answer, according to the data, is the Dolphins, who played 63 colored-jersey games over the past 21 seasons, compared to the Cowboys’ 67:
2. As you can see in the Dolphins’ chart, the team’s different shades of aqua over the years were taken into account — a nice detail. This makes it easy to see how certain teams have changed their primary team colors over the years. Many fans, for example, probably think that the Lions just wear Honolulu blue, period. But the chart indicates that their shade of blue changed quite a bit over the past 21 seasons:
3. You can see certain historical protocols and events play out in the color distribution. For example, remember how Washington used to routinely wear white at home, just like the Cowboys? Their chart makes it easy to see when they did that, and when they stopped:
4. Similarly, what’s that one yellow cell in the Eagles’ chart? It’s from that one time they wore the 1934 throwbacks:
5. And you can also see how the Bills abandoned their classic royal blue for navy for nine years, and then switched back to royal:
6. While most teams have at least two different non-white jerseys represented (the Broncos have four: orange, navy, yellow, and brown), the charts for some teams, like the Steelers and Kansas City, feature nothing but the team’s primary color and white:
I could go on, but you get the idea. I’m sure you can find your own examples of interesting things lurking in the chart. It’s all very satisfying to pore over!
The chart does have its limitations. Some teams have two different white jerseys, for example, or two different blue jerseys, or whatever, and the chart doesn’t distinguish between them. Similarly, there’s no information on what the opposing team was wearing, who won the game, or any number of other things that might be fun to explore. But within the limited framework of what the chart sets out to do, it succeeds spectacularly.
Reinhard posted the chart on his Twitter feed two Saturdays ago, which is how I became aware of it. He describes himself on his Twitter profile as a “visualizer of football data,” which sounded intriguing, so I began scrolling through his feed, where I found all sorts of fun graphics (mostly created by him, but also some retweets of other people’s work). None of them were about uniforms, but many of them used team logos, team colors, and so on, so it all felt connected to athletics aesthetics — maybe not uni-related, but uni-adjacent. Here are some of the ones I particularly enjoyed:
— Anthony Reinhard (@reinhurdler) September 2, 2020
I thought this would be a fun one for a Saturday: Which quarterbacks most often make the tackle after they throw an interception?
Patiently waiting for Andrew Luck to return next season as an inside linebacker.
— Anthony Reinhard (@reinhurdler) August 15, 2020
I plotted WR heights by team! Rodgers has the tallest WRs to work with while Josh Allen has the shortest. Sorted by median. These are pretty interesting to look at and think about. pic.twitter.com/kMP5a3pOiR
— PykingsAnalytics (@mnpykings) August 16, 2020
Which NFL teams are willing to be social on twitter?
– 23 teams follow every other team
– 3 teams (PHI, BAL, IND) do not follow ANY other teams
– @AtlantaFalcons are following every team that follows them back and none that don't! pic.twitter.com/3CGuiepqVa
— Anthony Reinhard (@reinhurdler) August 30, 2020
After exploring Reinhard’s feed, I wanted to learn more, so I contacted him and asked if he’d be willing to do a phone interview. He readily agreed, and we recently had a really good talk. Here’s a transcript of that conversation, edited and abridged for clarity:
Uni Watch: Let’s start with some basic information about you. How old are you, where do you live, and what do you do for a living?
Anthony Reinhard [shown at right with his wife, Gina, on the day of their engagement two years ago; click to enlarge]: I am 27 years old. I live in Columbus, Ohio. And for a living, I am a data analyst at a midsize insurance company.
UW: What does a person study to get into data analysis? Like what what was your major in college?
AR: I have a Bachelor of Science in actuarial science, which is basically just statistics and math that has to do with insurance.
UW: How and when did you start doing football data visualizations?
AR: I probably started doing this a little more seriously about eight months ago, close to the end of last football season. I’ve followed a few people on Twitter who talk a lot about advanced football metrics and stuff that is, you know, newer over the last six years or so. And one of the best ways to share information about these advanced metrics, which can be kind of complicated, is to make graphs about some of the information. I’ve worked with data a lot and in my professional life, so it seemed like an easy segue. And I’ve always loved football.
UW: Looking through your Twitter feed, it appears that you focus almost exclusively on the NFL — no college football, no other sports. Is that right?
