Magic The Gathering: Arena Dev Tracker

Mirrors are a tiny percentage, the system has a strong bias against them.

Edit: Correction, there are a non-trivial number of Cat mirrors due to more people selecting Cats than Dogs.

But anyway, the 51%/49% holds when limited to Dogs vs. Cats games.

The two decks win percentages are 51% and 49% so far, so it's very possible to win or lose with either.

Yeah, that Uncollected checkbox can really trip you up. We need a better solution for saving whether a user wants it on or off by default. There are actually 3 user stories with competing needs:

New User: Wants Uncollected off because they don't own very many cards and they want to make a deck with the cards they have.

Advanced User: Wants Uncollected on because they have a large collection and will craft what they are missing.

Uber Power User: Has all of the cards and wants Uncollected off to hide Planeswalker deck cards.

Thanks for posting your experience, it's something we are working on!

I don't think you can import it even if you did have it, but I might be mistaken. I will see what I can do about that.

I didn't write the original system, but I did write the simple text decklist importer. In general, I like having both. Relying on something to be spelled 100% correctly over 8 languages is pretty brittle. The CNs are written right on the card, so there really shouldn't be that many mistakes with them. Mostly, I use the simple text one myself because you can hand-edit it easier.

Import doesn't return styled cards right now, I can see about handling that situation and replacing with unstyled versions depending on ownership.

Set + CN is better for foreign languages because it will ignore misspelled names as of the 1.09 release. The only problem I see is the using promo CNs (full art cards, etc.), we do track them by CN, but there is some work to make the import treat them as actual cards.

In general, these are the formats that will work with import:

Number + Name:
4 Basri Ket

Number + Name + (SET) + Collector Number:
4 Basri Ket (M21) 7

You can misspell the name if everything else is right:
4 Bazzri Khet (M21) 7

-------

Here's what won't work:

Name + only SET or CN
4 Basri Ket (M21)
4 Basri Ket 7

Correct name, wrong SET/CN:
4 Basri Ket (M20) 78

Xs after the number:
4x Basri Ket (M21) 7

Collector Number is for a promo card or different style:
1x Basri Ket (M21) 280

------

There are two formats for deck sections. Each section must be separated by an empty line.

Unlabeled (or incorrectly labeled)
1st section is Maindeck, 2nd is the sideboard

Labeled (above section)
...

There are some changes we need to make to this scene, that I hopefully have some time for soon.

- Remove the colors. Regardless of how you read the colors, we are introducing bias by making them two different colors. That and our game is conditioned on the "orange" button being the one we are highlighting for you to press in general - so it's not a good way to get an unbiased sample.

- Also, the order should be randomized. Having the button on the same side every time also introduces bias, that's why presidential ballots are randomly ordered.

I certainly believe it. I'll take a look on Monday. #wotc_staff

try this:
t:rat -t:pirate -t:angrath

http://dtwtranscripts.blogspot.com/2013/05/52413-episode-35-blue.html

This is Maro's reasoning:

" Oh, real quickly. Why do we do that? I’ve talked about this in my column, but for those that haven’t heard me say it, when first Richard made the cards, the problem was that black and blue both start with B. So why not go to L then? Well, land—which we used to—card colors are used to signify the border, the kind of card it’s on. And so land has its own border. So L is for land. Well why not go to A for black? Well, A is artifact. And so U was the first we could get to in either of them, so we ended ...