My Gruul section is Bloodbraid Elf, Rhythm of the Wild and Ghor-Clan Rampager. I like Rampager over Sunder Shaman, fwiw.
As to Ikoria, I haven't gotten any for the peasant cube yet, so I don't have experience with them in that format. I plan on adding Grimdancer, Heartless Act, and Titanoth Rex. I think mutate is a bit too parasitic and the ones we have access to didn't seem good enough if you only get one trigger out of them.
Unfortunatley haven't been able to draft much with everything that's going on. Some day I'll be able to evaluate these additions in practice rather than just theory.
By some miracle, I actually got nearly a dozen people to put down that they'd have a real draft on the 11th of July. I'll try and see if I can do the C/u list and then after that I don't know which cube I want to try out the most. I think the Orzhov list needs the most eyes on it.
Conveniently, NY is almost out of the woods with this virus (lowest infection rate in the country, less than 1% of people test positive now)
Prerelease done, I managed to get all the cards I needed (dumpster diving the chaff people left) I also managed to get some foil Ikoria cards I was missing.
Now there is just one card I am missing and it's not even that good... Splendor Mare to replace sandsteppe outcast (which is weak). Is there some other 3 mana good stuff white creature I can use instead?
I also suggest Hookmaster straight away. Advokist's another multiplayer card that's strong in 2-player Magic. I feel it's a bit like a white Standstill for the battlefield in that it isn't hard to have a board state where the onus really is on your opponent to deal with it.
Private Mod Note
():
Rollback Post to RevisionRollBack
I wish all archetypes could be equally hated. Or loved, but the former better reflects the mindset of the vocal portion of the player base. My 540ish Peasant Cube on Cubetutor
I think I would still put them in the order you listed. Getting (pseudo-)evasion from some other source is relatively easy whereas the big thing Hexproof offers is survivability, making that extra toughness important in making the card more resistant to death by blockers and able to attack more freely to grind down an opponent's resources. Being one mana less than the Crusher also certainly helps.
At least I hold Behemoth above Crusher. I feel 6/7 Hexproof is better here than 7/6 with Trample added, and only 6 cmc makes it better just as a cube card too.
Private Mod Note
():
Rollback Post to RevisionRollBack
I wish all archetypes could be equally hated. Or loved, but the former better reflects the mindset of the vocal portion of the player base. My 540ish Peasant Cube on Cubetutor
It's all anecdotal evidence, but from my experience the lack of both trample and vigilance really limits the usefulness of Scaled Behemoth a lot. If you want to kill your opponent quickly he can block the Behemoth with tokens or cheap creatures all day long, if you need a defender you can't attack at all. I had Behemoth sit around for these reasons far too often to value him higher than the other two. From my pov they all have their drawbacks and I wouldn't be able to decide which one I like most.
If you don't go for raw power level alone it is also worth mentioning that hexproof is one of the most unfun mechanics in Magic. I personally don't mind if there are a few very high cmc creatures with it in my cube, but I try to keep the numbers as low as possible. And green has more than enough other options now, so I don't feel there is a need for Behemoth or Crusher in my cube anymore. If I had a larger cube I'd still consider them though.
I would say Fatal Push is best and after that it depends on your cube.
Disfigure is the card that fits into the most decks but tends to be more of a control/midrange card. Although Ulcerate hits more targets it is worse in control/midrange against aggro so if you want to boost aggro a bit (or your cube has no aggro decks) Ulcerate might be the choice.
Darkblast is great in B/x graveyard matter decks if you support them. If you don't, I would rate it below Disfigure/Ulcerate. It can be much stronger than those cards against the right deck (aggro) so it depends if you want a more general removal spell or one more specifically tailor to be used against aggro decks.
Defile is strong if you want to push mono decks. In two color decks it scales better into the late game. However, it is less consistent in the early part which to me defies the purpose of one mana removal spells. Why not play a good two mana removal spell instead then?
Overal rating if I had to:
1. Fatal Push
2. Disfigure
3. Ulcerate
4. Darkblast (is number 2 if you support B/x graveyard matters)
5. Defile ( is number 3 if you support mono decks)
Defile vs Disfigure
This isn't a monocolor issue at all. It's all about the scaling.
Average case, Defile only lags behind Disfigure on the opponents 2 and 3 drop turns.
(I am assuming you're running 9 swamps with the math)
Turn 1 Against 1 drops, they are essentially identical.