AR: I wouldn’t preclude myself from getting into any other sport in the future, but right now the data for the NFL is so accessible, and the community seems to be right there for me. So it’s definitely the thing I’m focused on most right now.
UW: It’s so interesting that you use that word community, because that was my next question: I get the impression from your feed that there is a community of data visualizers out there. How big would you say that community is? And where do you all congregate online? Are there particular message boards or websites?
AR: I would say it’s small enough so that any person with an interest in visualization or football, I think there’s a place for them if they’re willing to chip in their own stuff. This is going to be a little bit in the weeds, but most of the stuff that I do is in a statistical package called R, or a statistical program called R…
UW: Just the letter R?
AR: Yeah. It’s been around for a while, and it’s open-source. So it’s not something that’s owned, and it’s not for sale. You can download it on your own computer and do your own analysis. A lot of companies use it just because it’s so accessible, and a lot of academics use it because it’s available and you can use it across different universities or whatever. But within R you can download other programs that people have built inside this statistical program. So one of the ones that has been built kind of recently is called nflfastR. Basically, it’s a pipeline to get NFL play-by-play data very easily. So you can download a big table of data from every play, all the way back to 1999. And I would say that the folks who use that data and talk about it a lot are really at the nexus of the community.
Of course, there are people who work for ESPN or Pro Football Focus, who have access to other data. But the online Twitter community is mostly people who work with that nflfastR data.
UW: Are we talking dozens of people in this community? Hundreds of people? Thousands?
AR: Closer to hundreds, I would say.
UW: One thing I noticed while looking through your feed is that you often refer to data being “scraped.” Can you tell me more about that?
AR: I’m not an expert at web scraping, but basically it means going to a website, downloading parts of the web page, and then storing them in a secondary file so you can access them later.
For the graphic about the jersey colors, for example, I visited every web page on the Gridiron Uniform Database and downloaded the image of the uniform for each team in each game. And then once I downloaded the image, I could look at individual pixels in that image, and then those were the ones that gave me the ideas for the the colors of the uniform. Once I had all that information, I could make it into a graph.
UW: When you say you visited every page and downloaded the uniform images, did you do that, like, physically, one at a time? Or was there a global command?
AR: Yeah, that’s the magic of web scraping. I can write some code in R that will visit every web page for me — and quickly, too.
UW: I never really thought about this before, but now that we’re talking, it occurs to me that in order to create effective data-driven graphics, you need several distinct skills. First, you need to know what data is available and how to access it. Then you need to have ideas about how to harness or leverage that data in order to provide meaningful information — in other words, you have to know which questions to ask and how you can use data to answer them. And then you need good design skills in order to present that information in visually accessible and engaging ways. Which of those would you say is most important, and which one do you think is your biggest strength?
AR: I think you put it perfectly right there. I would say that to make something like what I made, in this case, you would need all three, probably in pretty equal doses. I think that probably the second one is ultimately the most important — knowing which questions to ask and then being able to, you know, answer those questions in interesting ways.
The third step — how you’re going to show it — is, I think, probably the most underrated of the three. I like to say that the information that you have is only as valuable as you can present it. If you have a really interesting finding but you can’t communicate it, whether that’s in a graph or even in words, then it isn’t worth much.
UW: In the data visualization community, are there people who are known as better number-crunchers, and other people who are known as better designers?
AR: Definitely. In my case, I’m probably a better designer than somebody who is a great analyst, and I think some of that comes with time. I’ve only been really looking at a lot of this data for eight months or so, while other folks have been doing this for a couple of years. For someone like me, I think a lot of people are interested in my designs, sometimes more than the information, which I’m okay with. Other folks might say they’d rather have a really interesting concept than an interesting design. So I think it depends. I think that there’s kind of a spectrum that folks land on in the football data community.
UW: You mentioned that you’ve only been doing this since the end of last football season. Now that the 2020 season is about to start, do you think watching a game will be a different experience for you? Will you be sort of processing what you see in a different way?
AR: Absolutely. One of the things that you quickly realize about the basic beliefs of the community is that a lot of folks think that teams run the ball too much, which is something I really hadn’t thought about too much in the past. And a lot of teams punt too often — probably they can be more aggressive on fourth down, and also go for it on two-point conversions a little more often. So I think those are things I’ll be looking out for. It’s very different once you’ve thought critically about some of the stuff that teams are doing in games.
