General CBS
- Thread starter James Stevenson
- Start date
I think it al depends on the unmorph cost. Borph is very doable but boring if the unmorph is hefty.
1/1 for one are also okay. The thing is, morph demands a special environment. At least if one wants that the morph is important and not that morph is on the card but almost never used. Like on the void age or cat. There it is often better to hardcast than to morph it.
1/1 for one are also okay. The thing is, morph demands a special environment. At least if one wants that the morph is important and not that morph is on the card but almost never used. Like on the void age or cat. There it is often better to hardcast than to morph it.
Chris Taylor
Contributor
There's something to it being a 2/2 because of how (pound for pound) they contribute to a game, where as the average 1/1 for 1 is just not worth the cardboard it's printed on.
It works better on the assumption they're going to be turned face up, but you're never putting off color morphs in your deck to distract/contribute (however marginally) to a gamestate
It works better on the assumption they're going to be turned face up, but you're never putting off color morphs in your deck to distract/contribute (however marginally) to a gamestate
1/1 for 1 makes morphing it very low-commitment, the decision of whether or not to play it face down becomes a lot less interesting. For most cubes I think a bear is fine, if you have something that looks more like retail limited you could experiment with 2/1 for 2, or 3/2 for 3, riptide lab's favorite statline.
I'd honestly reach for Goblin Cavaliers as my base size for morphs before I reached for Grizzly Bears.
Though that gives me an (overly cute) idea: a morph-focused cube where you can draft alternate stats for your morphs (in the form of vanilla/french vanilla creatures). The details of how that'd work are kinda hazy, but implementation shimplementation, amiright?
Though that gives me an (overly cute) idea: a morph-focused cube where you can draft alternate stats for your morphs (in the form of vanilla/french vanilla creatures). The details of how that'd work are kinda hazy, but implementation shimplementation, amiright?
Last edited:
finally a use for the Unfinity stickers!
When you open a booster pack with Timeless Dragon, you can't help but think "hey, a 5 mana 5/5 dragon, that's not bad!" but then you keep reading and realize the incredible flexibility on offer. Excellent modality is present at first blush, but what's even better is when you look through the pack more and figure out more neat things you can do with it, and appreciate how you have a great beater that loves to just sit in your graveyard.
This monster tells you exactly what to do with it: play instants and sorceries and get rewarded in a huge way. It's a tense mini game that doesn't add unnecessary complexity, a 2-drop that begs removal or an increase in aggression. It has a ton more play than that of course, but that baseline is delightful and very blue.
I've talked about Triarch Praetorian a lot here, but as someone who used to drool over the possibilities of getting in with a Skulking Ghost, having an efficient artifact beater that can draw cards if it's failed its primary mission represents a game plan I can immediately get behind.
Inti, Seneschal of the Sun is everything I want in a Magic card. Evocative art, flavor, name, etc. Powerful and practical. Easy to grok but has a ton going on. Your opponent will (rarely) get randomly messed up by not reading it fully. It makes you wonder what else you can make it work with -- you can tell immediately that it's not a downside by any stretch, but it gets the gears turning to figure out how to best take advantage of it.
Priest of Titania is a classic that visually evokes the beauty possible in a fantasy board game and, like Thing in the Ice, tells you what to do with it. The exciting thing here is the wonder and potential of "what can I possibly do with all that mana??" New players love the tribal aspect -- the flavor added to their draft makes them feel like they have a solid grasp of what they're supposed to do as well -- and veterans will smile at the nostalgia evoked from this Urza-era wonder.
What can be said of Scrapwork Mutt that hasn't been said already? A dreamboat, a darling, a pal. Maybe dogs are better than cats after all.
I don’t normally pay attention to the advertising on this website, but I was pleased to see this appear…
As a fan of cubing, I’m clearly the kind of dude who needs a cubesat propulsion system for my satellite.
As a fan of cubing, I’m clearly the kind of dude who needs a cubesat propulsion system for my satellite.
A really creative, flexible one-drop that stays relevant and rewards deckbuilding choices.
A creature with cool bonus options whose "base" mode isn't totally awful. So many old Kicker creatures were just Gray Ogres who also made your opponent discard a card for an extra 3 mana or something. I like the shift into Good Cards that can be Very Good, and I really hope we get some new Bestow creatures someday that show this shift. Unfortunately, they went a little insane with Evoke, but hopefully they'll find a sweet spot with that mechanic eventually.
Cards like Luminous Broodmoth have a bad habit of costing 6 mana. I really think 4 is the right spot, because the game isn't already over by the time you cast it. Also: I'm a big fan of ability counters. Titanoth Rex could be on this list, too.
A personal favorite since I pulled it out of a pack like 20 years ago. This card is just great, and was responsible for many victories and level-up moments for me. Would love to see an uncommon reprint of this guy.
Last edited:
Saw this on the Cubing Facebook group.
Love the idea, hate the execution.
I don't have Photoshop any longer (a recent development), so it took me much longer than I had planned for to make this:
But hell yeah. Will print out a few of them for the next session.
Love the idea, hate the execution.
I don't have Photoshop any longer (a recent development), so it took me much longer than I had planned for to make this:
But hell yeah. Will print out a few of them for the next session.
QR codes are very dependant on contrast so you might want to print the code itself on white!
How this big ass whale only a 5/5?
