How to Check a Pre-Baltimore Meta Call Against the Actual Lists

"Best deck" is four different sentences wearing one coat, and the four rarely resolve the same way on a Sunday.

• Highest raw win rate. The deck wins the most games across the field, regardless of how many people brought it.
• Best positioned. The deck beats what the creator expects other people to bring. This can be correct about the deck and wrong about the weekend if the expected field does not show up.
• Best conversion. The deck moves its pilots from Day 1 into Day 2 and then into top cut at a higher rate than its share of the field would suggest.
• Safest pick for a given pilot. A deck with fewer decision points and fewer unwinnable matchups, which is a coaching claim rather than a metagame claim.

When you rewatch a preview video after results are posted, note which of those four the creator was actually making. A call that a deck is "positioned well" is not refuted by a low raw share, and a call that a deck is "the worst" is not vindicated by a single pilot bricking on stream.

The same care applies to a "worst deck" read. The common failure mode is that the deck was fine and the counter-deck the creator assumed would punish it never arrived in numbers.

Once official standings and decklists are posted, three numbers matter more than the headline winner.

Raw share versus conversion

Raw share is the percentage of the Day 1 field on a given archetype. Conversion is the percentage of that archetype's pilots who reached Day 2, and then top cut. A deck can be the most-played archetype in the room and still be a poor choice if its conversion sits below the field average. The inverse also happens: a deck with thirty pilots and eight in Day 2 is telling you something a raw-share chart hides.

Sample size, stated plainly

A deck with nine pilots and three in top cut has not proven anything. That is a sample small enough that one good pairing run explains it. Before you rebuild your list around a top-8 finish, ask how many people brought the deck and how many of them converted. Regionals fields are large, but archetype sub-samples often are not.

The list, not the label

Archetype names flatten real differences. Two lists filed under the same name can differ by four or five cards — a second copy of a stadium, a different energy count, a tech line that only exists for one matchup. If you are testing a creator's read, compare the actual counts in the posted list against what the creator described. Check the posted list itself rather than a screenshot in a thread or a summary in a video, including yours truly's summary: counts get transcribed wrong constantly.

And one structural caveat that applies to every regional ever run. One event is one region, one field, one weekend of pairings and one set of matchup luck. It can show that an idea works. It cannot establish a national metagame, and it certainly cannot establish that an idea will still work after the next set legality change or rotation.

AzulGG also published a 30th Celebration set review and buy list, which per the brief names cards including Fuecoco ex, Greninja ex, a Pikachu ex listed as #54, Mew ex, Espeon, Espeon ex, Unown, Lucario, Yveltal, Eevee and Minior. That is collecting content, and it should be read separately from the tournament preview.

Two things are not confirmed here and should not be assumed. First, I cannot confirm from this material which of those cards are legal in the format played at the Baltimore Regional Championship — a set being new does not make its cards tournament-legal on any particular date, and English and Japanese releases do not map one to one. Second, the card identifications are partial: "Pikachu ex #54" gives a collector number without a set code, language or rarity, and a shared illustration does not make two printings the same card. If you are buying singles off a buy list, confirm set, number, language and printing on the listing itself.

It is also worth saying that a buy list is a market opinion, not a deck-building plan, and that booster packs are a collecting and entertainment product rather than a reliable route to the 60 cards a tournament list needs.

Preview content frequently carries commercial relationships, and AzulGG's description states them openly. The video description includes sleeve and deck box links and a singles retailer, each with a discount code.

Disclosed affiliate links are normal and they are disclosed. But when a preview names cards you are being encouraged to buy, the honest reading order is: watch the call, wait for the posted lists, then decide whether the call held. The lists are free and they do not have a discount code attached.

AzulGG's pre-event video is The Best and Worst Decks Going Into Baltimore on YouTube, published 18 September 2026; the description lists chapters beginning with a decklist segment followed by two games. Celio's Network's read is Last Minute Baltimore Regionals Metagame Thoughts on YouTube. AzulGG links deck lists and merchandise from azul.gg; those are the creator's own postings rather than event-verified lists.

When you go back to grade either call, do it with the official standings and the posted decklists open in another tab, and give each archetype's conversion rate the sample-size discount it deserves. A preview that got the top three archetypes roughly right is doing real work. A preview that named one deck that spiked is doing something closer to what a weekend of variance does on its own.