Frank Analysis – Abzan and Affinity, Worlds Weekend and Milan

I’ve played a lot of Magic over the last two weekends, so today, I’ll share some stories and decks from the World Magic Cup in Nice and the Modern Grand Prix in Milan.

Team Sealed at the World Magic Cup

I had the honor of leading the Dutch WMC team this year, and the team was excellent. From left to right: Roy van den Oever, Frank Karsten, Iso Been, and Thomas Hendriks.


First up was Team Sealed. When we practiced, we often ended up with a Mardu token/Warrior deck that abused powerful synergies, a Temur morph/tempo deck with a great mana curve, and an Abzan deck that was based around the plethora of Duneblasts and Siege Rhinos that we opened every single time.

Generally speaking, the way we split the colors was dictated by the multicolor cards, synergies, lands, creature counts, and mana curves in our pool. Those were the five most important factors that guided our decisions. If we found three decks that allowed us to play our most powerful multicolor cards and use the strongest synergies while providing a reasonable mana base, mana curve, and a good mix between creatures and spells for each deck, then we were happy.

Simon Goertzen taught me a useful way to lay out the pool. We used it at the World Magic Cup and I recommend it going forward, especially for team tournaments. The system is to make a 3×5 grid of three rows by five columns:

• In the top row, you lay out every color, but you order the colors such that enemy colors are adjacent to each other. So, green-black-white-red-blue.
• In the middle row, you lay out all of the enemy color cards and lands in between their respective colors. So, Ride Down and Wind-Scarred Crag go in between white and red.
• In the bottom row, you lay out all the allied color lands, tri-color cards, and tri-lands for a clan below that clan’s enemy color. So, Siege Rhino, Blossoming Sands, and Sandsteppe Citadel go below black. (Allied color lands can go in only one clan.)

This way, it is easy to see all of the gold cards and lands in one easy glance.

After using this layout to facilitate a discussion about the possible combinations and eventually agreeing on one that seems to be the best, it can be dangerous to get locked into that way of thinking. We had one practice pool in Nice where we agreed to start by building a U/G morph deck with double Secret Plans, a B/W Warrior deck, and a Jeskai deck with the leftover cards. The first two turned out to be fine, but the Jeskai deck didn’t have enough creatures and spells and didn’t have a good mana base either. So then we turned Jeskai into Temur, which had a much better creature and land base. But we failed to re-evaluate our initial assumptions at that point.

You see, the two Secret Plans would actually be much better in Temur. But since we already had a U/G deck with all the morphs, that option didn’t cross our minds. Until Ivan Floch dropped by and asked, “Why don’t you make a controlling Temur deck with the red burn spells as removal and Secret Plans for value and an aggressive U/G deck with all the cheap creatures, bounce spells, and pump spells?” We swapped a bunch of cards around and, indeed, the result was much better.

So with Team Sealed, an unbiased outside perspective can be invaluable. We actually considered blindfolding one player on the team, who would keep his hands over his ears for the first thirty minutes of the build. Afterwards, we could take off the blindfold, allowing him to give a completely fresh perspective on the pool, uninfluenced by prior conceptions.

Ultimately, this was deemed to be a little too drastic, so we just built in a sanity check at the thirty-minute mark where we would all try to re-evaluate all of our initial assumptions.

It worked: Team Netherlands ended up going undefeated in the Team Sealed portion of the WMC.

Unified Standard at the World Magic Cup

In Unified Standard, you essentially have to build three decks from a binder containing four of every card in Standard. So you can’t play Abzan Midrange, Green Devotion, GB Enchantress, G/R Monsters, and/or Sidisi-Whip alongside each other because all of those decks want 4 Sylvan Caryatid and 4 Courser of Kruphix. The other cards that saw the most overlap were Hero’s Downfall, Drown in Sorrow, Thoughtseize, Seeker of the Way, Elvish Mystic, Goblin RabblemasterLightning Strike, Wingmate Roc, Magma Spray, Llanowar Wastes, Temple of Malady, Battlefield Forge, Temple of Triumph, Polluted Delta, Flooded Strand, and Wooded Foothills.

This made for an interesting puzzle of how to find three good Standard decks with as little overlap as possible. For the top ten decks in Standard, I made a spreadsheet that indicated any overlap between them, and then used a computer algorithm to enumerate over all three-deck combinations. This yielded around 60 combinations with no substantial overlap, so we had to narrow it down. Thomas liked Mardu a lot, so we zoomed in on the combinations that allowed for Mardu:

• U/B Control + Abzan Whip
• U/B Control + Abzan Midrange
• U/B Control + Abzan Aggro
• U/B Control + G/B Constellation
• U/B Control + Mono-Green Devotion
• U/B Control + Temur Aggro
• Abzan Aggro + Mono-Green Devotion
• Abzan Aggro + Temur Aggro
• Abzan Midrange + Temur Aggro
• Abzan Whip + Temur Aggro

No one on the team was a dedicated control player, so U/B Control was out by playstyle preference. We liked Sylvan Caryatid and Courser of Kruphix over Herald of Torment and Rakshasa Deathdealer, especially since we expected many midrange decks, so Abzan Aggro was out as well. Finally, even though Abzan Whip may actually be a better deck than Abzan Midrange, I was afraid of getting too many draws with Abzan Whip because the life gain can draw out games and it’s a team event where everyone takes more time than usual because they want to discuss plays with teammates.

