Luck can be one of the most frustrating elements of Magic. In Constructed, luck can determine the decks you play against, the skill level of your opponents, your tiebreakers, and the draw of the cards. In Limited, and more specifically Sealed, one also is affected by the cards one is lucky or unlucky enough to receive.
There are some things you can do to try to limit the effects of bad luck. In Constructed, you can play consistent decks with stable manabases. In draft, you can stay flexible with your early picks. However, a big portion of Magic is going to involve luck no matter what you do, and there seems to be little that one can do about it.
Interestingly, other luck-based games have made efforts to solve this problem. In bridge, one set of partners–North/South–play against another set of partners–East/West. The cards that each partnership receives is determines solely by luck. Higher cards are better than lower cards, and the distribution of suits within the partnership is very important. Some days when you play bridge, you and your partner will run into weak hand after weak hand, and therefore have no chance of winning.
In order to avoid the effects of luck at tournaments, most bridge tournaments are run using a format called “duplicate bridge.” In
duplicate bridge, you and your partner are always either “North/South” or “East/West.” If you are part of a North/South partnership, you are evaluated only against other North/Souths, and similarly for East/Wests. In the tournament, each North/South pair and each East/West pair start out by playing Hand 1. After a hand is played, each East/West pair moves to another table and each table then plays Hand 2, moves again and plays Hand 3, etc., until every North/South pair has had an opportunity to play each East/West pair. After the tournament is over, each North/South pair is evaluated only against other North/South pairs, and similarly for each East/West pair. Using this approach, there is no luck associated with the cards you or your opponents receive or the quality of your opposition. All that matters is how well you bid and play your hand as compared to the other people with the same cards playing against the same opponent.
It’s provocative to think about how this might work in Magic. For example, the tournament could require each player to play various decks against various decks. Imagine an eight round tournament. 16 players would be in a pod, with one player in each of the first round matches closer to the front door (“close” players) and one player farther from the front door (“farther” players). In Round 1, all of the players sitting closer to the front door could play Jund against Red Aggro. In Round 2, the “farther” players would rotate to the next seat and the match would be Blue/White Control against Mythic. This would go on for eight rounds with varied decks and combinations. At the end, one could compare across pods to find the “close” winner and a “farther” winner.
This approach would limit the importance of deck construction; it would be a test of a number of skills. First, this would ensure that winning players were well-rounded. While Wafo-Tapa would be a heavy favorite playing Blue/White, could he take Red mages down with Jund? Second, it would ensure that people would not have bad luck with respect to matchups: sometimes one ends up playing three times against one’s worst matchup even if their nemesis is only a very small part of the metagame. Third, if the decks were pre-shuffled, each player would have the same manaflood or manascrew problems, or would find the right card at exactly the right time (of course, shuffling during the game would still introduce some luck). Finally, this approach would better equalize the quality of player’s opponents.
Unlike Duplicate Bridge, there are a lot of logistical problems with Duplicate Constructed Magic. In bridge, there is clearly a metric of success: tricks relative to bidding. The more accurately you predict the number of tricks that you will win, the more points you receive and the more successful you are. In Magic, although there is a winner and loser, the need for more precise tiebreaks is problematic. For example, one could look at life totals, but this would encourage bad play like chump blocking. Conversely, board position would overly discourage chump blocking, even if it was the play that might save you. In addition to these, cards in hand, cards in library, cards in play, the quality of cards in one’s play or library, etc., could all be considered metrics for success in a game. Honestly, I can’t really see how to make this part of the Duplicate Magic idea work for Constructed formats.
Sealed formats, which are normally thought of as the most luck-based, might be ideally suited for a duplicate approach. The usual proposal to make Sealed formats less luck-based is to turn the Swiss part of sealed PTQs into draft. However, the logistics of this are very hard with large events: running 20 or so draft tables with a small staff for judging is impossible.
A second possibility would be to have two card pools: one for the “close” and one for the “farther.” However, players would eventually learn the combat tricks of the opposing sealed deck and play would become routineized and overly draw-based. If you tried to fix that by having people keep rebuilding, it would take too long.
Here is how I think a Duplicate Sealed PTQ could work without too much trouble:
The players in the PTQ are randomly divided into 8 groups; each member of a group gets the same pool. You are competing against the players in the same group as you for a spot in the elimination rounds. Players in the same group do not sit next to each other. Each player in the same group receives the same sealed deck. In each of the seven rounds, the players in a given group play against a player from a different group based on the Swiss system of pairing. One could then put the top one or two players (using the normal tiebreakers) in each group into a Draft Top 16 (Top 8 would lead to x-1s missing out): this would be a completely manageable draft.
Obviously, this does not completely eliminate the luck in Magic. One could still get manascrewed or flooded, or mulligan into oblivion even if they build and play their pool much better than their counterparts. Because every player does not play against every other player, there would still be potential for being paired against better players. However, it would certainly be much less luck-based that traditional sealed formats.
Eliminating all the luck from Magic should not be a goal: there are games like chess for people who hate the idea of luck having any influence on a game. One of the great things about Magic is that there is luck: Joe Schmo can beat Luis-Scott Vargas. Conversely, in chess, an average player will lose to a master 100 times out of 100.
Duplicate Sealed is the perfect compromise. It takes some of the luck away from one of the most luck-intensive formats without completely eliminating it. It allows the best people to feel confident that if they play well and get a little lucky, they will make Top 8, without taking away all the hope from good old Joe.