PT: Why teams splinter into different decks

From Patrick Sullivan's PTDGM report:

Ten years ago*, Constructed Pro Tours were usually pretty well established formats, and the lead time for testing was pretty long. The teams I worked with usually settled on a deck with some weeks to spare, giving us plenty of time to tune sideboards, work on the mana, etc. Given the short lead time Pro Tours have now, that's simply impossible. So while we were largely undecided, most of the team seemed pretty unfazed by that; they had experienced it before. For me, it ran counter to expectations, and it was adding an appreciable amount of stress to the whole thing.

Here we are -- short lead time. Three weeks between Dragon's Maze prerelease and PT start time, new cards legal right away instead of staggered by a month, and no previous GPs in block constructed. So instead of just knowing the format from a bunch of previous events, players rely more on guesswork and taste / personal experience to figure their decks out, and Channel Fireball puts three different decks in the top 20.

It might also be true that most of the players on Channel Fireball have jobs, and all of them interrupted their testing to go play Modern at GP Portland a week earlier. "Three weaks for a pro team to break a format" isn't necessarily three weeks of all the players testing decks. I'd be interested to see if more teams start skipping GPs in the weeks before a pro tour.

PT: PTDGM report listings

Known PT Dragon's Maze reports, by finish:

1 -- Craig Wescoe
6 -- Andrew Shrout
7 -- Matej Zatlkaj
9 -- Reid Duke (premium)
10 -- Eric Froehlich
11 -- Luis Scott-Vargas
13 -- Gabriel Nassif (premium)
18 -- Brian Kibler (premium)
30 -- Larry Swasey
31 -- Matt Costa (premium)
32 -- Patrick Sullivan
40 -- Patrick Chapin (premium)
72 -- Conley Woods
77 -- Brian DeMars
209 -- Jim Davis
235 -- Paulo Vitor Damo da Rosa
371 -- Bill Comminos

I'll keep updating this as I discover more reports. Feel free to alert me to more in comments -- non-English is fine.

PT: Ranking the PT DGM teams

Using Lauren Lee's team statistics and the top decks page from the coverage, here are the known team performances from PT Dragon's Maze, ranked by average points in the Swiss and noting the teams' deck choices.

team	avg pts	decks	top finish
South Florida	29.2	Naya	David Sharfman
European Union	29.2	Esper	Matej Zatlkaj
Channel Fireball	27.3	Red, Esper, Junk	Josh Utter-Leyton
StarCityGames	24.9	Bant	Reid Duke
Cute By Comparison	23.4	Bant, UWR	Michael Bernat
TCGPlayer	21.6	Red, UWR, BWR	Raphael Levy
Legit	20.4	UWR, Esper	Andrew Shrout
Sweden	18.3	Gates, Junk	Denniz Rachid
Wilson Gone Wild	18.3	Esper, GW	Ross Merriam
Luxurious Hair	17.5	GW, Esper, RDW	Craig Wescoe
Latin America	16.4	Naya	Felipe Tapia Becerra
Northern Walkers	15.7	Esper	Andrejs Prost
Australia	15.0		Justin Cheung
Rourix	15.0	Naya	Jose Francisco Silva
Mana Deprived	14.8		Hao-Shan Huang
Netherlands	14.6	Esper	Jan van der Vegt

The "top teams" for the event seemed to each use a single deck choice (Florida Naya, European Precinct Esper, and arguably SCG Gatecreeper Bant), while a bunch of "medium-successful" teams did OK playing different decks. From what I keep reading in the PT reports, Channel Fireball gets consistent success despite splintering into factions playing different decks, but the results from this PT really make "solving a format" look like a much better plan than just "figuring out some good decks."

Players in Lauren Lee's team listings amounted to 40% of the field, included 51% of players who went 10-6 or better and 64% of the players who went 11-5 or better. Here are some pie charts:

Block: Composition of control decks

Here's a chart showing the components of top-finishing control decks at PT Dragon's Maze.

deck	lands	draw spells	removal	disruption	finishers
Andrew Shrout	26	12	15	4	3
Channel Fireball	27	7	15	6	5
Ari Lax	28	7	14	7	4
Richard Bland	27	11	14	3	5
Andrejs Prost	27	8	14	8	3
Matej Zatlkaj	27	8	14	7	4
Dusty Ochoa	28	8	14	4	6

("Disruption" is defined here as counterspells and Sin Collectors. These examples leave out the StarCity Bant deck and the Esper deck with Precinct Captains main; for now we're looking at "heavy control.")