Turn 2
Only about half of our two drops have 2+ toughness. So half the time, it snipes a 1 toughness guy just as well as disfigure. Against 2 toughness creatures, there's... a lot of gamestate to take into account. Worst case scenario, you were on the draw when they played a 2/2 on their turn 2, and you have a different play to make on your next turn. But in every other scenario we're just looking at the odds of having 2 swamps by your turn 2. That's 59-67%. Ballparking it all together, it seems like defile will be as good as disfigure on this turn like 80% of the time.
Turn 3
For turn 3, the odds of having 2 swamps ranges from 67% to 73%. Accounting for 1 toughness creatures, 2 drops being played on turn 3, etc, it seems like defile is as good as disfigure on turn 3 about 70-80% of the time again.
Turn 4
This is interesting. You only have a 73-79% chance of having 2 swamps by now. But you have a 39-47% chance of having three or more swamps by now as well. So while 24% of the time it's worse than disfigure, 43% of the time it's better.
Defile's entire value would have to come from it being close to terror in the late game, and it starts being able to do that occasionally as early as turn 4. Being able to moonlight as a 1 mana terror from turn 5 or 6 and onward is comparable to how we value slapping cycling on a 6 or 7 drop.
Perhaps defile is best thought of as a cross between disfigure and murderous cut? Murderous cut is only good because it can be a 1 mana terror eventually, but it does have a plan b of being about as good as murder at about turn 4 or 5. Defile can almost be a 1 mana terror in the late game but its plan b is instead to be a disfigure for the early game. Which isn't too bad.
How rare are three color decks in your cube for you to not even consider them for the average case of Defile?
I feel like at least 10-20% of all decks are more than two color in my cube.
For me the most important part about a one mana removal spell is consistency in the early game. So, at least for me, Defile vs Disfigure is (at least in part) a monocolor issue.
(I am assuming you're running 9 swamps with the math)
How often do you run 9 swamps? If your second colour has cards costing CC, 1CC or 2CC, you are probably running 3 nonbasics and 7 of each basic? In your cube there were about 9 such cards in each colour.
Looking at your cube, it seems that on average each player might end up with 3 nonbasics ( (1/2)*15*152/368=3,1)? This doesn't even count the coloured lands and dual lands, so the average might be higher. Given that you usually build 2 colour decks, the expected value of swamps in a Bx deck would be 7. What does the math look like for this case?
I agree with Arbor, for me being better late game (and often only very late game) doesn't make up for the lack of reliability in the early game when we talk about a one mana spell. I generally dislike spells that have a random factor to them and avoid them if possible. If I had to choose between Defile and Disfigure I'd always take Disfigure unless I had a mono black deck.
This is also why I cut Darkblast. It can be devastating against the right deck, but sometimes it's just trash. The graveyard matters support (and I support that archetype) doesn't make up for that kind of randomness.
How rare are three color decks in your cube for you to not even consider them for the average case of Defile?
Offhand I would have guessed that something like 5-10% of the cube's decks are more than 2 color. But we all run 1CC three drops and those also are almost reserved for 2 color decks. I think of defile's color requirements to be similar to that of vampire nighthawk, minus the total failure case when color screwed.
How often do you run 9 swamps? If your second colour has cards costing CC, 1CC or 2CC, you are probably running 3 nonbasics and 7 of each basic? In your cube there were about 9 such cards in each colour.
Looking at your cube, it seems that on average each player might end up with 3 nonbasics ( (1/2)*15*152/368=3,1)? This doesn't even count the coloured lands and dual lands, so the average might be higher. Given that you usually build 2 colour decks, the expected value of swamps in a Bx deck would be 7. What does the math look like for this case?
I'm actually having trouble following the math you did to arrive at this number of 3 (It actually took me a long time just to remember that different people use commas to separate wholes from tenths). Why did you divide 152 by 368?
In any case, I did neglect to account for nonbasics. By rough estimate, I'm used to seeing something to the effect of 1.5 nonbasics per deck. It isn't wildly uncommon to see 3 in a two color deck, but I am certainly more likely to see 0 than 3. This is probably because people don't always run every nonbasic in their colors. So I'll grant 1.5 per deck, which puts me on fair ground to expect 8 swamps with room to see perfectly common decks that have 9.
Just so you know, with 8 swamps in the deck the percentages drop by approximately 5 or 6 if I remember correctly. Not negligible but ok-ish.
