Over the last few weeks, I've started to feel that 2-land hands are traps. I've gotten manascrewed so many times with them that its not even funny.
Today I finally got around to doing the math, and it seems that they should be keepers.
Here are the probabilities for playing x lands by turn 7, on the play in a 40-card, 17-land deck:
First column: number of lands in your opening hand
Second column: probabilty of getting that number of lands in your opening hand
Third - Fifth columns: probability of drawing enough lands to have played 3 by turn 3, 4 by turn 4 or 5 by turn t, respectively. Since we're on the play that means we don't get to draw on turn 1.
3- and 4-land hands are obviously better, but 2-landers seem OK in this table.
Am I doing the math wrong, or am I just unlucky?
Do you keep 2-landers if you don't have any or only one 2-drop in your hand?
you shouldn't worry about x lands on turn 7. If you have a 2 land hand, you need to worry about hitting 3 lands on turn 3 and 4 lands on turn four.
edit: er.. maybe I'm reading it wrong. it looks like you included that. If i'm reading that right, more than half the time you miss your 4th land drop with a 2 land hand. That's what you should be judging your hand on. I don't see how you judge them to be keepers.
The math seems kinda irrelevant. Keeping a two-lander is fine provided you have a game plan. Ultimately that should be what determines what you keep and what you throw back. Something to consider if two-landers keep burning you is that maybe your curve is too high and you may need to focus on solving curve issues.
The math seems kinda irrelevant. Keeping a two-lander is fine provided you have a game plan. Ultimately that should be what determines what you keep and what you throw back. Something to consider if two-landers keep burning you is that maybe your curve is too high and you may need to focus on solving curve issues.
Thanks, and agreed my curve might be off.
However, it must be highly relevant to understand how often you can expect to draw those critical third and fourth lands? Or are you saying that you never assume you will draw more lands than you have in your opening hand.
I'm doing the math because it feels like I've gotten stuck on two lands for 4-5 turns about half the time I kept a 2-lander. That shouldn't happen nearly that often according to this math. There are many psychological obstacles when humans try to estimate probabilities, but I'm pretty sure I've gotten unlucky. The math seems to indicate its reasonable to keep 2-landers with the expectation of being able to cast both 3 and 4-CMC spells on turn 3 and 4, respectively.
As a general rule its pretty safe you assume you will hit your first three land drops with a two land hand on the DRAW. On the play expect to miss a land drop. Have a plan for both scenarios. Can it function if you draw five lands straight? Can it function if you draw five non land cards?
Of course you might just be running cold. It happens. Ride it out and think about your curve and whether or not it needs work.
If I have a two-lander and a hand full of two and three drops I'll play easily, then I keep it. If I can't cast anything, its a little more questionable.
Oftentimes mulligans to 6 will end up giving you average 2 lands anyway in a 16-17 land deck. You have to keep some bad hands in Magic, there's no way around that.
I sometimes mull 2-land hands (virtually never on the draw), but then it's usually because my hand is clogged with expensive spells and/or off-color spells. Basically, if my hand is ok if I draw a land within the next 2-3 turns, I keep it.
I believe it was LSV that once said that it is better to be mana-screwed than mana-flooded, because when you eventually draw lands you will have a lot of gas in your hand. Thus, in general a 2-land hand is probably inferior to a 3-land hand, but better than any other combination.
Yeah, I got burned by this very hard yesterday in a DE. I kept several pretty solid two-landers and just couldn't seem to find my third land. One of the hands had a walking corpse on two, as well as vampire nighthawk + ring of xanthid. Despite never drawing a third land and dying, I still think that hand was an easy keep. It had a two drop and if I ever hit swamp, I'm off to the races.
Two landers and five landers are often the toughest to evaluate. However, if you have early plays, or it's a pretty slow, grindy matchup, you can probably get away with keeping land light hands. In the case of early plays, at least you can get on the board. In the case of a grindy matchup (like a fog bank mirror), you want to have more to do than your opponent. Assuming both decks have stuff like divination, I would rather have a bunch of counterspells, removal spells, and filtering and try and win the game with my spell quality. Since he's not pressuring my life total early, I have a lot more time to draw into lands to make my spells more relevant. Obviously, your mileage may vary depending on your deck and what high end spells are in it.
When I have a 2 lander I will usually keep it unless my deck has a super high curve, or the other 4 cards are 4+ cmc.
More often than not, your average 6 card hand will probably have 2 land anyway, lol.
Its important to not place too much stock in those 2 land hands you have kept and then proceeded to not draw land for 6 turns, if you had 3 land instead you probably wouldn't have fared all that much better.
