FPG has gone mobile!
A while back I wrote an article about the origins of FPG, a game I designed, built, and hosted in streamlit. I wrote about the rules and went into detail regarding where the idea came from. At the time I had around 30 people beta testing the concept. Since then FPG has evolved into something with a much wider reach — A mobile app. Continue reading to hear about that process and what I have learned along the way.
Lessons from Beta Testing
So, after 6 months of FPG in the wild, what have I learnt?
The first thing I’ve learned from beta testing is that the concept works — FPG is a genuinely fun, fresh, and unique way to engage with the Premier League without demanding too much time. People are enjoying it, and as a player myself, I can personally vouch for how much fun it is.
One of the most surprising takeaways has been FPG’s ability to engage people with little interest in football. Naturally, I’ve roped in those closest to me, and out of what I assume was a sense of obligation, even people like my girlfriend’s mum agreed to play. But since then, they’ve been checking weekend scores to see how their FPG points stack up — which makes me think I might be onto something.
I can’t begin to describe how exciting it was to see people actually using something I had built. Suddenly, comments like “I got -2 points on FPG this week” or “There’s no way you’re doing better than me on FPG” started popping up in group chats. Knowing that these conversations were happening purely because of the code I had written was an incredibly cool feeling.
The second big lesson I learned is the importance of testing a concept — getting something in front of people early so you can see how they actually interact with your idea is invaluable.
Anyone who has ever managed software that others use knows how much people love to complain, and FPG was no exception. I’d regularly get texts about how painful the login process for the Streamlit app was or how I “needed” to build a mobile app because they had once again lost the URL. (Despite the fact that it’s in every weekly email, pinned in every group chat, and has been sent multiple times… seriously, how hard is it to bookmark a URL?)
Rule Tweaks
One thing I hadn’t fully considered was that, unlike me, FPG wasn’t going to be a top priority in everyone else’s life. This meant I had to figure out how to handle players who didn’t submit a pick each week.
The core mechanism of FPG relies on everyone making 38 selections over the season, with their available choices narrowing as the weeks go on. But if someone didn’t pick, this process broke down — and, in fact, not picking could actually be beneficial in two ways. First, it eliminated the risk of losing points. Second, it allowed players to retain all their available choices, giving them an unfair advantage later in the season. It quickly became clear that I needed to handle this programmatically, especially since some of my so-called trusted beta testers (not naming any names… Tom Meadows) couldn’t be relied upon to submit a pick each week.
This led to the first major tweak to FPG’s rules. Following a suggestion from Hannah Reeve, I introduced a system where, if a player didn’t make a selection, they would be randomly assigned a team from their remaining choices. However, if that team earned positive points, the player would be docked one point as a penalty for not picking. Negative points, on the other hand, would still apply in full. So if the randomly assigned team won a derby and earned +2 points, the player would only receive +1 due to the penalty.
The goal was to encourage players to make a pick while ensuring that missing a week wouldn’t ruin their season. Life happens — you might be on holiday or just forget for a couple of weeks — but this way, you could still return and keep playing without being at a major disadvantage. At the same time, it ensured that those who actively participated weren’t at a disadvantage compared to those who skipped selections.
Another rule tweak came from a conversation with Tom Dent, who suggested a cool idea — awarding a bonus point if you were the only person to pick a winning team. So, for example, if Bournemouth won and I was the only one who had chosen them that round, I’d get an extra point as a reward for the unique prediction.