General How many Scry is worth a card?
- Thread starter Jason Waddell
- Start date
Laz
Developer
Index is a black sheep there, but it seems that Scrying at CMC 2 basically reads 'fine whichever cards you need, it is cool'
Which of these is better?
Back when M11 debuted, Chapin posited that scry 4 ≈ draw 1. After playing with the mechanic since that time, I'm inclined to believe him. He was also one of the first to push Preordain in the month after its printing, and convince the world that it was miles better than Ponder in control decks. I mean, it's like drawing 1.5 cards for
!
Back when M11 debuted, Chapin posited that scry 4 ≈ draw 1. After playing with the mechanic since that time, I'm inclined to believe him. He was also one of the first to push Preordain in the month after its printing, and convince the world that it was miles better than Ponder in control decks. I mean, it's like drawing 1.5 cards for
!
I don't know if Index is a fair comparison because putting to the bottom is most of the point of Scrying.
If the spell is doing anything other than card drawing, that 4:1 ratio probably drops a bunch. If you're just after raw card draw/quality, you need more scry to make up for not being a card draw, but as a rider on a spell instead of a cantrip I think it's probably different.
Chris Taylor
Contributor
Aprox 3?
<3 prog sphinx
Does this draw roughly one card every two turns?
Depends on the deck. If it helps you hit 6 drops, yes. If it hits 2 drops, no.
In limited, assuming the card does something else I've been counting scry as .4 of a card (i.e. magma jet is somewhere around 1.8 cards in my mind, in Th block limited). Scry gets worse in large numbers though, since at some point you hit a card you want and further scry becomes dead. It's just a guideline I made empirically (read as, poorly) during limited though.
Disagree, putting some awful card / your 14th land on bottom is still good. The game doesn't end when you find something you want, yo.
I was introduced to scry around the time I got out of the pool of calculations and just started riding my gut on both card and play evaluation (see: "bad"). I value scry as random, but I feel it's generally stronger than "one card is scry 4." Leaving the cards exactly the way they were feels pretty bad IMO, even if it lets me plan my plays a little more precisely. On the other hand, dodging 2+ awful draw steps as a side effect of a shock is the greatest thing ever and the real reason I want the effect. It clearly shows a lot harder in topdeck mode, but if you crystal ball your millionth and millionth+1 lands to the bottom and draw an actual spell didn't you just basically take 2 extra turns?
I actually think scry gets better in large numbers, because you can do more upfront planning at once. You can more safely send things to the bottom knowing that you'll have desirable cards left on top.
That is to say, I think scry 4 is more than twice as good as scry 2 as a one-shot effect. But, as goldenpineapple alludes to, if you're leaving things on top anyways, scrying continually isn't as valuable. So being able to scry 4 every turn compared to scrying 2 repeatedly is arguably less than twice as good.
Unless you're digging for a specific combo or engine piece, and are heavily incentivized to bottom everything almost always. Then it's more than twice as good again!
That is to say, I think scry 4 is more than twice as good as scry 2 as a one-shot effect. But, as goldenpineapple alludes to, if you're leaving things on top anyways, scrying continually isn't as valuable. So being able to scry 4 every turn compared to scrying 2 repeatedly is arguably less than twice as good.
Unless you're digging for a specific combo or engine piece, and are heavily incentivized to bottom everything almost always. Then it's more than twice as good again!
As a general rule Scry gets much better if you know what you want, specifics or not. That's why the T1 Preordain play was often really shit, because you just don't know what you need at that point. Do you need the lands? The wrath? Who knows.
Scry also gets much better the more context dependent your cards are. If you're deck is all cycling lands, goblin guides, and putrid leeches, and your opponent's deck is all basics, doom blades, and dark betrayals, (let's say, with an aetherling somewhere in there ) then they will value scry way more than you. In fact, in that kind of scenario, scry 2 or 3 is possibly better than draw a card, since such a high percentage of your cards are blanks.
So really, it seems like if we want to figure out when Scry X=draw a card, we need to know more about how you value the cards in your deck.
In the simplest case, we have a 50-50 split, where half the cards in your deck are worthless, and by extension, the other half are worth "two cards" each
Thus, scry 2 has 4 equally likely outcomes: a. two good cards, b. good-bad, c. bad-good, d. bad-bad. these correspond to:
a. draw 0
b. draw 1 (but somewhat worse, since you don't "draw" the extra card until two turns later, when you would have been hitting the dead card)
c. draw 1
d. draw 2
Average: draw 1 (but a turn slower 25% of the time)
This doesn't take into account the extra information you get from scrying, but that's much more complicated.