So, we eventually decided on Mardu + Abzan Midrange + Temur Aggro.

There was actually very little overlap between these three decks. We would’ve liked to have access to 5-6 Lightning Strike and 5-6 Thoughtseize, but that was it—apart from that, our decks would’ve been the same if we would enter an individual Standard event. There was no overlap between Caves of Koilos and Temple of Silence because both Mardu and Abzan Midrange only wanted to play two of each. There was no overlap in Elvish Mystic because Temur Aggro only wanted to play three and Abzan Midrange only wanted to play one.

Lightning Strike and Thoughtseize are nice to have, but are far from essential. Lightning Strike doesn’t even kill all that much in Standard, and Thoughtseize can take up a turn where you’d rather add a threat to the board. Besides, the Abzan deck could play an extra Nissa, Worldwaker and an extra Erase in the sideboard instead of Thoughtseize and the Mardu deck could run an extra Magma Jet and an extra Stoke the Flames in the main deck instead of Lightning Strike. We felt that these small changes didn’t really affect the decks in any significant way and were happy with our configuration.

Since I had the most experience with Abzan Midrange, I was tasked with playing that deck:

Abzan Midrange – Frank Karsten, WMC Unified Standard


My version was a throwback to the midrange-style deck with Fleecemane Lion and Wingmate Roc. I like this pro-active style better than the slower one with more removal and planeswalkers. Some cards choices of note:


It’s particularly great at turning Sidisi into a lowly Hill Giant. With the rise in Whip decks, I felt that Anafenza was well-positioned for maindeck inclusion. Once you have Anafenza, maxing out on Fleecemane Lion and Wingmate Roc makes sense.



This was a late addition to the sideboard, and it served me well. Against Whip decks, you need an answer to Hornet Queen, but drawing too many reactive cards like Drown in Sorrow or Bile Blight doesn’t fit with the role you have to take in the matchup, as your late-game is inevitably worse. Doomwake Giant is a pro-active card that can deal with Hornet Queen, so it fits the game plan. It was also great against token decks with Raise the Alarm and Hordeling Outburst.

Speaking of sideboards, we made sure to discuss and write down broad sideboard plans before the event started. It’s a waste of time to discuss “how should I sideboard against Mardu?” with teammates during the event when you can do that beforehand.

At the WMC, a few people expressed interest in my sideboard sheet, and I promised them that I would add it to my next article, so here it is. I hope my handwriting is decipherable, but if you want to play my exact deck at an upcoming FNM, then feel free to print out this picture and refer to it in between games.


If you’re interested in Iso’s Temur Aggro deck or Thomas’s Mardu deck, you can find them here. They’re fairly standard. The Temur deck put all Temur Charms in the sideboard because it’s unspectacular and mana-inefficient except against Hornet Queen decks.

We did well in Unified Standard on Day 1, but then failed to win enough matches in the second pool on Day 2, so eventually we stranded in the Top 16. A tad disappointing, especially for Iso, Thomas, and Roy (who would’ve snatched Pro Tour invites with a Top 8 finish) but we played well and had a great time, so I can’t complain. The tournament was a lot of fun, and I hope to lead the Dutch team next year again.

“Danish Magic” After the World Magic Cup

On Saturday night after the WMC, Olle Rade introduced me to a booster-battle variant that he called “Danish Magic.” The rules are as follows: Both players crack open a booster and, without looking at the contents, they take out the token and land and shuffle the rest. You start with zero cards, draw a card per turn, don’t lose to decking, and have infinite mana. This makes the format very similar to infinite-mana Magic. But the catch with Danish Magic is that you have to play like the Danes do. Or at least, how in the minds of some Swedes the Danes tend to play.

In Danish Magic, you have to play a card at the first possible opportunity. You have to activate any ability at the first possible opportunity. You have to attack whenever you can, and you have to block whenever you can. Basically, there are almost no decisions to make at all (except for which creature to block or which creature to target with a kill spell) so the game is almost completely random, but it was still a lot of fun to see how this rule set worked out.

Olle’s first card drawn was Kheru Bloodsucker. As a result, he had to immediately eat any creature that he drew. This led to a rather suicidal, amusing damage race, but the best part came when Olle drew Kill Shot and, to the hilarity of onlookers, was forced by the rules of Danish Magic to kill his own attacker, only for me to reveal that I had had just drawn Cancel and was forced to counter it! It was a beautifully bizarre game, and I’m looking forward to playing more in the future.

Of course, the Danes won the World Magic Cup the day after.

