'Duplicate' Magic - Could It Work?

[Kieran Symington]

Everyone hates variance - it's the one thing most people wish wasn't part of the game. Equally, everyone wants to know which of their friends is the best player. So I've been giving some thought to doing an experiment with Magic similar to how they play Duplicate Bridge - basically have a bunch of people play games against each other where the starting conditions are exactly the same - including the exact deck order - then awarding points based on how far the result deviated from the normal one (some matches would be clearly going to one side or the other, but the player who lost 'best' would get as high a score as the one who won 'best'. There are a lot of things I'd need to sort out before trying it - mostly the scoring system and whether to use draft or cube decks for the games - but would anyone be willing to give it a try, bearing in mind it will probably all end in tears!

[Dan Barrett]

I think duplicate sealed might be a better place to start? Potentially allowing players to mutually agree to "write-off" games decided purely on mana flood/screw?

[Simon Baldwin]

What happens if you need to shuffle the pre-ordered deck?

There are potentially two things to be "testing" - if you go for the duplicate sealed, then deck building is being tested more than play, whereas if using pre-built and pre-"randomized" decks, then it is almost exclusively play skill. 

I think in both cases, there is a chance that it will not only be play skill being testing, but also playing around a known quality (i.e. a card pool or deck order)

[Russell Davies]

Sounds like a lot of effort and complexity - designing a new game with less inherit variance might be easier  ;)

Nice idea, though.

[Thomas David Baker]

You'd have to remove all shuffle effects from both decks as well as anything like Vendilion Clique.  Then you could potentially make the decks considerably smaller if you're prepared to assume the game won't go longer than a certain number of turns.

If you make the decks I'll play :)

If you make both decks exactly the same that's probably too weird as then you always know the contents of your opponent's hand.

[Simon Baldwin]

If both decks are made in advance (even if different) and both keep the same order, you will know the order of cards for the opponents deck if you play both decks in at least one game each, which would make sense as you can try both sides of the match. 

Thomas David Baker said:

If you make both decks exactly the same that's probably too weird as then you always know the contents of your opponent's hand.

[Kieran Symington]

The original idea was to make eight pre-constructed decks where I knew the exact order of every card, get eight people together and split them into A and B sides. Then each A player would play deck A in each of their games against the B players playing with the B decks. After all the games were played, I could then award points based on how well each player did with each deck. Currently the only scoring method I can think of is how many turns the game was won in - the longer you survive with the 'bad' deck the more points you get and the less your opponent would. Ties could be settled by difference in life totals at the end of the game.

[Thomas David Baker]

You think the same deck will win in each case?  If you prearrange reasonable draws between evenly matched decks I bet you don't get the same result from each A v B game.  No need for tiebreaks, just best A and best B player.