So in this super simple case, scry 2 is almost equal to draw a card, but obviously decks aren't made up of half 2 for 1s and half blanks (although it sometimes feels that way in topdeck mode).
Ultimately, the value of Scry X is going to depend on the variance of your draw step, the less variable your card power is, the more you want draw, and vice versa.
) then they will value scry way more than you. In fact, in that kind of scenario, scry 2 or 3 is possibly better than draw a card, since such a high percentage of your cards are blanks.So really, it seems like if we want to figure out when Scry X=draw a card, we need to know more about how you value the cards in your deck.
In the simplest case, we have a 50-50 split, where half the cards in your deck are worthless, and by extension, the other half are worth "two cards" each
Thus, scry 2 has 4 equally likely outcomes: a. two good cards, b. good-bad, c. bad-good, d. bad-bad. these correspond to:
a. draw 0
b. draw 1 (but somewhat worse, since you don't "draw" the extra card until two turns later, when you would have been hitting the dead card)
c. draw 1
d. draw 2
Average: draw 1 (but a turn slower 25% of the time)
This doesn't take into account the extra information you get from scrying, but that's much more complicated.
So in this super simple case, scry 2 is almost equal to draw a card, but obviously decks aren't made up of half 2 for 1s and half blanks (although it sometimes feels that way in topdeck mode).
Ultimately, the value of Scry X is going to depend on the variance of your draw step, the less variable your card power is, the more you want draw, and vice versa.
Scry also gets much better the more context dependent your cards are. If you're deck is all cycling lands, goblin guides, and putrid leeches, and your opponent's deck is all basics, doom blades, and dark betrayals, (let's say, with an aetherling somewhere in there
So really, it seems like if we want to figure out when Scry X=draw a card, we need to know more about how you value the cards in your deck.
In the simplest case, we have a 50-50 split, where half the cards in your deck are worthless, and by extension, the other half are worth "two cards" each
Thus, scry 2 has 4 equally likely outcomes: a. two good cards, b. good-bad, c. bad-good, d. bad-bad. these correspond to:
a. draw 0
b. draw 1 (but somewhat worse, since you don't "draw" the extra card until two turns later, when you would have been hitting the dead card)
c. draw 1
d. draw 2
Average: draw 1 (but a turn slower 25% of the time)
This doesn't take into account the extra information you get from scrying, but that's much more complicated.
So in this super simple case, scry 2 is almost equal to draw a card, but obviously decks aren't made up of half 2 for 1s and half blanks (although it sometimes feels that way in topdeck mode).
Ultimately, the value of Scry X is going to depend on the variance of your draw step, the less variable your card power is, the more you want draw, and vice versa.
This is a very good post. Kudos.
ok my problem with this, the reason i made that post: they don't correspond. no matter how much you scry, you will never be able to get at that card right then, you will never be able to activate psychatog an extra time, you will never be able to save a single card currently in your hand from liliana of the veil's +1
filtering is very strong but it is i think it is very very imprecise to say that scry can ever be equivalent to a draw.
it can be worth a trade off of a card to do it, and if that's what you mean, carry on.
filtering is very strong but it is i think it is very very imprecise to say that scry can ever be equivalent to a draw.
it can be worth a trade off of a card to do it, and if that's what you mean, carry on.
Yeah, but a huge part of Magic theory is about trying to assess the value of conversion of resources. Nothing but a card is ever a card, but I think it's an interesting discussion that's presented some insightful statements.
How about a different question then: What if one of the modes that wasn't counter/draw on cryptic command was scry x? How deep would it have to be for people to choose scry over draw? If it's two, and you bottom one, you're in the same "draw location" (your draw order? something) as if you'd just drawn with cryptic- the third card down would be next after your draw step. And since having a physical card is usually more advantageous than knowing your next single draw, it'd probably have to be scry 3 on sheer probability that you bottom more cards?
But that's why I value it randomly, because there are many factors involved in scrying, so it's difficult to weigh against simpler-yet-still-powerful things like drawing cards. My players like scry, and when it keeps you off floods or something it's a real brow-wiper. Foresee is played and digs deep often in my list. It was a shaky add but has been welcomed and loved. Every draft there's at least one instance of scry 4 lands to the bottom draw gold. Screw concentrate.