The seven decks in the table are all Esper, except Shrout (UWR) and Prost (straight UW). The Channel Fireball deck is a list that was played to 8-2 by Luis Scott-Vargas and Conley Woods, which has 30 of 33 nonlands in common with Matej Zatlkaj's version.

Observations on the control decks:

Removal: All the decks had 4 Supreme Verdict, and all except Lax had 4 Azorius Charm. All the Esper decks had 3-4 Far // Away, while the non-Esper decks had 3-4 Renounce the Guilds.

Mana: Lax's "28 lands" are really 26 lands and 2 Orzhov Keyrune. Ochoa's 28 lands are really 28 lands. Shrout ran only 26 lands but had 3 Prophetic Prism.

Counterspells: All the decks had 3-4 counterspells, except for Prost's deck which had 8 (Syncopate and Render Silent).

Big spells: All the decks had 4 Jace, 3-4 Sphinx's Revelation, and 2-3 Aetherling. None of the decks had Ghost Council of Obzedat.

4-0 control decks in recent block dailies:

• Youbun playing Swedish-style Maze's End control, using 26 / 6 / 21 / 7 / 0 (i.e., 21 removal spells and 0 finishers outside of the mana-base)
• JIsOrange playing a close match of the CFB / Zatelkaj deck, using 1 Maze's End and 2 Psychic Strike

Block: Composition of aggro decks

Aggro decks at PT DGM:

• The Florida Naya deck played 21 lands / 32 creatures / 7 other (going by Chris Fennell's list), and the creatures included 8 one-drops and 16 two-drops.
• The Rourix Naya deck used 24 lands / 29 creatures / 7 other (going by Jose Francisco de Silva's list and counting "make a token" as creatures). They used no one-drops and 16 two-drops, including 4 Zhur-Taa Druid.
• Wescoe's GW deck played 23 lands / 32 creatures / 5 other. He used 12 one-drops and 12 two-drops.
• Josh Utter-Leyton's Rw deck played 21 lands / 28 creatures / 11 other. He used 12 one-drops and 12 two-drops, and he had Legion's Initiative maindeck over Boros Reckoner.

Aggro decks that have 4-0'ed in the last two published dailies:

• MagicLair playing the same 37 spells as Wescoe
• xMiMx playing 35/37 of the same spells, except with 2 Vitu-Ghazi Guildmage instead of 2 of the Judge's Familiar
• Puddingtime playing 35/39 of the same spells as Utter-Leyton, with 4 Viashino Firstblade instead of 4 Frontline Medic
• Shatun playing 19 Mountains, 33 creatures, 4 Dynacharge, and 4 Mizzium Mortars (similar to Jonas Koestler's 8-1-1 deck from the PT)
• Medvedev playing a sweet Bant tokens deck featuring 4 Lyev Skyknight and 3 Rapid Hybridization

Finance: Price Shakes in PT Week

Via MTGGoldfish, here are the cards that went up in value this week:

These are all cards that saw play at the Pro Tour. Voice of Resurgence was identified before last weekend as the best card in the set, but the "verification" of dominating games on camera boosted it from 29 to 38 tix.

(Only two of top 8, Wescoe and Castellon, actually played GW and got to use Voice of Resurgence. GW might not make it in standard, so maybe the value is to do with eternal formats and Birthing Pod Voice in Modern.)

8 of the 10 winners on this list are from the RTR block. A couple of them (Maze's End and Jace Architect of Thought) are unlikely to matter much outside of block constructed.

Ral Zarek takes the same drop Vraska and Domri Rade took after the release hype. New Planeswalker turns out not to be powerful. The other losers are mostly cards rotating from standard after this summer, plus Sphinx's Revelation.

LSV's top 16 report claims Sphinx's Revelation wasn't overpowered in the PT format. There were 18 Sphinx's Revelation in the PT top 8, but they were sort of kept in check by Aetherling, Dispel, and Sin Collector. Probably Sphinx's Revelation (and Rakdos's Return) will hold value; the PT results just reflect a format without mana acceleration.