Hey

I ran a parser over the battle logs from about 3000 ranked matches in the last 14 days (approximately). My main objective was to get stats on win rates depending on you take the first turn ("white"), and resource meta balance.

I haven't bothered to format the results, so you find them here: http://pastebin.com/tbXj0vzV

Some explanations:

"DECAY: 47.77475, as white 50.760452 (total: 1101)" means: Out of all games where a decay deck met a non-decay deck, it won 47.7% of the matches. It took first turn in 50.7% out of matches it won.

"SPECIAL" means that the deck resource couldn't be determined (at least 3 in two resources, and neither being more than double than the other).

The list of numbers are the turn counts where the resource won and lost. Note that it includes surrenders, which explains a lot of the early wins. Matches that were shorter than 5 turns were discarded.

Scrolls are counted by the number of games they appear in either player's hand. Even if a player plays Burn 7 times in a game, it will only count as one in the statistics. Because I also counted which scrolls appear on the field, certain scrolls get misleading stats (for example Beast Rat, as it's counted when Rat King is played).

Scrolls' "win rate" is the number of games it appeared in and won, versus number of games it appeared in and lost. A scroll can count for both a win and a loss from the same game if both players had it in their hands.

Conclusions?

There's a small benefit of taking the first turn, especially for Energy players (not sure why). Energy is over-represented right now due to the current meta.

To get better results I would need to crossreference the matches with the rating of the players, because I suspect a lot of the low-performing scrolls come from preconstructed decks.