I agree with Arbor, for me being better late game (and often only very late game) doesn't make up for the lack of reliability in the early game when we talk about a 1 mana spell
I know you can't believe this is universally true because I know you don't think this of path to exile. So there is some amount of being good at different stages of the game that is acceptable. Defile never reaches the high point of PtE, obviously, and it requires you to be majority black, but it at least actually functions in the early game where PtE is probably a serious misplay.
Look, I drew a thingy https://imgur.com/a/2p0M6X6
Usefulness on the y axis and turn count on the X axis. My MS Paint skills aren't what they used to be, but I think this is reasonably fair.
Defile is essentially as good as disfigure on turn 1, then worse for turns 2 and 3. Slightly better on turn 4 on average, then just better from there. All with the caveat of being only playable in a deck that would be happy casting a spell for 1BB. Turns 5 to 7 aren't so late in the game to be outliers, they're perfectly relevant. Hence why we entertain at least conversationally the value of cycling on a 2 drop.
Even if defile just upgrades to 1 mana last gasp for the whole game, it's really tough for last gasp to lose its value except against ramp.
//
Your other complaint about randomness is interesting, although the variability in defile vs a card like... vampire nighthawk is only that nighthawk has an on/off switch for when it works and defile uses a sliding scale. Defile's variance functions sorta similarly to rise from the tides.
I'm actually having trouble following the math you did to arrive at this number of 3 (It actually took me a long time just to remember that different people use commas to separate wholes from tenths). Why did you divide 152 by 368?
I assumed your grid draft that you described in another thread.
4 grids of 36 cards = 144 cards
2 players get 1 extra card pr grid = 8
This means you play with 152 out of 368 cards, assuming you haven't changed your ratios from your list on cubetutor.
You have 15 nonbasics in your colourless section. This means that the players see 6.2 rainbow/colourless nonbasics between them. I guess if you don't pick up every land, you will arrive at a lower number than 3.1/player.
If each player go for two different guilds, you would have 7 mono coloured lands (8 if noone plays black). There are 1 dual land for each of your guilds, resulting in another 2 lands. The expected value of relevant lands/draft would be:
Relevant lands in cube*Cards used/cube size
(15+7+2) * 152/368 = 9.9
If you pick up every land, each player would have about 5 nonbasics in their draft pool. With your grid (quilt?) drafting, I can see why you probably won't end up picking all the lands.
I know you can't believe this is universally true because I know you don't think this of path to exile. So there is some amount of being good at different stages of the game that is acceptable. Defile never reaches the high point of PtE, obviously, and it requires you to be majority black, but it at least actually functions in the early game where PtE is probably a serious misplay.
Well, now you got me. Obviously not every one mana spell needs to be great in the early game. But Path to Exile will remove any creature no matter what and it will do it when you need it and when you want it and you know exactly what kind of drawback you will get for it. It's your decision, and sometimes even a PtE on turn 2 or 3 can be the right play (I know for sure I cast it at that point many times and didn't regret it), while a Defile on turn 2 will often not give you any option because it simply can't do anything useful at that time. And sometimes it only hits a 2 toughness creature even on turn 10 - the chance to have three swamps by that time is only about 70% in a deck with 8 swamps if you go first, which means that will happen in almost 1 out of 3 games. That's quite a lot.
Vampire Nighthawk seen as on/off switch is a good example because that's exactly why I play as few CC cards as possible and cut otherwise good cards like Consul's Lieutenant or Gifted Aetherborn from my cube. And that even though I run better mana fixing lands than most people. I still do not consider Nighthawk a great pick unless you have good fixing lands in your deck or play mono black. And Defile is worse than Nighthawk from that perspective as you absolutely need swamps to make it useful, so a City of Brass or Gemstone Mine or Thriving land or Paradise Druid or Coldsteel Heart etc won't do anything for it, while all these cards will allow you to cast Nighthawk. If you want Defile to be better than Disfigure it's like a CCC card that can only be cast from basic lands, which no one right in their mind would run if it was an on/off type card (not that anyone would run it if it was possible to cast it from any mana source...).
Last but not least your graph is a bit exaggerated when it comes to late game usage of Defile vs Disfigure vs Path. On one hand it looks as if Defile becomes way better than Disfigure very quickly after just a few turns, which just isn't true when you look at how slowly the percentage where the average case becomes better rises. And while the average case may be reached after ~3.5 turns the randomness should be taken into account as well. I'd rather have a card that does what it says it does when I want to play it than a card that sometimes does more and sometimes does nothing. What would you prefer to play? A basic swamp or a land where you flip two coins during your upkeep/etb and for each flip won you get B when you tap it?
