When Julián and I started playing Pokémon, I thought the difficult part would be learning the rules. The dangerous part turned out to be discovering that our decks could be better.
Each player brings sixty cards. Some are the creatures doing the fighting. Others give them the energy to attack or help you find what you need. The basic idea is simple enough. The number of ways to make it complicated is apparently endless.
We began with ready-made battle decks, the kind designed for people who would prefer not to earn a minor degree before their first game. You open the box, shuffle the cards, and theoretically begin having fun.
Theoretically.
Beginner decks are built to teach the game, which means they do not always work particularly well. Sometimes you draw a powerful Pokémon and nothing it can use to attack. Sometimes you draw the card needed to evolve a creature you have not found yet. Sometimes your hand contains several perfectly legal cards that have apparently agreed not to participate.
After a few games, I started noticing which parts of our decks were causing the trouble.
This was my first mistake.
I learned that certain cards could help us find the Pokémon we needed. Others could get energy into play faster or bring back something we had already used. A few small changes could make a deck feel less like a random stack of cardboard and more like something with an actual plan.
This was useful information. It was also extremely compatible with my personality.
Soon I was sitting on the floor with cards spread around me, combining two Lightning decks while waiting for a third one to arrive so we could make a super deck. I had a Master Ball. I had Rare Candy. I had a Zapdos deck on the way. I was asking which cards should stay, which should go, and how many copies of everything we needed.
The fact that I had barely known how to play a few days earlier did not appear to trouble me.
Improving the decks became its own game. There was always another card that could make one work more reliably, another weak link that could be replaced, another way to get the main Pokémon into play sooner. Each answer produced a better question, and each better question seemed to require a small purchase.
The original goal had been to build a few enjoyable decks for a father and his six-year-old.
Then I asked what it would take to make one of them competition-ready.
The question did not seem unreasonable. A deck that works properly is more fun than one that spends four turns doing nothing while the person holding it wonders whether the game has begun. We could remove the cards that had no real job, add better ones, and make the whole thing less dependent on luck.
Then we reached the Mega Diancie deck.
For the Pokémon-uninclined, Diancie is a small, pink, jewel-covered creature that looks as though a rock and a princess reached an agreement. The deck’s plan was easy to understand: get Diancie onto the table, give it enough energy, and hit things very hard.
The deck was already pretty good.
Naturally, I wanted to know what would make it insane.
There was an answer.
It involved replacing much of the original deck with specialized cards that worked together in ways Julián would not understand without help. The model came from a deck that had won a Japanese City League tournament, the kind of event where people arrive with carefully engineered card lists rather than a six-year-old who may need to be reminded that his turn is not over.
In the right hands, the stronger version could dismantle another deck with terrifying efficiency.
I was halfway through admiring it when I remembered that the intended operator was six years old and occasionally needed to be reminded that his turn was not over.
We had gone from making the deck a little smoother to bringing the City League guillotine.
A few minutes later, the phrase became more specific: the bedtime City League guillotine, a deck so efficient that Julián could ask whether Diancie was ready to attack and receive a short lecture before pajamas.
The funny part is that I was not rescuing him from some outside pressure. Nobody else had insisted that his deck become a tournament weapon. I had asked for the guillotine. I had researched the guillotine. I found the guillotine interesting.
That made the problem more complicated than the usual story about an overcompetitive parent. I was not trying to turn Julián into a champion or use him to satisfy some unfinished dream of my own. I was enjoying the construction. The system had become interesting enough that I had briefly stopped asking what the system was for.
Leaving the decks alone would not necessarily have served him better, either.
A beginner deck can be simple because its plan is easy to understand. It can also be simple because it does not work very well. Those are not the same thing. Losing because another person made a smart move can teach you something. Losing because your cards supplied nothing useful mostly teaches you that shuffling is cruel.
We did not have to choose between a clumsy starter deck and a tournament machine that required its own operations department.
There was a middle.
We removed cards that had no clear purpose. We added better ways to find Diancie and keep the deck moving. We adjusted the Lightning and Psychic decks so one would not flatten the other every night. The decks became more dependable without becoming harder to recognize.
The Diancie plan remained intact: find the sparkly rock Pokémon, give it energy, and hit things very hard.
Julián could see what his deck was trying to do.
That became the distinction I had been missing. I had been measuring improvement by what the deck could do. A better question was what the deck allowed the person playing it to do.
Could he understand why a card belonged there? Could he recognize when the plan was beginning to work? Could he make a choice and see what happened because of it? Could he still care that the Pokémon on the table was one he liked?
That last question mattered more than I expected.
At one point, while looking through possible upgrades, I asked for options involving more recognizable Pokémon. A competitive deck does not care whether a six-year-old has heard of the creature doing useful work in the background. It cares whether the card is efficient.
Julián is not a competitive deck.
Recognizing Pikachu, Charizard, or Zapdos is part of the experience for him. The picture matters. The name matters. A Pokémon he already loves can make an imperfect card more valuable than a better card whose only personality is that it improves the odds.
Later, while building another deck, we kept two enormous buffalo Pokémon in the lineup. There were probably cleaner choices. There were certainly more efficient ones.
But the deck had TWO FUCKING BUFFALO.
That was not the design failing.
That was the design remembering who would use it.
Improvement has a way of continuing after it has solved the original problem. Once a system begins getting better, every remaining flaw starts to look like unfinished work. The person adjusting it can always see one more refinement, one more advantage, one more reason the current version is not quite done.
The trouble is that the measurements usually belong to the system. They tell us whether the deck wins faster, whether the process uses fewer steps, whether the tool produces more output, whether the machine has become more efficient.
They do not always tell us what happened to the person using it.
The deck can become stronger while the game becomes harder to enjoy. The process can become more sophisticated while fewer people can understand it without help. The designer can become increasingly proud of something the intended user no longer experiences as theirs.
We still improve the decks. Julián will understand more of the strategy as he plays, and eventually some of the complicated cards may become the fun part. The point is not to keep him inside a beginner version of the game. It is to let the complexity arrive at a pace that gives him something to do with it.
For now, the Diancie deck works. Its main Pokémon shows up. The attacks make sense. There is enough strategy to reward paying attention and enough sparkle to remind us that we are still playing with jewel-covered monsters at the kitchen table.
The stronger list is still there if we ever need it.
But bedtime was never asking for a guillotine.
P.S. Since writing this, I have built enough Pokémon decks to require a documented arsenal, binder inventory, shared-card architecture, exact swap packages, and multiple Standard configurations. I also returned to Magic: The Gathering after twenty years and built a necromancer deck called DEAD AS F*.
The bedtime City League guillotine was not a warning. It was an origin story.