Článek v češtině zde / click here for Czech translation.

Where I’m from, the NFL is labeled “American football”. If you only say “football,” you mean soccer. And if you say “soccer,” people think you’re an asshole. That alone tells you that NFL fans are in the minority in the Czech Republic. I was a kid the first time I saw a football game, and I had no idea what was going on. Five on each side, big men pushed each other around for a few seconds, then the whistle blew. After thirty or forty seconds of nothing—just waiting— the ball was snapped again. By then, my attention shifted back to my comic-book. That day, the modern gladiators remained a mystery to me.

A couple of years later, at a summer baseball training camp, I finally understood. During a lunch break, an American coach offered if we want to toss around a football. I was a pitcher, and in this group, I was the one with the best arm. It did not help: when I tried to throw that strange thing, it wasn’t a good look. I threw a damn duck. The ball wobbled in the wind until it was mercifully pulled by the gravity at the feet of a kid I wasn’t even aiming for. We then received a quick brief of the mechanics when throwing a football, along with the basic rules of the game. I happily forego quarterbacking, lined up as a receiver, and we played several downs. That day I’ve learned that just by explaining the ‘offense has four downs/attempts to gain ten yards’ goal, people will start getting the sport. They still might not enjoy it, but they will get it. It took couple more years before I was able to watch the games regularly on our television. Once that happened, though, my attention level quickly grew from indifference to a hobby to a source of countless sleepless nights to an extra income.

In 2014 I published free betting picks on the Czech betting site KolemDvou. I’ve had excellent results, and my bets had thoughtful analysis behind them. Shortly after Super Bowl, I received an offer: to write a five-part series about football to help potential new viewers better understand, which I happily did. I worked my ass on that series, too. Shortly after finishing it, another offer came by: to sell my picks in a premium part of the site, accessible after paying, which I happily did also. I didn’t see anything wrong with that. I already established myself as a credible figure, so promoting myself wasn’t even needed. On top of that, in my young, cocky mind, I considered myself an expert, a prodigy who had the game figured out. Yeah, right. I had one thing goin for me. A new weapon in my arsenal: in January 2015, right before the post-season started, I created a rating/betting model. It used play-by-play database I’ve put together. Transforming the plays from the messy ESPN webpage to tidy excel spreadsheets kept my Mondays busy for the next couple of Autumns. That model helped me pull statistical comparisons for any two teams I wished. After my first primitive spreadsheet successfully predicted Colts upsetting the heavily-favored Broncos in the divisional round, it was on.

This is a late version of my model Anthony. It was named after Matthew McConaughey’s role of football tout in Two For The Money.

The 2015 was supposed to be the first on a long list of seasons I worked as a betting advisor. And I started extremely well. I went 2-0-1 on the Thursday opener and 4-1 the first Sunday. At this point, I felt unstoppable. Next two Sundays, a reality showed its ugly head, as I went 3-7 and 5-7. This erased my winnings from the first week and then some. Week four I was back in black, as I went 6-1. I could go on and on, but this was my experience in a nutshell. Ups and downs, mostly downs. Highs and lows, more lows. Heavy swings. Not a single week of average results to settle down a bit. Mentally, I was as wrecked as a pitbull’s favorite chew toy. I wasn’t made for this game and tapped out about three months into the season.

They say hindsight is 20/20. Today, the pattern seems obvious. Sure, I liked playing baseball—but I loved analyzing baseball. Just like I enjoyed betting football—but I loved discussing, writing about, and learning about football. Certainly, I didn’t particularly appreciate losing other people’s paychecks. I also wasn’t built for big swings, even when the conditions were in my favor. Like when I was making okay money playing online poker back in 2011. Poker was very beatable those days. Even then and there, I didn’t take the big swings very well, but I could handle myself. But now, instead of facing István or Ference from Budapest, I was up against the most efficient betting market in the world.

After this epic failure, I switched my attention entirely to analyzing stats and learning the game on a deeper level. I added more data, advanced stats, and layers into the model. I worked backward to fill the database with another two or three years worth of data every single off-season. Of course, I wasn’t the only one who collected the play-by-play details and who appreciated the flexibility that comes with owning a neatly organized database. There even were people and projects that charted advanced data, the sort of things that I felt I was missing in Anthony.

To chart a play, someone has to actually watch it and mark whatever he’s interested in manually. It’s next to impossible to do it without a fairly big team. Automating it is not really possible unless you’re working for Nextgen Stats. They are the league’s official source of data, and as such, they have unprecedented access. They put microchips in players’ pads and fully automated the process of data mining—even the NFL realized how valuable this kind of data could be. Sure they are: they provide context normally isn’t there. For example, the quarterback happens to be under pressure on a specific play. But with the benefit of additional information, you might know that the play call was a screen pass. In that case, the pressure is half-expected because the linemen block downfield instead of protecting the QB. Or there’s a late-game interception that kills a team’s chances to win the game. The context can tell that the ball was intercepted only after the receiver tipped the ball—not the QB’s fault.