Chalice Affinity at Grand Prix Milan

Since GP Milan was the weekend after the World Magic Cup and I had to work in between, I didn’t have a lot of time to test. Hence, I stuck with my trusty Chalice Affinity deck that I built for Grand Prix Madrid:

Chalice Affinity – Frank Karsten, GP Milan

I only made two changes to the sideboard that I had in Madrid:

• I cut Back to Nature because I didn’t expect a lot of Bogles and I added a second Grafdigger’s Cage. There were three Birthing Pod decks in the Top 8 of GP Madrid, and stopping their namesake card is one of the best things you can do in that matchup. Moreover, Grafdigger’s Cage would stop Josh Utter-Leyton’s Jeskai Ascendancy combo deck from unearthing Fatestitcher and playing Gifts Ungiven for Unburial Rites and Elesh Norn. And it’s an artifact, so it turns on my Mox Opal.

• I cut Blood Moon and added Choke. The reason why I originally didn’t have Choke was that I ran into too many Sulfur Falls and the like when I was testing on Magic Online. However, when looking at the deck lists after GP Madrid, I saw that most U/R Delver and Scapeshift lists still contained a ton of Islands, so I made the switch.

Round 4 was incredible. I was paired against Jeremy Dezani, who was on some sort of Martyr of Sands deck. The match can be found in the Twitch replays starting around the 28-minute mark.

In game one, I got Jeremy down to a low life total, but after he made a pillow fort with multiple Sphere of Safety and Ghostly Prison, I didn’t even have enough lands in my deck to ever be able to attack. But Jeremy still had to find a way to kill me. He had fewer cards in his deck thanks to Squadron Hawk and Ranger of Eos, and I had several huge flyers (thanks to Steel Overseer) that could block his Serra Ascendants alongside a Chalice of the Void that prevented him from playing Path to Exile. Now, most of these white life-gain decks play a Mistveil Plains to prevent decking (or maybe some combination of Wrath of God plus Proclamation of Rebirth to break through) but I decided to play on for a while just to see how he would kill me. To my astonishment, Jeremy picked up his cards after realizing that he simply had no way to win.

In game two, pretty much the same thing happened. I could never attack him, but Steel Overseer made too many big flyers for him to push through. At some point, I had activated Steel Overseer so many times that we ran out of dice in the feature match area.


Although Jeremy tried to fight back with a Disenchant, it simply wasn’t enough. Once again, I decked him. “I misbuilt my deck,” he told me afterwards. Indeed, if he had just put one Mistveil Plains in his main deck and one Creeping Corrosion in his sideboard, or anything along those lines, he would’ve easily beaten me 2-0.

After that bizarre match, my luck ran out. I lost a match to Restore Balance, drew an Affinity mirror that took a long time due to Arcbound Ravagers on both sides of the board, beat a Storm deck, and finally lost to Slivers. As it turns out, Harmonic Sliver plus Sliver Hive is pretty good against my deck.

On the GP Milan Format

Due to a public transport and air traffic control strike in Milan, several players were unable to get to the venue in time, so it was announced that on Saturday the tournament would be split up in two shifts: one that started at 9 a.m. and one that started at noon. Moreover, as soon as a player had 21 points on Day 1, he would be dropped and automatically added to Day 2. Day 2 was a new tournament with 7 rounds where match records and tiebreakers from Day 1 didn’t apply.

I appreciate that the tournament organizer tried innovative solutions to accommodate as many players as possible, and I would support a similar effort when there are huge transportation disruptions in the future. However, I was not a fan of the implementation. The format had various difficulties:

Since Day 1 records vanish, the Top 8 is decided on a smaller sample of 7 rounds as opposed to 15. This is a disadvantage to better players and players with byes, whose higher chance of finishing Day 1 at 9-0 is swept away. Moreover, due to tiebreaker math, you can make Top 8 by going 7-2 on Day 1 and then 6-1 on Day 2 if the loss on Sunday comes late enough, but you can miss out on Top 8 by going 7-0 on Day 1 and then 6-1 on Day 2 if the loss on Sunday is in the first round. It’s strange that a player with a 13-3 record can finish higher than a player with a 13-1 record.

In the first, larger shift on Saturday, you could 6-0 and then triple ID to make Day 2 with a 6-0-3 record. I ran the numbers on this and found that, given the number of players on 5-0-1, it was beneficial to do so. Almost all 6-0 players decided to play, perhaps because it gave them the feeling of being more in control, but the probability of being paired down and losing in rounds 8/9 was lower than the probability of losing three rounds in a row. Although few players picked up on it, the incentive implied by the tournament structure was strange.
Some players had booked flights on Sunday evening and suddenly couldn’t make them anymore after a seventh round was added.

I think it would’ve been better if the Day 1 records had not been reset. In other words, if we had the same as when a Grand Prix is split up on Day 1 due to size and merge the tournaments at the beginning of a six-round Day 2. This structure might give a small advantage to players in the second, smaller flight who would have a larger chance of being paired down against a player with a worse record, but I think the disadvantages of the system used in Milan are even greater.

And with that, I come to the end of today’s mixed bag of topics. Happy holidays!


Scroll to Top