Pokémon TCG: Team Rocket

Written by Bryce Summers on 2.6.2025

A few days ago, my beloved partner Emily learned that there was a 'vintage' Pokémon trading card game set based off of Team Rocket. She told me that she wants to play the game with two decks based on this set.

My relationship with the Team Rocket set goes back much farther to my childhood when it first came out. When I was growing up, my dad used to bring home packs of Team Rocket cards on random days of the week. More than 2 decades later, he still discovers packs in nooks and crannies and other hiding places in his bed room.

I enjoyed opening up packs of cards back in the day, but had a hard time using the Team Rocket cards in my decks back then. Many have uninspiring attacks and they all have low hp. I wasn't the only one with a sour opinion of the Rocket set as Jason Klaczynski points out in his article on the Base-Rocket format.

Historical grievances aside, I'm motivated to build 2 new decks. First, I have many more Pokémon cards now to use than I had as a kid, including a full set of the unlimited cards from the Team Rocket expansion. Second, I'd like to use some of the underappreciated Pokémon with cool and/or unique abilities, such as Dark Machamp that could be more competitive when they faces Pokémon of similar hp.

Format Definition: Rocket Only

Before building the two decks, I need to define the restrictions I am working under. Here is a list that describes all cards that I may legally use when building the decks:

A card is legal iff:

• its from the Team Rocket Set.
• its a Team Rocket Promo card.
• its a basic energy card from Base Set or reprint.
• its Double Colorless Energy.
• its a trainer from the Base Set, Jungle, or Fossil.

Guidelines:

• Make sure that cards from the Team Rocket Set are featured.
• The decks should interact well with each other.
• Both decks should be fun to play with each other.

Energy Quantity Theory

The energy counts for the example decks in Klaczynski's article are as follows:

These numbers seem incredible to me, since I never would have put only 7 energy in my deck as a kid and I've played far more Magic: The Gathering (MTG) than Pokémon (PTCG) in recent years. Compared to MTG, PTCG has astoundingly powerful card draw and tutoring abilities. I think these abilities are balanced by the usual* speed limit of only being able to knock out 1 Pokémon per turn, regardless of how many cards are drawn.

Before building the deck, I want to understand the relationship between the quantity of energy I put in a deck and the number of turns I can reliably attach energy before running out given a certain set of trainers / attacks / abilities that I can use to find energy. I also want to understand how trainer cards that tutor or draw cards affect this relationship.

No Trainers

First, I've calculated the probability of hitting X turns of energy attachments given a quantity of relevant energy in a deck that contains no trainers. I prefer at least 80% reliability in my MTG decks, so I've highlighted the energy count I would put into a trainer less deck for each turn goal.

Notice that we would need 11 energy in the deck to be able to attach an energy on the first turn, 17 to attach energies on the first 2 turns, 22 on the first 3 turns, and 26 to attach energies on the first 4 turns.

Also note that without trainers, there is no way to balance out drawing too few or too many energy cards. I've chosen 80% as by standard because I would prefer to draw too many energies than draw too few, because at least I can play the game, just perhaps with less Pokémon.

4 x Bill

Let's add Bill, the simplest base set trainer, to the deck. We'll assume that a deck has 4 copies of Bill in it for our probability calculation.

Notice that these results correlate with the Xerox principle from MTG. For each 4 high quality cantrips in our deck, we should expect to need 2 less energy cards to achieve the same attachment goals.

Although the xerox principle means that we'll need more cards (energy + bills), the bills can be used to find cards other than energy, which is a great step towards regulating situations when we draw too few or too many energy cards.

4 x Professor Oak

Next we'll do the calculations for a deck containing 4 Professor Oaks. From a MTG perspective, Oak is like a one-sided wheel of fortune for free!

When comparing the Oak table to the No Trainers Table, it looks like the Oaks reduce the energy count by 3 for a first turn attachment, then they become even more relatively effective as our time turn goal increases. By turn 5, we need six fewer energy cards (23) in an Oak deck compared to a Trainerless Deck (29 energy).

If it comes down to putting Bill in my deck of Professor Oak, then I'm definitely going to choose Professor Oak. The good news is that we can play both and they operate independently.

4 x (Bill + Oak)

I then tested a deck that has 4 Bills and 4 Oaks. Based on my results, Bill should be played on any turn where I can play an an energy card, but don't have one in my hand. Oak should be played in the same situation, but only if I don't have a Bill in my hand.

Bill+Oak performs better than either alone.

4 x Computer Search

If Computer Search is the only trainer in a deck, then it is fairly straightforwards to do the probability calculations. We want energy and search for energy. Later in this article when we add Professor Oak to the deck, we will need to need to decide whether it is best to search for an energy or another trainer card.

According to these results, Computer Search Outperforms Bill across all turns for decks that have less than 28 energy.

Computer Search only outperforms Oak for decks with 10 or less energy.