I was looking for that type of project for about a year. Every once in a while, I googled charting projects and ended up disappointed again. There was work available, don’t get me wrong, you just had to be from the USA or Canada to apply, even if it was an unpaid voluntary job. Just like the fat kid in Moneyball, I had problems managing my weight I was trying to step into a world that wasn’t very welcoming to my kind. But one day, I finally got lucky. I found a blogpost at Armchair Analysis. In the past, its owner Dennis Erny was doing a similar thing I was. Only he did it longer, way better, and monetized it. In 2016, he was looking to expand into this new territory of play charting. The blogpost wasn’t older than three days, so I wasn’t late, and as a matter of fact, it didn’t specify that you had to be based in North America. I thought maybe he forgot to add that condition, or perhaps the option of Euros applying for this type of job didn’t even cross his mind. I went for it. I typed a little motivational letter, added a bunch of my old analysis, which must have been humorous because that shit wasn’t written in English—but hey, there are charts, he’s gonna get it! I really went overboard trying to make the first impression. If I was going to be rejected, I wanted to do it right.

The first season I only worked four hours a week, getting intimate knowledge with players on Carolina Panthers and Baltimore Ravens, as they were “my” teams. I was part of a huge group of volunteers just like me, and I fell in love with the process. Before the next season — this was 2017 — Dennis wanted more data and installed several additions. My excitement went through the roof. So much that I couldn’t control myself. At times, I seriously tested the fine line between being a valuable, enthusiastic employee and being a downright annoying twat. But at the end of the day, that was for the best. After all, while Armchair had a long history of selling data, the charting project was brand new. It takes a while to build a dedicated, well-established team, especially when dealing with Joe and Jane from the internet if those are even your real names. To make things more interesting, several people quit within the first two weeks without as much as saying ‘this job sucks, bye‘. In that situation, without the benefit of an experienced group of reliable people with solid coaching and know-how, having a psycho on your staff can go a long way.

Every year, the number of things we looked at and collected increased. New data led to new clients. More clients led to extra requests about future additions. The broader range came with learning new things and figuring out new guidelines. After all, if ten or twenty different people are doing the same task, you want them to look at it as close to identical as you can get. Ironically, as our quality rose, the fewer people we had on our staff. In September 2017, just as the new season kicked off, we felt quitters crashing the team morale for the first time. You kinda have to expect it in this predicament; that’s the beauty of doing anything on the internet. One week the guy is proposing to you, the next week you’re getting gthosted. According to Google, step 1 in dealing with being ghosted is ‘Accept your feelings and allow yourself to hurt‘. Fuck that. As much as I cared for the work and for Dennis, whom I was quickly becoming friends with, I didn’t mind seeing the unreliable twats going. Like Jocko is saying: I have to do this guy’s work? Good. I stepped up, fearing that one of these M.I.A.’s might be the first domino to fall, and the chain reaction could kill the whole project. I took plenty of workload on my back. Not a single deadline was postponed, and by October, I was promoted to the lone senior position. I was now consuming so much football that I stopped watching live games altogether. Well, except for the Super Bowls, and I only tune in to see the commercials, okay?

Until now, I was getting paid few dollars here and there. In 2018, I worked under contract for the first time, so I was officially getting paid for watching the NFL. I also had a voice when we discussed which direction the project should go. Dennis made things comfortable enough that the relationship blossomed into friends/colleagues rather than boss/employee. Maybe not the best strategy for most companies, but it worked wonders for us. For 2019, I was named a ‘Head of game charting’. What it meant? The contract read pretty much like this: more responsibility. More workhours. Some nights you’re gonna hate yourself. You will be constantly self-aware that you’re close to burning out. Hey, sounds good to me!

The whole thing even started to make sense as a profitable business and potential full-time job. We rolled into 2020 with a new attitude. This time we didn’t even use the offseason to take a break. As the world fought the pandemic, we just kept on rolling. Balancing the budget, figuring out the correct pricing, scouting for the right people, writing manuals, coaching. On the back of the NFL covid-season, we’ve put together the biggest downloadable data collection on the web for about a third of the prize our competition asked. We collected everything—you name it, we did it. Time to pass? Absolutely. Time to pressure? Damn right. Result of the first contact between ballcarrier and defender? You betcha. Stunts and zone blitzes? Yes, and yes. Specific players lining up on the field on any given play? Actually, no, we didn’t do that one. But we did everything else, and it all paid off: at the end of the year, the project attracted investors and eventual buyers of the project.

One of the features we installed for the 2020 season was tracking the field on the vertical and horizontal levels. Among other things, this gave us information about how many yards the ball really traveled through the air. Our numbers lined up with Nextgen Stats, which benefit from using automated tracking systems using microchips installed in players’ pads.