Shiinotic vs. Sylveon EX in Solgaleo EX Decks

If there is one thing that is happening with every expansion in Pokémon TCG Pocket, it's that the game is getting faster. On the first expansion, we only had Professor Research and Poké Ball as cards to draw from the deck (and Meowth or so). Then slowly but surely, we started seeing cards like:

• Sprigatito: a "Find a Friend" user with a stronger Meowscarada followup.
• Rare Candy: speeding up the evolution timing of Stage 2 Pokémon.
• Shiinotic: with an ability to also "Find a Friend" but for any characteristic.
• Sylveon EX: which basically feels like having 4 Professor Researches.
• Silvally: that can hit 100 on turn 4 and it's not an EX.
• Baby Pokémon: allow you to use a move on turn 1, or speed up a benched Pokémon.
• And most recently Suicune, Raikou, and Entei: allowing you to draw an extra card on every turn!

Although fallen out of the meta, one of the fastest decks during the Eevee Grove set was Solgaleo EX. This is because it only requires 2 energies and, with the help of Rare Candy and cards that help thin your deck faster, it can hit up to 120.

Now, the question remains: what is the fastest way to get Solgaleo EX active?

To answer this question (and after losing plenty of games on ladder) I decided to use a bit of programming and data analysis. The idea was simple: simulate a lot of games, half with Sylveon EX as support and half with Shiinotic as support.

I chose these two since they are the most common supports seen both on ladder and in tournaments.

Experiments

To make sure the simulations were counting the fastest way to play Solgaleo EX, the bots prioritized moves in this order:

1. Use Poké Ball (if possible)
2. Use Shiinotic's Illuminate Ability (if possible)
3. Use Professor Research (if possible)
4. Use Rare Candy (if possible)
5. Evolve all possible Pokémon (and say "Yes" to Sylveon's Ability)
6. Attach Energy and Attack

Playing Poké Ball first ensures it doesn't go unused — in theory Shiinotic could pull your last Basic from the deck, leaving the Poké Ball with nothing to find. Similarly, Professor Research is played after Poké Ball and Shiinotic, since it could accidentally pull Pokémon, leaving both of those without impact. Evolving Sylveon is also left until after Poké Ball, though the order between it and Professor Research doesn't matter much.

The decks used for the simulation were:

The remaining item slots are ignored by the bots in this analysis since they don't play a role in the speed of Solgaleo being online. There may indeed be more optimal versions of these decks out there, but for measuring raw setup speed all the ingredients are present.

Results

To my surprise, both variants are fairly similar when it comes to speed, but Sylveon EX is slightly faster. Specifically:

• Shiinotic variant: Solgaleo EX is active by turn 4 in 48 out of 100 games.
• Sylveon EX variant: Solgaleo EX is active by turn 4 in 51 out of 100 games.

In terms of raw speed, Sylveon EX wins. But. and this is a very big but, given the utility that Shiinotic provides against the most threatening counter in the format, the one and only Oricorio, I think we should conclude that Shiinotic is probably the better choice.

In general, Shiinotic is also worth fewer points to the opponent as a KO target, and it can actually attack in an Energy-only deck. Those two additional benefits from Shiinotic outweigh the 3-game-per-100 speed difference.

To be fair, Sylveon EX has only 1 retreat cost while Shiinotic has 2, which pulls me back toward thinking it's not all that simple.

That is why I think the only way to fully understand the optimal deck configuration is to check it out empirically, play out (or have bots play out) thousands, if not millions, of games with different configurations. This way we can account for all the complexity the beautiful game of Pokémon TCG Pocket entails.

What's Next?

The analysis above is just a narrow comparison of Shiinotic and Sylveon as supporting Pokémon for Solgaleo. There are many more aspects to consider. For example: is it better to run Giant Capes or Rocky Helmets in this deck? To answer questions like this, the only reliable approach is simulating millions of games with different variants, since the game is too complex to analyze theoretically otherwise.

This analysis was thanks to the open-source game engine Deckgym uses. A slim re-implementation of the game built for large-scale simulations as we strive to find the most optimal decks out there.

The code is on GitHub: bcollazo/deckgym-core. Feel free to check it out and build your own experiments; it's also a great way to learn Rust or practice using modern AI-assisted coding tools!