But that's why I value it randomly, because there are many factors involved in scrying, so it's difficult to weigh against simpler-yet-still-powerful things like drawing cards. My players like scry, and when it keeps you off floods or something it's a real brow-wiper. Foresee is played and digs deep often in my list. It was a shaky add but has been welcomed and loved. Every draft there's at least one instance of scry 4 lands to the bottom draw gold. Screw concentrate.
....Owl i guess? but green and blue decks are not created equal, and flying vs. being an elf is... well, it depends..On scry 3; for your consideration. Which of these do you prefer?
Weird idea, incoming..... Maybe this isn't actually useful, but in terms of "Is X worth a card?" perhaps you can consider the following test:
Suppose you start the game with an emblem that has the following ability. "Discard a card at random: Do X. Activate this ability only once per turn."
How often does it get used? This takes mana costs out of consideration, but it seems to me to be a decent metric for how often effect X is "Worth a card".
If X is Lightning Bolt, it probably gets used all the time, most decks would kill for that emblem, so it stands to reason that bolt is often worth more than a card..
If X is put a 1/1 token on the field, the desirability goes way down.... Lightning bolt is (usually) worth more cards than memnite... (i know bolt costs mana and memnite doesn't but we're not interested in costs here. The question is Whether the effect is worth a card)
The more useful examples though, invole Xs that, on their own, can't win the game. (That way the tempo of a free spell every turn doesn't encourage you to use the effect regardless of whether its a good deal or not.)
Suppose X is Divination. we obviously have a problem here, as you almost always should just activate this ability whenever it is available to you. 2 cards> 1 card (Again, this is an idealized world, don't fret about running out of library, or extra value from madness, or whatever. Think abstract magic. Cards are good. More cards are better. )
How about if X is Draw a card? This case is interesting, and actually why I believe the metric may have value. Thinking in the most abstract terms, it is correct to activate the emblem precisely when the average value of a card in your hand is less than one in your deck, that is to say, when you value "draw a card" more than an average card in your hand. So, according to this metric, "draw a card" is worth more than one card when your hand is worse, on average, than your deck, and worth less than a card when your deck is worse than your hand, which is exactly what conventional wisdom says about random looting. Basically, this metric says that a player should activate the ability whenever effect X is worth more than a card to them, and abstain whenever X is worth less.
So now, what if X is scry 1? This is probably almost never desirable... except when you truly have nothing but garbage to pitch, so it's usually worth less than a card.
The real question is, what if X is Scry 3 or scry 4? how often is it correct to pitch a random card for this effect?
In conclusion: i have no idea...
Does any of the above sound reasonable, or helpful?? It's just a crazy thought that popped into my head...
So this is a little left field, but lets work backwards from something where we know how much value the top of someone's deck is.
In Netrunner (drink), one way the runner can win the game straightforwardly is to be able to access more cards from the top of the corp deck than the corp can draw in one turn. As the corp can draw four cards, you need to access between 4 and lets say 8 cards (based on stealing agendas and trashing assets).
So metric: fateseal is best when you're fatesealing more cards than your opponents draws per turn. Corollary: scry of greater than the number of cards you're likely to draw is best scry? So, say, magma jet is well positioned; the red beatdown deck is unlikely to draw more than it's one card per turn. Similarly foresee is well positioned as it provides you two draw on your 4 scry.
Thinking about this, this sort of makes sense; the reason to run foresee is to dig for cards and it can dig you 6. Scry 1 on the temples is to smooth out draws, and that matches our scry ~= cards drawn metric. It also sort of matches what Jessie says, from the other direction, sort of.
This has been early morning thinking about why netrunner is better than magic with cbob.
In Netrunner (drink), one way the runner can win the game straightforwardly is to be able to access more cards from the top of the corp deck than the corp can draw in one turn. As the corp can draw four cards, you need to access between 4 and lets say 8 cards (based on stealing agendas and trashing assets).
So metric: fateseal is best when you're fatesealing more cards than your opponents draws per turn. Corollary: scry of greater than the number of cards you're likely to draw is best scry? So, say, magma jet is well positioned; the red beatdown deck is unlikely to draw more than it's one card per turn. Similarly foresee is well positioned as it provides you two draw on your 4 scry.
Thinking about this, this sort of makes sense; the reason to run foresee is to dig for cards and it can dig you 6. Scry 1 on the temples is to smooth out draws, and that matches our scry ~= cards drawn metric. It also sort of matches what Jessie says, from the other direction, sort of.
This has been early morning thinking about why netrunner is better than magic with cbob.