Nerd Notes, July
Hey there nerds…I’ve got 3 things running through my head this month:
- A sneak preview of where Pursuit of Play is (might be) headed
- Continuing to chase the white whale - the Pokemon Deck Generator
- Some thoughts about online versions of in-person games and how they affect the in-person competitions
So, let’s get into it…
Pursuit of Play season 2
I mentioned in the last edition of PoP that next season is going to look a bit different, in part because we’re going to be indifferent to World Championship qualifications and therefore play fewer events, and in part because I need some conceptual evolution. The Pokemon Travelogue was a bit anthropological and involved me as a spectator to this phenomenon, this first season of Pursuit of Play was a bit more sociological but added the dimension of me acting as a player. There’s an obvious route if I want to keep this contained within the world of Pokemon, which is to join the team of event judges; the main problem is that I have no desire whatsoever to be a Pokemon Judge.
But I do have an idea that’s taking shape somewhere in my cerebral cortex, so I wanted to share it here because you all might want to add on to it and/or discourage me from it. Also, this new approach may have more opportunities for newsletter readers to join in.
For the next season of Pursuit of Play, I think I’m still in a sociological mode and concerned with a greater variety of spaces that invite people to engage in collective play. Pokemon tournaments are fine - and we’re still going to play 5 events this season - but the more time I spend thinking about play and especially feeling more and more drawn to it as a social and creative activity, the more I see the limitations of just focusing on a single game.
So, over the next year I want to visit a few places where other forms of play are happening, with a particular focus on environments for play for adults. The question that I’m becoming more interested in is how people show up differently in environments for play than in environments that are ostensibly more serious…and how the two influence each other.
I have a loose plan in place:
- August - Gamescom in Koln, Germany (probably)
- January - the MIT Mystery Hunt in Cambridge, Mass.
- March - a combo trip to the Game Developers Conference in San Francisco and a visit to the IDEO Playlab
- May - spend the World Play Day at the LEGO House in Billund, Denmark
- July (or maybe August? Dates TBD) - Attend the PlayMakeLearn conference in Madison, Wisconsin. Maybe do some sort of workshop or presentation there.
And then alongside the 3 US trips - and probably alongside any other US travel that I do for other work engagements - I want to do a pop up event for Read Me Like a Book, the storytelling & connection card game that I’ve been developing. So while I was a spectator in the first year and a player in the second year, in this third year I want to focus on being a creator and give some time and energy to ReadMeLAB. I’ve only done the one test with it, but in the 5 months since I haven’t been able to shake the feeling that there’s something to it…and that’s a long time for an idea to stay with me.
What do you all think? Is this angle interesting to you? Are there other places I should plan to visit? Want to join in on any of these? Drop me a line and let me know.
Chasing the White Whale
Back in October, I mentioned my attempts to build a Pokemon Deck Generator, and at that time I said:
A few months ago I did a little experiment: I tried to build an AI tool that would generate novel Pokemon decks, thinking that it might provide a slight edge to come up with an unexpected rogue deck. This experiment was a dismal failure - in part because of the complexity of the game relative to something like chess[6] , but even more because of the nature of LLMs being trained on historical data and Pokemon being a game that renders its own history obsolete. Since then, I have spent a lot of time thinking about the technical feasibility of a Pokemon AI, and while I think it’s possible, the amount of compute needed is massive.
And then I went in depth in the first installment of Nerd Notes on why it didn’t work…I’ll spare you the details. All of this is to say, I probably should have just given up on the idea then.
But I haven’t. Over the past couple weeks, I’ve been back at it using a combination of Claude, n8n, and Lovable to see if I could build a better version of the deck generator. At the very least, I think I can get one that actually makes recommendations of Standard Legal decks. But the process of developing it feels like a couple steps forward where I feel like, "yeah, this thing is feasible" followed by somewhere between 1 and 3 steps backward where I think, "actually no, the decision tree involved in this is stupidly complex." The essential tension that I'm navigating is whether the ability of the computer to hold all of the traits of all of the cards in mind is more useful than the ability of the player to identify and understand the synergies of different cards and how they are useful with regard to specific game mechanics and dynamics.