Treat each hand/deck on its own merits, don't place too much weight on the random ordering of your deck in the past, and if you do mulligan and choose to look at the top card before you shuffle (which some people do, just to see if they would have drawn a land if they kept) DO NOT take that into account at all for future decisions. Alot of people keep dodgy hands because the last time they had a hand like that and mulliganed, they looked at the top 4 cards and it was 2 land and 2 spells, so they should have kept.
The answer to this depends more on your opponent's deck than your own, I think. I'll keep a 2-land hand clogged with 3-CMC spells on the play if I know that my opponent is playing a deck that doesn't want to beat down, assuming I'm not playing a very aggressive deck either. Against a slow deck you often want to get manascrewed (as long as you don't have to discard) just because it means that when you eventually do draw a few lands you'll have all spells.
The way I build M13 decks, I'm most comfortable with 2-3 land in my opening hand. I try to keep the curve low, though.
The answer to this depends more on your opponent's deck than your own, I think. I'll keep a 2-land hand clogged with 3-CMC spells on the play if I know that my opponent is playing a deck that doesn't want to beat down, assuming I'm not playing a very aggressive deck either. Against a slow deck you often want to get manascrewed (as long as you don't have to discard) just because it means that when you eventually do draw a few lands you'll have all spells.
The way I build M13 decks, I'm most comfortable with 2-3 land in my opening hand. I try to keep the curve low, though.
Even against an aggressive deck, if you are on the draw, I would more often bet on there being a land in the first 3 cards on top of your library, than on a 6 land hand being better than the 7 land hand you have.
If you have at least something you can cast, and/or have 1 or more (relevant) spells you can cast if you draw one more land, I think the hand is a keeper, it's when you need to draw multiple lands before you can do anything that its a definate mulligan. Be careful though, land light hands have a habit of looking better than they really are, why wouldn't they, there are so many sweet spells
Routinely mulliganing 2-land hands just because they have only 2 lands seems like a bad idea to me.
In a 40-card deck with 17 land, if you mulligan once, your 6-card hand will have:
* 0 land about 3% of the time
* 1 land about 15% of the time
* 2 land about 31% of the time
* 3 land about 31% of the time
* 4 land about 16% of the time
* 5 land about 4% of the time
* 6 land about 0% of the time
By mulliganing, you will get a hand with 3 or 4 land about 47% of the time. Thus, if we view land count and card advantage as the only important factors, then mulliganing a 2-land hand is more likely to make your situation worse than it is to make your situation better. (You have a greater than 50% chance of either by getting an even less desirable land count, or getting another 2-lander and being down a card.)
Of course, you are going to get mana screwed more often with 2-land hands than you are when you have more land, but mulliganing all 2-landers isn't going to help matters.
However, there are a couple things that may help, such as:
* Knowing when building your deck that you need to keep some 2-land hands, and adjusting your mana curve accordingly by including enough spells costing 1-3 mana
* Mulliganing 2-land hands that will be a disaster if you don't hit a fourth land for a while.
I often keep 2-landers on the play with only one 2-drop provided I have a couple 3-drops as well (and hopefully other 2-drops in my deck that I might draw). I also sometimes keep 2-landers with no 2-drops on the play, provided I have several good 3-drops that I could play if I draw any of the lands in my deck.
I'm not necessarily happy with these hands--they might be below-average quality hands for a typical deck--but I think that mulliganing them often just makes matters worse.
In my opinion, the 2-land hand problem is one of the most significant drawbacks to playing first. Two-land hands make me much less apprehensive on the draw.
Awesome analysis - I'm taking another thing away: when you feel your deck is weak on 2- and 3-drops you should really consider drawing if you have the chance.
Private Mod Note
():
Rollback Post to RevisionRollBack
Cry 'Havoc' and let slip the dogs of war!
To post a comment, please login or register a new account.
Today I finally got around to doing the math, and it seems that they should be keepers.
Here are the probabilities for playing x lands by turn 7, on the play in a 40-card, 17-land deck:
First column: number of lands in your opening hand
Second column: probabilty of getting that number of lands in your opening hand
Third - Fifth columns: probability of drawing enough lands to have played 3 by turn 3, 4 by turn 4 or 5 by turn t, respectively. Since we're on the play that means we don't get to draw on turn 1.
3- and 4-land hands are obviously better, but 2-landers seem OK in this table.
Am I doing the math wrong, or am I just unlucky?
Do you keep 2-landers if you don't have any or only one 2-drop in your hand?
edit: er.. maybe I'm reading it wrong. it looks like you included that. If i'm reading that right, more than half the time you miss your 4th land drop with a 2 land hand. That's what you should be judging your hand on. I don't see how you judge them to be keepers.
Agreed - but that's not what the table shows. I will add some explanation in the OP.