I’ve always been a Cleveland Browns fan, so I think I’m very lucky that the Browns are, at this moment, one of the more analytically driven teams in the league. So I think I’ll be more comfortable watching Browns games than I would be if I were maybe a fan of the Seattle Seahawks, who are kind of working against some of the larger trends in the league — which I’m sure is very frustrating for a lot of the data-driven Seahawks fans that I know.
UW: You’re still new at this, but would you be interested in doing this type of thing for a living, like if a team wanted to hire you?
AR: That’s a tough question. I’m really happy with my current job, so it would be tough for me to imagine, you know, a whole different world like that. This is a hobby, I think — it’s a great getaway from what I normally do and also allows me to leverage skills that I can still use in my day job. If the right opportunity came along, I guess I would consider it, but I definitely see it as a hobby for now.
UW: OK, now let’s talk about the graphic you made about NFL regular season jersey colors. Is this the first uniform-related graphic you’ve created?
AR: Yes, I would say this first one that deals explicitly with uniforms.
UW: Why did you choose to document 21 seasons, instead of a round number like 20? Is it because of that data package you mentioned earlier, which goes back to 1999?
AR: Yeah, it is. We have really good data for those last 21 seasons. The Gridiron Uniforms site goes back to, I think, like the 1920s, so I certainly could have gotten more data, but most of the data that’s available for other metrics is from 1999 on, so I thought that would be a good year to stop.
UW: You mentioned before how you scraped the data — see, I can say it too! — and then got pixel values from the Gridiron Uniforms images. So is that how you were able to include all the slight color variations from year to year?
UW: So, basically, you didn’t track the color changes — the Gridiron Uniforms site did.
AR: Yeah, they did all the hard work.
UW: Were there any patterns or revelations that surprised you? For example, I would have assumed that the Cowboys wore white more often than any other team during this period. But that honor actually goes to the Dolphins!
AR: Yeah, I guess I never really thought about that either, because I think of the Dolphins having that iconic aqua or teal color — that’s kind of how I think of them — but I guess they wear white a lot more often than I imagined.
UW: Did you happen to tabulate which teams wore white least frequently?
AR: No, I didn’t. That’s an interesting thought, though. I would have to think it would be a team in the NFC East, because of the Cowboys.
UW: But not Washington because there was a period where they also wore white at home!
AR: Oh, yeah. So then I’d have to think it would definitely be the Eagles or the Giants.
UW: From a visual perspective, do you find it more satisfying to see the charts for teams that just have white plus one color, like the Steelers and Chiefs? Or do you like the ones that have additional jersey colors scattered into the mix? I use the word satisfying because that’s the feeling I get from organizing information this way.
AR: I would say that the most aesthetically pleasing one for me was Tampa Bay, where they’re all almost the same shade of red and then their whites and then they have like four of the Creamsicle ones, which is a great color.
I also like the Seahawks, how they’ve kind of gone through waves with their three different uniform colors and then they have — I mean, I don’t like the green ones that they wear, but I think it looks good when there’s just like four specks of the tennis ball color.
UW: Do you think you might be doing any other uniform-related graphics now that you’ve done this one? And for that matter, are you, like, a uniform guy to begin with? Were you aware of my work, or aware of Uni Watch, before I got in touch with you?
AR: I had heard of Uni Watch before, but I wouldn’t say I have a super-big interest in uniforms. I definitely have an interest in design and color, so that relates to being interested in uniforms. Now that I have the data [from Gridiron Uniforms], I could definitely see myself doing a few more things with this in the future.
And there we are. I should add that Anthony was an absolute pleasure to talk to — really nice guy, and obviously really smart as well. Big thanks to him for sharing his time and expertise.
(Special thanks to Greg Boone, who was the first to bring Anthony’s jersey chart to my attention.)
Click to enlarge
Too good for the Ticker: Hmmm, what’s going on in the photo shown above? That’s a shot from 1973 edition of the Baseball Hall of Fame Game, an annual exhibition game that took place in Cooperstown from 1940 through 2008. The ’73 game featured the Pirates vs. the Rangers (a routine interleague matchup nowadays, but a major novelty back in the day), and at least one Rangers player — outfielder Tom Grieve, who would later become the team’s GM — wore a Pirates batting helmet! It’s not clear, at least to me, if Grieve’s misplaced his regular helmet or if the teams were just having a bit of fun.