I was kind of hoping I'd have a rough version to share in time for this newsletter...but the basic feasibility is still very much in the works. I've overcome the problem of the first Playlab version - that the model resorts to its full corpus of training data and consequently recommends rotated cards - but not the standard of "suggest a legal 60 card deck" much less "suggest a competitive 60 card deck" much less "suggest a novel, competitive 60 card deck." If I can get to that second tier, then what I'd like to do this season is use the generator to craft a deck 2-3 weeks ahead of each tournament I play and then use that deck in tournament play. Right now, I think it's about 50/50 whether or not that actually happens.
Online Poker & Online Pokemon
In the early 2000s, online poker became a thing as a few different phenomenon all converged at once - 1 part the success of the Matt Damon/Ed Norton movie Rounders (we’ll never be able to unhear John Malkovich’s Russian accent), 1 part ESPN trying to fill up its 24 hours of programming every day with low production cost live sporting events, at least 2 parts the dot com boom and the desire to find ways to monetize the internet, 1 part the rise of broadband internet making online gaming possible.

By roughly 2003, we probably all knew someone who was making a living playing online poker. My friend Mike - already on a full academic scholarship with a nice stipend - was that person for me (actually one among many, but probably the most successful one). In the course of a couple months, he was suddenly making a few thousand dollars a month in poker winnings. Because I saw him succeed, I conversely realized that I probably would not. I watched him play one day and could recognize how quickly he was calculating odds in his head. My math skills were sharp, but they were nowhere near his level…still, they were good enough for me to be able to calculate the probability that playing in an infinite online casino I’d run into another player whose math skills were as good as his (ie, high).
2003 was also a breakthrough year for professional poker because it marked the first time that the main event of the World Series of Poker was won by a player who had qualified through an online tournament. That was a catalytic event in a chain reaction - the world of online poker had already broadened the game from something that was happening in back rooms and casino floors into something that people were doing from the comfort of their living rooms, and then anyone who was paying a little bit of attention could see that the online game was actually a credible pathway to professional success. Poker became mainstream.
It’s a common internet story, right? A new platform that broadens access to something that had been niche if not actively gatekept surfaces an even larger of people who excel at that thing, and usually a more diverse group. Techno-optimism comes in for a lot of well deserved critique, but this democratization phenomenon is one of the great things that has come of the digital age.
And I say all of this because I do find myself wondering if Pokemon is in the midst of it. I’ve mentioned the arrival of the Elo ranking system in the online game, and while just the existence of the game itself lowered access barriers, the addition of Elo ranking at the top now means that online players who work their way up there can guarantee themselves of always playing high quality matches…and playing high quality matches is the best way to level up skill.
If you’re only paying attention to the Masters division, I don’t think you’ll see this working itself out yet…though there are some indicators that I’ll come back to in a second. At that level, the skill level of the best players has been earned over such a long period of time, and they’re also actively playing in the Elo system so it’s not like it gives online players an advantage. In the Junior division, though, you can see it. Kids in that age division can build and refine a deck and test against really good competition before they ever show up to their first tournament. In person is different from online because there’s so much more that the player has to mind & manage themselves that is just enforced in code in the online environment, so it is an adjustment, but there were so many competitive junior players this year.
On the Masters level, the story to pay attention to involves a player named Ray Chen. She only played 3 tournaments this year, finishing 69th at the Vancouver Regional in March and then making the top cut at the Portland Regional in May and landing in the top 8 in the North American International Championship. In those latter two, she was piloting a deck that no one had ever seen before built around the Toedscruel ex. I don’t know for certain that she was playing a ton online between March and May, and if she had only posted a result in 1 of those last 2 tournaments I’d just chalk it up to luck…but the consistency of those 2 finishes within a month suggests a level of fluency with the deck and awareness of how to play it against the best players piloting the best meta decks.
Is Pokemon blowing up in the same way that poker did? I’m not sure. It’s undoubtedly growing - every region set records for largest regional and international tournament this year and had done so the year before as well. At the Masters level, Europe didn’t have a repeat tournament winner this year. The field of play is pretty open right now. I could name a handful of players who should be considered favorites at the World Championship (Brent Tonisson, Henry Chao, Isaiah Bradner would be at the top of that list), but you could make a credible case for at least 2 dozen players - and last year, the player who won wouldn’t have even entered into the discussion, wouldn’t have even been a name that anyone would have considered because nobody knew who he was. I think all of that is good for the game.
That’s it for now. Watch out for a side quest edition of PoP coming out later this month…what kind of opportunities does a person (me) find coming their way when they are known to be curious and open to random possibilities? I’ve got a good one.
Member discussion