And Defile certainly isn't closer to Path than to Disfigure on turn 8 (and apparently way closer on turns 9+ if you'd extend these lines).
Disfigure is great in the early game. Path to Exile is great in the late game. Both can be useful in all stages of the game. Defile is...good in the mid game, but worse than Disfigure in the early game and worse than PtE (or, for better comparison, something like Go for the Throat) in the late game. And while it randomly can be very good in each stage of the game it can also be randomly utterly useless in each stage of the game. All of that of course outside of mono black, where it is a great card. That's why I wouldn't run it over Disfigure unless I wanted to support mono black.
As to Ikoria, I haven't gotten any for the peasant cube yet, so I don't have experience with them in that format. I plan on adding Grimdancer, Heartless Act, and Titanoth Rex. I think mutate is a bit too parasitic and the ones we have access to didn't seem good enough if you only get one trigger out of them.
MTGS Average Peasant Cube 2023 Edition
Follow me. I tweet.
Sprite Dragon
Grimdancer
Bastion of Remembrance
Titanoth Rex
Chainweb Aracnir
Irreverent Revelers
Embereth Shieldbreaker
Order of Midnight
Syr Konrad, the Grim
Unfortunatley haven't been able to draft much with everything that's going on. Some day I'll be able to evaluate these additions in practice rather than just theory.
Same.
2023 Average Peasant Cube|and Discussion
Because I have more decks than fit in a signature
Useful Resources:
MTGSalvation tags
EDHREC
ManabaseCrafter
Voltaic Brawler is absurd if you support green aggro.
Draft it on Cubetutor here, and CubeCobra here.
Treasure Cruise did nothing wrong.
;_;
Conveniently, NY is almost out of the woods with this virus (lowest infection rate in the country, less than 1% of people test positive now)
My CubeCobra (draft 20 card packs, 2 packs.)
430, Peasant, Very Unpowered
Why you should take your hybrids out of your gold section
Manamath Article
Now there is just one card I am missing and it's not even that good... Splendor Mare to replace sandsteppe outcast (which is weak). Is there some other 3 mana good stuff white creature I can use instead?
3 drops
Splendor Maremaybe I have too many white 3 drops hmm
Pioneer:UR Pheonix
Modern:U Mono U Tron
EDH
GB Glissa, the traitor: Army of Cans
UW Dragonlord Ojutai: Dragonlord NOjutai
UWGDerevi, Empyrial Tactician "you cannot fight the storm"
R Zirilan of the claw. The solution to every problem is dragons
UB Etrata, the Silencer Cloning assassination
Peasant cube: Cards I own
Have you tried Soltari Champion or Kor Hookmaster?
MTGS Average Peasant Cube 2023 Edition
Follow me. I tweet.
My 540ish Peasant Cube on Cubetutor
Scaled Behemoth
Plated Crusher
Warden of the Woods
I really like the 7 toughess on the Behemoth, but the lack of trample concerns me...
My Peasant Cube: @ mtgsalvation---- @ cubecobra
My 540ish Peasant Cube on Cubetutor
If you don't go for raw power level alone it is also worth mentioning that hexproof is one of the most unfun mechanics in Magic. I personally don't mind if there are a few very high cmc creatures with it in my cube, but I try to keep the numbers as low as possible. And green has more than enough other options now, so I don't feel there is a need for Behemoth or Crusher in my cube anymore. If I had a larger cube I'd still consider them though.
My Old School Battlebox
My Premodern Battlebox
Might as well ask for a ranking:
I'm running all but disfigure, and am looking to remove defile to test other cards.
My CubeCobra (draft 20 card packs, 2 packs.)
430, Peasant, Very Unpowered
Why you should take your hybrids out of your gold section
Manamath Article
Disfigure is the card that fits into the most decks but tends to be more of a control/midrange card. Although Ulcerate hits more targets it is worse in control/midrange against aggro so if you want to boost aggro a bit (or your cube has no aggro decks) Ulcerate might be the choice.
Darkblast is great in B/x graveyard matter decks if you support them. If you don't, I would rate it below Disfigure/Ulcerate. It can be much stronger than those cards against the right deck (aggro) so it depends if you want a more general removal spell or one more specifically tailor to be used against aggro decks.
Defile is strong if you want to push mono decks. In two color decks it scales better into the late game. However, it is less consistent in the early part which to me defies the purpose of one mana removal spells. Why not play a good two mana removal spell instead then?