Computer Search decks require 4 less energy than Trainerless decks, since each of them can count as 1 energy card.

I believe this is because it is very likely to find an energy in 2 random cards when almost half of the cards are energy cards. Furthermore, a card like Bill can always be played, whereas Computer search could be a brick in a hand without 2 extra cards to discard.

4 x (Computer Search + Bill)

Based on a quick experiment, we should search for energies if there are 26 or fewer energy cards in our deck, but should search for bills if the deck has 27 or more energy cards.

Searching for Energy requires 3 fewer energy cards for a reliable 80% first turn energy attachment.

Computer Search finds energy cards.

Computer Search prioritizes finding Bill.

4 x (Computer Search + Professor Oak)

Here is where things get complicated.

Computer Search for Energy, then use Oak last

For this experiment, I had my simulation use Computer Search whenever it didn't have a needed land and it didn't discard the last Oak in a hand.

Computer Search for Oak if not in hand, else search for energy.

This is an astounding improvement! It looks like I should always search for Professor Oak, unless I am playing with a deck that has 5 or fewer energy cards.

For decks with 30 energy cards, searching for Professor Oak leads reliably hits 11 more turns of energy attachments! Wow!

Other Strategies

Always computer search for energy, discarding other cards at random. I ran a quick simulation and found that Blindly using computer search for an energy, while discarding any random 2 cards was an inferior strategy.

I also checked playing professor Oak first. It doesn't look like an improvement either.

I also tried searching for enrgy, prioritizing discarding Oak last. I saw no major differences in behavior to prior strategies I've run.

4 x (Bill + Computer Search + Professor Oak)

Now we'll run a simulation using all three trainers.

4 x (Bill + Computer Search + Professor Oak + Computer Error)

Finally, Let's run a simulation for those rare decks that will want to play computer Error.

According to these results, the Base - Rocket Rain Dance deck with 14 energy cards could be expected to hit the first 12 energy attachments.

Pokémon Count theory

Looking thru different articles on Vintage Pokémon decks and Modern Pokémon decks, I've seen some trends in Pokémon Card counts. It looks like we want 4x of any card that we will be using as our primary attacker if it is a basic Pokémon, 3x for a primary attacker that is an evolved Pokémon. In modern decks that use multi-prize Pokémon, there might be less of a need for so many Pokémon as they will have lost after losing 3 attackers, compared to a vintage deck that will lose 6 Pokémon prior to defeat.

Utility Pokémon families (Used for their Pokémon Power's / other helpful abilities) that aren't meant to attack, such as Fossil Muk/Grimer might be included as 2x/3x or 2x/2x quantities, since we only need 1 Muk and it can last us the full game.

Based on my calculations, I would want to have at least 12 Pokémon in my deck to guarantee that I have at least 1 basic Pokémon in my hand >= 80% of the time. This would prevent my opponent from being awarded 2 cards as I take a Mulligan. Beyond the first turn, trainer cards can be used to find other basic Pokémon to setup the gameplan.

Stage 1 Pokémon can be removed entirely by the inclusion of 4x Pokémon Breeder or 4x Rare Candy.

When can we expect to play our Pokemon?

Assuming a deck has 4xBill, 4xOak, and 4xComputer Search, and 4 copies of a Pokemon, then a player should be able to get at least one copy of that Pokemon into their hand by turn 1 with 86% probability, assuming they don't care about anything else like playing energies and they started the game with at least 1 other basic Pokemon in their hand.

  # Assume 4 copies of a Pokemon in deck.
  Pokemon in hand by turn 1: 0.86
  Pokemon in hand by turn 2: 0.90
  Pokemon in hand by turn 3: 0.93
  Pokemon in hand by turn 4: 0.95

More realistically, we want to play an energy each turn and find our Pokemon. The following table shows the chances of attaching an energy each turn and being able to play at least 1 of 4 copies of a card in our deck, assuming we have 4xBill, 4xOak, and 4xComputer Search.

4xPokemon, 0xTrainers that tutor for Pokemon.

What if we have 4xPokemon + 1xTrainer that tutor for that Pokemon?

What if we have 4xPokemon + 2xTrainers that tutor for that Pokemon?

What if we have 4xPokemon + 3xTrainers that tutor for that Pokemon?

(Or 3 Pokemon + 4 Tutor Trainers, which is a likely scenario for evolved Pokemon.)

What if we have 4xPokemon + 4xTrainers that tutor for that Pokemon?

Note: New graph shows what happens if we play needle before searching for energy, then search for needle.

Old graph shows what happens if we search for energy before searching for/playing the needle.

When can we expect to play a stage 1 Pokemon or a stage 2 Pokemon

This is a more involved question and will require me to extend my simulation to include tutors that might cover more than one type and to include Pokemon Breeder / Rare Candy. Pokemon Trader / Erika's Kindness

Let's Imagine a relevant tutoring situation.

Deck 1