General Removal density

[japahn]

How much removal do you run in your cube? Is there a philosophy behind the density of it?

Just for the sake of consistent benchmarking, let's define the lines of what to count as removal for the purposes of the percentages:

Things that I considered removal:


Things that I didn't consider removal:


For my cube, I got 71 removal spells out of 309 nonland cards, which is 23.0%, which was surprisingly high to me. In terms of colors: White (13), Blue (6), Black (13), Red (20), Green (1), Multi (9), Colorless (9).

I tried lower amounts of removal but found too many games ending by an unchecked flier while all the rest didn't matter. My hypothesis is that evasion is costed for constructed, where removal is near unlimited.

Therefore, in a low removal format, creatures need to be interacted with by blocking, so evasive creatures, first strikers, and ones that don't need to attack to generate values become much stronger. On the other hand, flying can be mitigated by reach, and menace and trample are generally ok. I think we're just starting to break ground in low removal formats and finding that there is a lot of room to explore in designing combat (shoutout to @Lady Lynn and their ramblings on P/T dynamics and evergreen keywords).

Edit: Considering counterspells as removal as well, I forgot to consider them.

 

[landofMordor]

I think we're just starting to break ground in low removal formats and finding that there is a lot of room to explore in designing combat (shoutout to @Lady Lynn and their ramblings on P/T dynamics and evergreen keywords).

Couldn't agree more. Lynn's thoughts have been really great, and these low-removal formats are also exciting. (Have we made reference to Retail Limited yet to interrogate these combat dynamics? retail asfan is usually quite low.)

I think I'm gonna count the following as removal, in addition to the examples listed:


This yields 92/466 ~ 20% removal. If we only count spells, the number goes up to 92/376 ~ 25%.

I'm concerned with the same sort of unchecked fliers as japahn, with three additional concerns:
- checking snowballing threats. I want games to be short and brutal, but I also want players to feel like they exercise agency. As such, I want them to be able to cleanly answer Uro and Oko in some portion of their games.
- keeping board states comprehensible. Since so many of my threats are snowballing, it can get hard to analyze board states. Removal helps keep combat at the scale where it fits in the human brain.
- sideboard and deckbuilding flexibility. Since most of my cube's decks are shades of midrange, which is already the greediest theater of Magic for spot removal, I want extra removal so that everybody gets enough, and they're able to flex into a different removal configuration after sideboarding.

 

[Chriskool]

Great topic, @japahn ! Have you or @landofMordor delved into conditionality of your removal suites (and any non-countermagic interactions that might stop if from effectively answered the recipient(s))?

I know that a lot of removal fits "burn" (I assume 3 damage) / "murder/hero's downfall" / "disenchant" (and countermagic, kinda!) classifications in a more traditional cube. How do the different classes of removal stack up against the threats?

I will deep-dive into a grid with many multiples and post my ramblings this week!

 

[Brad]

https://www.coolstuffinc.com/a/alexullman-030414-removing-removal-in-cube
TL;DR:
ROE: 1.95 asfan
INN: 1.7 asfan
Not clear what they counted as removal.

I go for asfan of 2. Asfan is how many you'd expect to see per pack "as fanned" out, so 2/15 cards, or ~13.33%.
We found in another thread that many of us were running about 2% of the total cube as Wraths.

360: ~50 removal, ~7 Wraths
450: ~60 removal, ~9 Wraths

I personally don't count bounce as removal and do count counterspells, but I don't run a density of bounce where it would sway the numbers significantly, either. I also tend to end up a little over 2.

 

[blacksmithy]

huh, i am at 48 removals / (224 total cards -60 lands) = 29.27% removal, seems pretty OK

 

[TrainmasterGT]

Right now, I am at 75 removal spells out of 311 nonland cards, which equates to 24.11% removal. I included Thoughtseize effects that could hit creatures in this calculation.

My removal As-Fan is 3.1245.

 

[Nanonox]

Somewhat related question.

Assuming you play with less than a full pod (3-5 people in my case), but still draft the whole cube. Do you change the numbers to reflect the lower amount of players or do you just keep it as if there were 8 drafters (even if it won't happen)?

I am currently designing with a full pod in mind, but I don't know if this is right!

 

[TrainmasterGT]

Somewhat related question.

Assuming you play with less than a full pod (3-5 people in my case), but still draft the whole cube. Do you change the numbers to reflect the lower amount of players or do you just keep it as if there were 8 drafters (even if it won't happen)?

I am currently designing with a full pod in mind, but I don't know if this is right!

As long as you’re still drafting the whole Cube, your players are still getting cards in the same ratio, they’re just getting more of them. The primary difference is that the resulting decks are going to be more consistent. I would say that you only need to scale back on removal if your players are running “too much” of it, depending on what that means to you.