Overal rating if I had to:
1. Fatal Push
2. Disfigure
3. Ulcerate
4. Darkblast (is number 2 if you support B/x graveyard matters)
5. Defile ( is number 3 if you support mono decks)
My C/Ube on Cube Cobra
This isn't a monocolor issue at all. It's all about the scaling.
Average case, Defile only lags behind Disfigure on the opponents 2 and 3 drop turns.
(I am assuming you're running 9 swamps with the math)
Turn 1
Against 1 drops, they are essentially identical.
Turn 2
Only about half of our two drops have 2+ toughness. So half the time, it snipes a 1 toughness guy just as well as disfigure. Against 2 toughness creatures, there's... a lot of gamestate to take into account. Worst case scenario, you were on the draw when they played a 2/2 on their turn 2, and you have a different play to make on your next turn. But in every other scenario we're just looking at the odds of having 2 swamps by your turn 2. That's 59-67%. Ballparking it all together, it seems like defile will be as good as disfigure on this turn like 80% of the time.
Turn 3
For turn 3, the odds of having 2 swamps ranges from 67% to 73%. Accounting for 1 toughness creatures, 2 drops being played on turn 3, etc, it seems like defile is as good as disfigure on turn 3 about 70-80% of the time again.
Turn 4
This is interesting. You only have a 73-79% chance of having 2 swamps by now. But you have a 39-47% chance of having three or more swamps by now as well. So while 24% of the time it's worse than disfigure, 43% of the time it's better.
Defile's entire value would have to come from it being close to terror in the late game, and it starts being able to do that occasionally as early as turn 4. Being able to moonlight as a 1 mana terror from turn 5 or 6 and onward is comparable to how we value slapping cycling on a 6 or 7 drop.
Perhaps defile is best thought of as a cross between disfigure and murderous cut? Murderous cut is only good because it can be a 1 mana terror eventually, but it does have a plan b of being about as good as murder at about turn 4 or 5. Defile can almost be a 1 mana terror in the late game but its plan b is instead to be a disfigure for the early game. Which isn't too bad.
I think I just talked myself into keeping it in.
My CubeCobra (draft 20 card packs, 2 packs.)
430, Peasant, Very Unpowered
Why you should take your hybrids out of your gold section
Manamath Article
I feel like at least 10-20% of all decks are more than two color in my cube.
For me the most important part about a one mana removal spell is consistency in the early game. So, at least for me, Defile vs Disfigure is (at least in part) a monocolor issue.
My C/Ube on Cube Cobra
How often do you run 9 swamps? If your second colour has cards costing CC, 1CC or 2CC, you are probably running 3 nonbasics and 7 of each basic? In your cube there were about 9 such cards in each colour.
Looking at your cube, it seems that on average each player might end up with 3 nonbasics ( (1/2)*15*152/368=3,1)? This doesn't even count the coloured lands and dual lands, so the average might be higher. Given that you usually build 2 colour decks, the expected value of swamps in a Bx deck would be 7. What does the math look like for this case?
This is also why I cut Darkblast. It can be devastating against the right deck, but sometimes it's just trash. The graveyard matters support (and I support that archetype) doesn't make up for that kind of randomness.
My Old School Battlebox
My Premodern Battlebox
Offhand I would have guessed that something like 5-10% of the cube's decks are more than 2 color. But we all run 1CC three drops and those also are almost reserved for 2 color decks. I think of defile's color requirements to be similar to that of vampire nighthawk, minus the total failure case when color screwed.
I'm actually having trouble following the math you did to arrive at this number of 3 (It actually took me a long time just to remember that different people use commas to separate wholes from tenths). Why did you divide 152 by 368?
In any case, I did neglect to account for nonbasics. By rough estimate, I'm used to seeing something to the effect of 1.5 nonbasics per deck. It isn't wildly uncommon to see 3 in a two color deck, but I am certainly more likely to see 0 than 3. This is probably because people don't always run every nonbasic in their colors. So I'll grant 1.5 per deck, which puts me on fair ground to expect 8 swamps with room to see perfectly common decks that have 9.
Just so you know, with 8 swamps in the deck the percentages drop by approximately 5 or 6 if I remember correctly. Not negligible but ok-ish.
I know you can't believe this is universally true because I know you don't think this of path to exile. So there is some amount of being good at different stages of the game that is acceptable. Defile never reaches the high point of PtE, obviously, and it requires you to be majority black, but it at least actually functions in the early game where PtE is probably a serious misplay.
Look, I drew a thingy
https://imgur.com/a/2p0M6X6
Usefulness on the y axis and turn count on the X axis. My MS Paint skills aren't what they used to be, but I think this is reasonably fair.
