Three level draft

Just now, an idea for an interesting draft format for two teams of three players each popped up in my head, which I want to share with you and get some initial feedback on it (obviously, it is completely untested right now). It is a draft format to be used with a cube (it can, with a little adaption, also be played using real boosters, but since it needs the equivalent of 23 boosters – even more, if you use boosters with a relevant number of unplayables – I don’t think that is a realistic option). It uses a card pool of 320 cards (the cube can be bigger, of course, meaning a number of cards does not enter the draft pool).

I will present two slightly different ways of drafting this format. One is simple and elegant, and also well suited for just two players each drafting three different decks, but takes longer (although not as long as the Rochester drafts of old did). The other is a bit less elegant and looks complicated at first glance (it isn’t actually, just a little unintuitive), but is a lot faster and reduces “downtime” for drafters (when they cannot make choices while waiting for other drafters) greatly.

The basic idea in any case is that both teams draft three different decks, which will wildly vary in power level. Afterwards decks of similar power level will be played against each other in matches. Obviously, the team able to take two out of those three matches wins the draft. What makes this draft variant special is how different the decks in these three matches will play: One pair will be very strong, one mediocre and one weak.

In variant A of this format, players are seated as follows: All members of one team next to each other, facing their opponents. Cards will be drafted from one side of the table to the other. Let’s call the teams A & B, and the players sitting closest to the cards A1 & B1, the ones in the middle A2 & B2, and the ones at the far end A3 & B3.

———-A1—– A2—–A3—–
cards >         >           >

The cards of the draft pool are prepared as 40 packs of 8 cards each (totally random shuffling will lead to very extreme distributions, so it makes sense to sort the card pool into subpools containing different colors before and build packs with equal input from these pools instead).

Then a pack is passed to the players. In this variant, I strongly recommend that all cards from the pack are layed out on the table for all players to see, so that a player whose turn it isn’t to pick a card can use that time to think about possible picks. Alternating from pack to pack, the pick order is either A1-B1-A2-B2-A3-B3, or B1-A1-B2-A2-B3-A3. Each player picks a card; the last two cards of the pack go unused and are put in a discard pile (preferably facedown). In the end, each player will have drafted a pool of 40 cards from which to build a limited deck (according to the usual rules, and of course with the addition of basic lands).

It’s easy to see why this produces decks on three different power levels, since players will obviously almost always chose more powerful cards first – passing a card in the hope that it will end up in a teammate’s deck is a really risky proposition, since an enemy drafter will be able to get it first. Thus, by and large, the decks of A1 & B1 will be mainly made up of the strongest 25% of each pack and so on.

This variant takes some time, because drafting consists of 240 picks which have to be made in succession. That’s 2/3 of the time a classic Rochester draft takes, so it’s not THAT bad, but many drafters will probably prefer the faster variant. Note, though, that this variant works really well for two players building three decks each (they just need to put their picks in clearly distinguishable heaps – deckboxes can help here). In that case, the drafters’ downtime is obviously much shorter. Also, the longer playing time per player (they will play the three matches one after another, of course) will relate better to the time drafting takes.

Variant B cuts down on drafting time immensely by allowing players to pick cards at the same time (just like in booster draft, where the number of successive picks is cut down to an eighth). However, to preserve the element of building decks on three levels, players from each team must still pick cards from each booster in order. But if you do this while having players pick simultaneously, there is the issue of one team’s players picking cards before their opponents (either always, or at least 2/3 of the time). To offset this, you obviously have to change the order in which players draft, but you cannot do that too often, or players will again have to pick one after another a lot of the time. So, what’s needed is a balance between a good number of packs flowing undisturbed around the table, and a few breaks to balance the picking orders.

I feel a good solution is the following: The 40 packs to be drafted are separated in three groups of 10, 20, and again 10 packs. Packs in each group will be drafted in the same picking order, group 1 and group 3 in an order where A1 comes first, group 2 in an order where B1 comes first. Also, the picking order is defined such that not all three players from one team get earlier picks than their later opponents, just two out of three. This leads to the following picking patterns:


cards ^  v   >  ^   >  v


cards v  ^   >  v   >  ^

It’s really not complicated: A1 & B1 alternate at getting the first shot at a pack in both patterns, and for each player, they alternately pass to a teammate and to their opponent (A3 & B3 alternate between passing to their opponent or putting the last 2 cards in the discard pile).

Maybe it is even clearer in simple notation:

Pattern 1: A1-B1-B2-A2-A3-B3

Pattern 2: B1-A1-A2-B2-B3-A3

This way, the number of picks made in succession is reduced to 55 (packs still need to pass from one end of the table to the other, so not all cards can be picked at the same time), which isn’t that much higher than in a normal booster draft (42 nowadays, since boosters contain effectively 14 cards). Each player will pick the same number of cards before and after their direct opponent overall, with one team having two players drafting first with picks 1-10 and 31-40 and one player with picks 11-30, and the other team the other way around.

So that’s the idea: Using M12 limited as an example, players at table 1 will end up with decks full of stuff like Fireball, Mind Control & Overrun; those at table 2 will run decks full of quality cards like Assault Griffin, Mana Leak and Gravedigger, but probably without bombs; and those at table 3 will have to make do with a lot of unexciting stuff like Bloodrage Vampire, Slaughter Cry and Brindle Boar.

I believe that this will result in a fresh and entertaining drafting experience, forcing players to adapt to quite different sub-environments. What do you think?

Explore posts in the same categories: Next Level Cube

Tags: , , , , , , ,

Both comments and pings are currently closed.

5 Comments on “Three level draft”

  1. schmirglie Says:

    Ich dachte ich helf dir mal mit deinen Modo-Problemen. Falls du die Tickets noch nicht hast reicht eine Mail an den Kundenservice, das Problem wird normalerweise fix behoben. Wenn man Umlaute und “ß” aus seiner Paypaladresse rausnimmt (oder zahlst du mit Kreditkarte? Dann da irgendwo) sollte das Problem behoben sein. Ansonsten kannst du auch zum Spieler deines Vertrauens mit zu vielen Tix gehen (zum Beispiel mir), und da Tix billiger als im Shop bekommen. Ich vermute mal allzu viele Tix wirst du nicht mehr kaufen, aber nunja… just sayin’.

    • Danke für das Angebot, aber das löst leider mein Problem nicht, denn sie Leute haben mein Geld bereits genommen, wissen aber nichts von einer Bestellung – und jetzt beginnen sie zu behaupten, es sei gar nicht mein Account! Das wird noch höchst lustig…

  2. jashinc Says:

    Ich habe den Artikel gelesen, das System verstanden und sehe worauf das ganze abzielt.
    Was in meinen Augen dagegen spricht:
    Zum einen ist mir das System zu schwierig für meine Casualrunde – da gibt’s schon im normalen Draftablauf zu viele Probleme…
    Zum anderen werden A3 und B3 keine besonders nachgefragten Positionen sein – wer will schon mit dem Bodensatz der Karten spielen?
    Ich glaube allerdings, dass, wenn man eine Limitedumgebung nur für dieses Format selbst gestaltet, mit ausreichend fähigen Spielern ein paar interessante Runden spielen kann – ob das aber die Mühe wert ist…?

    • Das ist die Frage!

      Ach ja: Ich spiele gerne mit dem “Bodensatz” – das sind häufig interessantere Spiele als zwischen stärkeren Decks! Aber natürlich ist das Geschmackssache.

      Im Moment komme ich leider auch nicht dazu, es zu testen. Aber vielleicht probiert es ja irgendeiner meiner Leser schon einmal aus – hoffen darf man ja!

Comments are closed.