General CBS
- Thread starter James Stevenson
- Start date
Onderzeeboot
Ecstatic Orb
Hahahaha, the page break makes this the perfect troll XD
I actually switched some of green's ramp from 2MV to 3MV in my current cube, because ramp can easily take up so much space space in 1-2 MV that it crowds out everything else in green. I'm trying to some aggro and madness elements to green after reading the thread below, and I think a slower ramp will still be fine. The three mana guys do more interesting stuff to support multiple archetypes, too.
https://riptidelab.com/forum/threads/what-can-green-do-thats-not-ramp.3363/
Chris Taylor
Contributor
Just don't run green ramp 
If your goal is to let everyone have MV 2 ramp, stick to that for green... but make sure that green ramp spells do other stuff. Lean more towards stuff like Poison Dart Frog than Nature's Lore.
Yes, this is why I haven't run them in a decade. It made any green ramp outside of explicitly Rofellos, Llanowar Emissary / ten-elves-and-a-Craterhoof Behemoth seem quaint by comparison.
Talismans or Signets also indicate to drafters that colors don't matter. The hard part of fixing is done for you, you don't need green for Farseek in that world.
The talismans aren't even that good in my environment. In the past 13 drafts, they have all together been maindecked 28 times and they only won 30 matches. That's barely a 1.07 match win rate. At the same time, my green ramp spells have been played 54 times and collected 82 wins, which is a 1.52 match win rate. I think the reason though is mostly synergy with powerful themes like landfall.
Onderzeeboot
Ecstatic Orb
Could it be that you run a bunch of Talismans, so they end up both in the winning decks and the losing decks? Just wondering!
That would conflict with the (low) amount of matches they have won.
If X goes into a winning deck it gets some matches won. If X also goes into some bad decks, it gets even more matches won.
at my power level the ramp is built into the lands themselves so it's not really a question i've had to reckon with
They are definitely notfor every deck. Also, if they were average, they would be closer to 1.5, right?
Onderzeeboot
Ecstatic Orb
I don't really know how you calculate your numbers, but on second thought, shouldn't a "winning rate" be between 0 and 1? If two decks with a Talisman in them face each other, one wins, and one loses, you've got 2 matches played, and only 1 won, for a 0.50 winning rate. It looks like you are dividing the number of matches won by the number of maindeck inclusions, which is... well, I don't know how you would call that number
For a good calculation you need to know the total number of games played, and you need to know te draws, because those need to be ignored. I don't have those numbers, but let's assume everybody played three games, and they lost the games they didn't win. For the Talismans, 28 maindeck inclusions would translate to 84 games played, of those 30 were won, for a win rate of 30/84 = 0.36. For green ramp, 54 maindeck inclusions translates to 162 games played, of those 82 were won, for a win rate of 82/162 = 0.51. So it looks like Talisman deck are actually losing in your cube, and ramp decks are about even (meaning they lose as much as they win). Of course, that's with me doing a bunch of assumptions, but you get the picture
It's just calculating win percentage relative to going 3-0 or 0-3 instead of every individual match, the number tells you the expected outcome of a deck with talismans is barely above 1-2. It erases the nuance of someone going 2-0 every match compared to 2-1, but, well, ultimately it doesn't matter how many games you lose as long as you win your matches, so in a sense this models reality and is more lenient on decks that might have a bit more variance.I don't really know how you calculate your numbers, but on second thought, shouldn't a "winning rate" be between 0 and 1? If two decks with a Talisman in them face each other, one wins, and one loses, you've got 2 matches played, and only 1 won, for a 0.50 winning rate. It looks like you are dividing the number of matches won by the number of maindeck inclusions, which is... well, I don't know how you would call that number
Onderzeeboot
Ecstatic Orb
But the difference can be huge. Obviously the following calculation is going to be skewed to the max, but say the Talisman decks got a 1-2 in all of their losses, but won their matches with 2-0, and the Ramp decks instead won their matches with 2-1, but their losses were 0-2's.
Talisman: 30/84 matches were won, meaning there were 84 - 30 = 54 losses. This adds up to (30 * 2) + 54 = 114 games won, and 54 * 2 = 108 games losses. This gives us a game win rate of 114 / (114 + 108) = 0.51.
Green ramp: 82/162 matches were won, meaning there were 162 - 82 = 80 losses. This adds up to 82 * 2 = 164 games won, and lost 82 + (80 * 2) = 242 games lost. This gives us a game win rate of 164 / (164 + 242) = 0.40.