Defile is essentially as good as disfigure on turn 1, then worse for turns 2 and 3. Slightly better on turn 4 on average, then just better from there. All with the caveat of being only playable in a deck that would be happy casting a spell for 1BB. Turns 5 to 7 aren't so late in the game to be outliers, they're perfectly relevant. Hence why we entertain at least conversationally the value of cycling on a 2 drop.
Even if defile just upgrades to 1 mana last gasp for the whole game, it's really tough for last gasp to lose its value except against ramp.
//
Your other complaint about randomness is interesting, although the variability in defile vs a card like... vampire nighthawk is only that nighthawk has an on/off switch for when it works and defile uses a sliding scale. Defile's variance functions sorta similarly to rise from the tides.
My CubeCobra (draft 20 card packs, 2 packs.)
430, Peasant, Very Unpowered
Why you should take your hybrids out of your gold section
Manamath Article
I assumed your grid draft that you described in another thread.
4 grids of 36 cards = 144 cards
2 players get 1 extra card pr grid = 8
This means you play with 152 out of 368 cards, assuming you haven't changed your ratios from your list on cubetutor.
You have 15 nonbasics in your colourless section. This means that the players see 6.2 rainbow/colourless nonbasics between them. I guess if you don't pick up every land, you will arrive at a lower number than 3.1/player.
If each player go for two different guilds, you would have 7 mono coloured lands (8 if noone plays black). There are 1 dual land for each of your guilds, resulting in another 2 lands. The expected value of relevant lands/draft would be:
Relevant lands in cube*Cards used/cube size
(15+7+2) * 152/368 = 9.9
If you pick up every land, each player would have about 5 nonbasics in their draft pool. With your grid (quilt?) drafting, I can see why you probably won't end up picking all the lands.
Well, now you got me. Obviously not every one mana spell needs to be great in the early game. But Path to Exile will remove any creature no matter what and it will do it when you need it and when you want it and you know exactly what kind of drawback you will get for it. It's your decision, and sometimes even a PtE on turn 2 or 3 can be the right play (I know for sure I cast it at that point many times and didn't regret it), while a Defile on turn 2 will often not give you any option because it simply can't do anything useful at that time. And sometimes it only hits a 2 toughness creature even on turn 10 - the chance to have three swamps by that time is only about 70% in a deck with 8 swamps if you go first, which means that will happen in almost 1 out of 3 games. That's quite a lot.
Vampire Nighthawk seen as on/off switch is a good example because that's exactly why I play as few CC cards as possible and cut otherwise good cards like Consul's Lieutenant or Gifted Aetherborn from my cube. And that even though I run better mana fixing lands than most people. I still do not consider Nighthawk a great pick unless you have good fixing lands in your deck or play mono black. And Defile is worse than Nighthawk from that perspective as you absolutely need swamps to make it useful, so a City of Brass or Gemstone Mine or Thriving land or Paradise Druid or Coldsteel Heart etc won't do anything for it, while all these cards will allow you to cast Nighthawk. If you want Defile to be better than Disfigure it's like a CCC card that can only be cast from basic lands, which no one right in their mind would run if it was an on/off type card (not that anyone would run it if it was possible to cast it from any mana source...).
Last but not least your graph is a bit exaggerated when it comes to late game usage of Defile vs Disfigure vs Path. On one hand it looks as if Defile becomes way better than Disfigure very quickly after just a few turns, which just isn't true when you look at how slowly the percentage where the average case becomes better rises. And while the average case may be reached after ~3.5 turns the randomness should be taken into account as well. I'd rather have a card that does what it says it does when I want to play it than a card that sometimes does more and sometimes does nothing. What would you prefer to play? A basic swamp or a land where you flip two coins during your upkeep/etb and for each flip won you get B when you tap it?
And Defile certainly isn't closer to Path than to Disfigure on turn 8 (and apparently way closer on turns 9+ if you'd extend these lines).
Disfigure is great in the early game. Path to Exile is great in the late game. Both can be useful in all stages of the game. Defile is...good in the mid game, but worse than Disfigure in the early game and worse than PtE (or, for better comparison, something like Go for the Throat) in the late game. And while it randomly can be very good in each stage of the game it can also be randomly utterly useless in each stage of the game. All of that of course outside of mono black, where it is a great card. That's why I wouldn't run it over Disfigure unless I wanted to support mono black.
My Old School Battlebox
My Premodern Battlebox