Thanks
Thanks, and agreed my curve might be off.
However, it must be highly relevant to understand how often you can expect to draw those critical third and fourth lands? Or are you saying that you never assume you will draw more lands than you have in your opening hand.
I'm doing the math because it feels like I've gotten stuck on two lands for 4-5 turns about half the time I kept a 2-lander. That shouldn't happen nearly that often according to this math. There are many psychological obstacles when humans try to estimate probabilities, but I'm pretty sure I've gotten unlucky. The math seems to indicate its reasonable to keep 2-landers with the expectation of being able to cast both 3 and 4-CMC spells on turn 3 and 4, respectively.
Of course you might just be running cold. It happens. Ride it out and think about your curve and whether or not it needs work.
Oftentimes mulligans to 6 will end up giving you average 2 lands anyway in a 16-17 land deck. You have to keep some bad hands in Magic, there's no way around that.
I believe it was LSV that once said that it is better to be mana-screwed than mana-flooded, because when you eventually draw lands you will have a lot of gas in your hand. Thus, in general a 2-land hand is probably inferior to a 3-land hand, but better than any other combination.
Two landers and five landers are often the toughest to evaluate. However, if you have early plays, or it's a pretty slow, grindy matchup, you can probably get away with keeping land light hands. In the case of early plays, at least you can get on the board. In the case of a grindy matchup (like a fog bank mirror), you want to have more to do than your opponent. Assuming both decks have stuff like divination, I would rather have a bunch of counterspells, removal spells, and filtering and try and win the game with my spell quality. Since he's not pressuring my life total early, I have a lot more time to draw into lands to make my spells more relevant. Obviously, your mileage may vary depending on your deck and what high end spells are in it.
*DCI Rules Advisor*
More often than not, your average 6 card hand will probably have 2 land anyway, lol.
Its important to not place too much stock in those 2 land hands you have kept and then proceeded to not draw land for 6 turns, if you had 3 land instead you probably wouldn't have fared all that much better.
Treat each hand/deck on its own merits, don't place too much weight on the random ordering of your deck in the past, and if you do mulligan and choose to look at the top card before you shuffle (which some people do, just to see if they would have drawn a land if they kept) DO NOT take that into account at all for future decisions. Alot of people keep dodgy hands because the last time they had a hand like that and mulliganed, they looked at the top 4 cards and it was 2 land and 2 spells, so they should have kept.
The way I build M13 decks, I'm most comfortable with 2-3 land in my opening hand. I try to keep the curve low, though.
Even against an aggressive deck, if you are on the draw, I would more often bet on there being a land in the first 3 cards on top of your library, than on a 6 land hand being better than the 7 land hand you have.
If you have at least something you can cast, and/or have 1 or more (relevant) spells you can cast if you draw one more land, I think the hand is a keeper, it's when you need to draw multiple lands before you can do anything that its a definate mulligan. Be careful though, land light hands have a habit of looking better than they really are, why wouldn't they, there are so many sweet spells
But if you need to get lucky and hit *two* land or even *three*, the odds turn against you.
As the table in the opening post points out, you can really only bank on drawing one land consistently in the first few turns of the game.
In a 40-card deck with 17 land, if you mulligan once, your 6-card hand will have:
* 0 land about 3% of the time
* 1 land about 15% of the time
* 2 land about 31% of the time
* 3 land about 31% of the time
* 4 land about 16% of the time
* 5 land about 4% of the time
* 6 land about 0% of the time
By mulliganing, you will get a hand with 3 or 4 land about 47% of the time. Thus, if we view land count and card advantage as the only important factors, then mulliganing a 2-land hand is more likely to make your situation worse than it is to make your situation better. (You have a greater than 50% chance of either by getting an even less desirable land count, or getting another 2-lander and being down a card.)
Of course, you are going to get mana screwed more often with 2-land hands than you are when you have more land, but mulliganing all 2-landers isn't going to help matters.
However, there are a couple things that may help, such as:
* Knowing when building your deck that you need to keep some 2-land hands, and adjusting your mana curve accordingly by including enough spells costing 1-3 mana
* Mulliganing 2-land hands that will be a disaster if you don't hit a fourth land for a while.
I often keep 2-landers on the play with only one 2-drop provided I have a couple 3-drops as well (and hopefully other 2-drops in my deck that I might draw). I also sometimes keep 2-landers with no 2-drops on the play, provided I have several good 3-drops that I could play if I draw any of the lands in my deck.
I'm not necessarily happy with these hands--they might be below-average quality hands for a typical deck--but I think that mulliganing them often just makes matters worse.
In my opinion, the 2-land hand problem is one of the most significant drawbacks to playing first. Two-land hands make me much less apprehensive on the draw.