Now, admittedly this scenario is very unlikely, and likely even mathematically impossible, considering that a number of games should be mirrors. However, it does show that in the most extreme and lopsided case, the game win rates of two decks can completely flip in when compared to the match win rates of those decks. Looking at the actual game win rate may give a much better idea of how good certain cards (or at least the decks that play them) perform than merely looking simply at match win/loss statistics.
See, @Mown is right in terms of how I've been doing it. I check how many matches a card won per draft, which means for one use (either maindecked or brought in from the sideboard in a relevant manner) it can get 0, 1, 2 or 3 points.
You @Onderzeeboot are probably right though, that game wins would be more accurate. Wow, thanks, now I have to invest many hours once again. But you explain to my girlfriend what I'm doing this time
You @Onderzeeboot are probably right though, that game wins would be more accurate. Wow, thanks, now I have to invest many hours once again. But you explain to my girlfriend what I'm doing this time
But now I seriously have a problem here. See, my document looks like eight pages of this:
The first column is the matches won, the second the number of drafts the card was used in, and then the quote I calculated from the first two numbers. Now, I could of course just re-do this table with game wins per draft, but I didn't keep track of game results for the most part. I could probably recall the two most recent ones, maybe just the last.
How can I convert that first number into one that gives me an estimation of games won? Maybe some kind of formula?
So, Archaeomancer here had 4 uses, that means he was there for 12 matches. If a match is, on average, 2.5 games, that would be 30 games. Since he won 5 matches, he must have been part of at least 10 game wins. But then, there are 20 games left, where I have no certainty how many were wins or losses. If we say, half the losses are 1-2s, half are 0-2s, then ... that would add to 5 more wins in those lost matches. So I'm adding a quarter of the remaining games. So now, the card would have 15 wins out of 30 games. Is this even a useful way to do this?
The first column is the matches won, the second the number of drafts the card was used in, and then the quote I calculated from the first two numbers. Now, I could of course just re-do this table with game wins per draft, but I didn't keep track of game results for the most part. I could probably recall the two most recent ones, maybe just the last.
How can I convert that first number into one that gives me an estimation of games won? Maybe some kind of formula?
So, Archaeomancer here had 4 uses, that means he was there for 12 matches. If a match is, on average, 2.5 games, that would be 30 games. Since he won 5 matches, he must have been part of at least 10 game wins. But then, there are 20 games left, where I have no certainty how many were wins or losses. If we say, half the losses are 1-2s, half are 0-2s, then ... that would add to 5 more wins in those lost matches. So I'm adding a quarter of the remaining games. So now, the card would have 15 wins out of 30 games. Is this even a useful way to do this?
If you wanted to switch to games, but still use the data from these matches, you could figure it's a 50-50 chance at a 2-0 or a 2-1 and count the match wins as 2-0.5. Conversely, a match loss is 0.5-2.
You could also start over.
Are these stats based on the card being in the deck that won? I assume you're not tracking whether or not the card was drawn or cast, as that's far more difficult.
You could also start over.
Are these stats based on the card being in the deck that won? I assume you're not tracking whether or not the card was drawn or cast, as that's far more difficult.
Yes, I'll do it like this. And round to full numbers where needed. This way I can also get win percentages:
4 cards done, 496 to go.
Yes, exactly. It's just not possible to have a good time with friends, play and joke around AND also keep track of all the details. No GIHWR here.
4 cards done, 496 to go.
Yes, exactly. It's just not possible to have a good time with friends, play and joke around AND also keep track of all the details. No GIHWR here.
Help me get some quick data. How many cards are in your cube that you'll usually cast at mv>6? Meaning something like Myr Enforcer or Virtue of Persistence don't really count, as they're cast for mv<7 and mv=2, primarily.
I want some idea of how many a normal cube runs so that I can see how far to push my cube.
I want some idea of how many a normal cube runs so that I can see how far to push my cube.
None. Hardcasting in my cube caps out at 6
Chris Taylor
Contributor
Yeah I think I cap at 5
I think Nexus of Fate is my only one.
Forgot about that one. So one for me.
I run like one 6 and one 7 (technically two more with Elvish Aberration and Krosan Tusker but those have alternative modes, and I guess Honored Hydra sort of does too. But Avenger of Zendikar is a pure example) in green as ramp payoffs and then cap out at 5 in every other color. I think you'd need to make games pretty slow to go much further (like original RoE did with all the defender nonsense and nerfed aggro).