If you assume it has a similar eye to body ratio of a blue whale and that the perspective makes the eye approximately the size of the back of that ship, it's ~1150 meters long.
Blue whale 4-6 inch eye
Blue whale 70 feet long
Back of ship 20-25 feet
(~70 feet / ~6" = ~150) * 25 feet = 3750 feet = 1143m
Blue whale 70 feet long
Back of ship 20-25 feet
(~70 feet / ~6" = ~150) * 25 feet = 3750 feet = 1143m
So, dumb question, but...
Is there a way to scrape the ELO numbers from Cube Cobra programatically? I'm currently taking a text mining course and decided to use Magic cards for the final project (because Scryfall's API is great). The issue, though, is that my big "research" question is whether or not you can algorithmically determine how good a card is just from its oracle text (my assumption is no, but why not try, right?), which runs into the issue that Scryfall doesn't really have any good metrics for how "good" a card is other than price data and EDHREC rank, which aren't the best metrics for card quality in my opinion? I mean, I'll use them if I have to, but I'd love to have some EDHREC rank equivalent for other formats.
Cube Cobra has that in the form of ELO, of course, but there's no way in fuck that I'm going to manually tag thousands of cards for a class project if I don't have to.
(Also, if anyone can think of another website that has a similar ranking for various constructed formats — or, better yet, for individual sets — I'd love to hear about it.)
Is there a way to scrape the ELO numbers from Cube Cobra programatically? I'm currently taking a text mining course and decided to use Magic cards for the final project (because Scryfall's API is great). The issue, though, is that my big "research" question is whether or not you can algorithmically determine how good a card is just from its oracle text (my assumption is no, but why not try, right?), which runs into the issue that Scryfall doesn't really have any good metrics for how "good" a card is other than price data and EDHREC rank, which aren't the best metrics for card quality in my opinion? I mean, I'll use them if I have to, but I'd love to have some EDHREC rank equivalent for other formats.
Cube Cobra has that in the form of ELO, of course, but there's no way in fuck that I'm going to manually tag thousands of cards for a class project if I don't have to.
(Also, if anyone can think of another website that has a similar ranking for various constructed formats — or, better yet, for individual sets — I'd love to hear about it.)
It's an interesting idea! My first thought is that the best way to rank cards for power level across formats would actually be price, once you adjust for date/rarity of last printing, format legality, etc. That data is maybe even a bit more granular than ELO or EDHREC rating, and forces voters to put their money where their mouth is.
My only hesitation about price is that, well... how do you handle cards like Lightning Bolt, which are fantastic but only cost ~$1 because they've been printed into the ground? Would my attempts to account for that artificially inflate the value of all of the random commons that get reprinted all the time because they play well in Limited but aren't actually all that good?
The thing that I'm kinda bummed about, though, is that Scryfall's API doesn't let you access any of the tags from the tagger project, because that's where a LOT of the real power is.
The thing that I'm kinda bummed about, though, is that Scryfall's API doesn't let you access any of the tags from the tagger project, because that's where a LOT of the real power is.
You'd have to play around with your terms, but even a linear correction for number printings seems a fine first-order approximation to me. There have been 55 printings of lightning bolt!
If you add in a couple more terms past your text mining like so
Value = baseline + (your word term) + (b1 * number_of_printings) + (b2 * years_since_last_printed) + (b3 * rarity_last_printed_at)
my guess is that you'd do an okay job controlling for these factors. You could also try it as an exponential and see if that gets you a better fit. I'm also not at all sure that using a player opinion-derived power metric will regress out the number of printings/extraneous factors as much as you seem to think it will.
At any rate, I think that the data is easy enough to access that you could run the correlations on both. It might yield some amusing discrepancies!
The thing with value is that it is artificial. A deck with rares and uncommons will have that the rares are more worth money wise, while they do not necessarily are better. That is why I proposed to look at the top 8 deck lists of major tournaments in eg legacy, modern, and standard. And please do not forget the banned lists (except ante for legacy) since those cards are on that list for a reason.
If you succeed, you could use it to predict the good cards in spoiler season…
If you succeed, you could use it to predict the good cards in spoiler season…
Even with an incredible AI I'd be surprised unless it can somehow intuit game mechanics just from drawing correlations through a massive dataset.
Anyway, unless mtgtop8 has an api your best bet is probably making a scraper for it, but you run into the issue that playability is relative to environments and a large number of cards have only seen use in what are now extremely dated formats, so the fact that Dralnu du Louvre was an archetype in TSP standard doesn't change the fact that Dralnu is a horrendous card by modern standards, so then you have to, I don't know, make a relatively scale of standard environment by how many of its cards currently see play in legacy or some nonsense (which is kind of suspect considering, well, let's not forget that RAV-TSP has dredge, bridge from below and dread return).
In other words... good luck.
I'm fully aware that I'm pretty much doomed to failure here — I picked this as my project almost purely because Scryfall's API is unusually nice, and I've got about a month to get everything done. 
I just thought that having a big, philosophical question (my "big question" basically boils down to "how much of a card's power is intrinsic?") would be a fun way to justify playing around with smaller stuff like predicting what colors a card is based off of its oracle text.
I just thought that having a big, philosophical question (my "big question" basically boils down to "how much of a card's power is intrinsic?") would be a fun way to justify playing around with smaller stuff like predicting what colors a card is based off of its oracle text.