Moreover, although it’s a little hard to tell in that photo, both teams wore their road greys, because they were both in the middle of road trips when they detoured to Cooperstown for the game. Here’s a photo of the two managers — Texas’s Whitey Herzog and Pittsburgh’s Bill Virdon — where the grey tones are a bit more apparent:
(My thanks to @indywestie and @PolyesterUnis for this one.)
Having been a teacher for 11 years, I have used a wide variety of sports to create connections and build relationships with my students. This video explains what I'm doing this year. @UniWatch @sportslogosnet @Extratime @MiamiDolphins @TimbersFC pic.twitter.com/BSbShZslYh
— Trevor Williams (@TrevorTeaches) September 15, 2020
A teacher who deserves an A+: Longtime Uni Watch reader Trevor Williams is a third grade match/science teacher in Texas. He also has what has to be one of America’s most uni-themed classrooms. To get a sense of how he incorporates sports into his teaching style watch the video shown above — it’s two minutes very well spent, I promise.
ITEM! Color Remix cap launch: I’m happy to announce that the first monthly round of Uni Watch Color Remix caps is now ready to go. The four designs, which I teased a few weeks ago, are as follows (click to enlarge):
Nice, right? All four are 100%-cotton strapbacks. They’re available here and will remain available through mid-October, when we’ll launch a new batch of color combos.
Meanwhile, in case you missed it last week, we also have corresponding T-shirts in the same four color combos:
Here’s where you can get the Black/Yellow, Royal/Orange, Navy/Red, and Red/Navy versions. (The T-shirts will remain available in the Uni Watch Shop indefinitely, and we’ll launch new shirts next month to match up with the October cap designs.)
My thanks, as always, for your consideration of our products.
Seam ripper update: I am once again restocked on all five colors of Uni Watch Seam Rippers. They’re available here.
By Lloyd Alaban
Baseball News: The Black Lives Matter stencil on the mound was obscured by the score bug in last night’s Cleveland/Cubs game, at least for the Cubs’ TV feed. Cleveland’s feed had a smaller bug, which meant the stencil was unobscured (from Phillip Santos and @CLETribe). … Padres P Zach Davies was still wearing the team’s Roberto Clemente sleeve patch on his cap last night, despite the fact that Roberto Clemente Day was last week (from Jakob Fox). … The Orix Buffaloes of NPB (which merged with the old Orix Blue Wave) wore Ichiro Suzuki-era Blue Wave unis yesterday. The Blue Wave was Ichiro’s former team before he joined the Mariners (from Dustin L. Meador). … The West Coast Baseball League has established a new team in Edmonton (from Wade Heidt). … Also from Phillip: The Paper Stadiums Twitter account made a 1968 replica of Wrigley Field entirely out of paper.
Football News: A vintage 1960 Raiders sideline cape is on display at the Raiders Image store at Allegiant Stadium. Amazing script! (From @khaled74.) … New uniforms for Bemidji State (from @doubleasterisk).
Hockey News: Here’s a video showing the Stars’ equipment staff adding the Stanley Cup Final patch the team’s jerseys (from Bill Larkin).
Basketball News: Here’s a look at some of the uniforms the Spain national team has worn throughout the years (from Jeremy Brahm). … High school player Nathan Bittle’s dad announced his commitment to the Oregon Ducks with a graphic showing Nathan Photoshopped into an Oregon women’s jersey and men’s shorts (from Derek Buchheit). … Louisville’s arena, which had arguably the most embarrassing corporate name of any sports facility in the country, will be getting a new name, because the current naming rights holder has opted not to renew its deal (from Timmy Donahue).
Grab Bag: The U.S. Space Force will soon let some members test out dress and physical training uniforms (from Timmy Donahue). … Also from Timmy: Apparently President Trump wanted First Lady Melania Trump to have a hand in designing some of the Space Force’s uniforms. … Tokyo’s Ohta City Gymnasium has created a completely socially distanced seat map for the entire arena. Here’s the layout for basketball (from Jeremy Brahm).