 

[japahn]

As long as you’re still drafting the whole Cube, your players are still getting cards in the same ratio, they’re just getting more of them.

Disagree. Players will have access to a larger number of removal spells, so if the average removal is better than the average card they will likely play a larger proportion of them.

Whether you want to adjust for that depends on if you want to keep the cube balanced for a full pod or optimize it to drafted with 3-5 people.

 

[sigh]

This, as mentioned, requires a draft style where the whole cube is still seen. If you are just doing 3x15, the amount of cards seen scales with the number of players, and the ratios of cards seen is randomized. In theory 4 players could see all of the removal from the entire cube, putting the ratio to 40 or 50%. Highly unlikely of course, but some inconsistency will be there.

 

[blacksmithy]

yeah if you draft the whole cube then everyone is going to get exactly the ratios of fixing/threats/removal/etc that are actually in the cube, regardless of how that gets distributed among players or maindecked. which i THINK is what train was saying too?

 

[japahn]

People still choose what to maindeck, even if all cards end up in someone's pool. Of course if you create a draft format where all the cube is seen but decks are worse than in an 8-player regular draft (say, 6 booster sealed with extra steps to pretend it's a draft), then effective removal density might actually be lower.

 

[Hemlock]

The non-land portion of my current (untested) list is ~22% removal. And ~2.5% sweepers.
Some of this is STP/Lightning Bolt level, though much of it is strong but conditional, or attached to a body or other effect and appropriately less efficient as removal. I am experimenting with running efficient spot removal but slightly less efficient sweepers and multi-target removal.

Good points about the value of evasion relative to removal. I have noticed that games often revolve around who can stick an unanswered flier. Something to keep an eye on.

 

[Brad]

Somewhat related question.

Assuming you play with less than a full pod (3-5 people in my case), but still draft the whole cube. Do you change the numbers to reflect the lower amount of players or do you just keep it as if there were 8 drafters (even if it won't happen)?

I am currently designing with a full pod in mind, but I don't know if this is right!

Each player is going to get 72-120 cards instead of 45? Seems unnecessary.

Disagree. Players will have access to a larger number of removal spells, so if the average removal is better than the average card they will likely play a larger proportion of them.

That's a big "if." I agree it'll gravitate more heavily towards whatever is supported, though.

 

[Nanonox]

Each player is going to get 72-120 cards instead of 45? Seems unnecessary.

Depending on number of players, drafters get between 48 and 56 cards (draft method explained in cube overview in my sig).

It is a lot, but it allows me to have more powerful decks without necessarily increasing power level of individual cards (thus tricking my drafters who want high powered game play, but avoiding what I consider GRBS).

 

[blacksmithy]

Depending on number of players, drafters get between 48 and 56 cards (draft method explained in cube overview in my sig).

It is a lot, but it allows me to have more powerful decks without necessarily increasing power level of individual cards (thus tricking my drafters who want high powered game play, but avoiding what I consider GRBS).

bigger pools are great! i do the same thing (pool of 54-56) to allow drafters to pick up 10+ lands for their deck and still have a decent sized pool. like you said, gives decks more power/consistency

 

[TrainmasterGT]

yeah if you draft the whole cube then everyone is going to get exactly the ratios of fixing/threats/removal/etc that are actually in the cube, regardless of how that gets distributed among players or maindecked. which i THINK is what train was saying too?

That's exactly what I was trying to say.

 

[Jason Waddell]

Disagree. Players will have access to a larger number of removal spells, so if the average removal is better than the average card they will likely play a larger proportion of them.

Whether you want to adjust for that depends on if you want to keep the cube balanced for a full pod or optimize it to drafted with 3-5 people.

Agree with Japahn here. Maybe it's easier if you visualize it as bombs in a retail environment. You'll have the same number divided among fewer drafters, this more bombs per drafter. It's definitely a dynamic change.

 

[shamizy]

I'm actually a big fan of deploying more conditional removal to create decision trees for my players or trying to figure out an cutting off point for certain effects. For example, I like "unconditional" white removal to start showing up around 3 CMC on average with stuff like Oblivion Ring or Council's Judgment. I have Swords and Path in my environment, and they are cream of the crop when it comes to removal at their CMC, but I'm not interested in flooding my cube with other like removal if we were to suddenly have a bunch of similar printings. I also like having wrath effects at 4 CMC on average with Wrath of God and Damnation being the clear signposts.

I think where the biggest struggle happens for me now is with black removal spells. There are just so many efficient options nowadays that there is just a ton of saturation. My current offerings include:



When Hero's Downfall came out I remember how big of a deal it was, mostly due to the ability to snipe walkers efficiently at instant speed. That felt like a good baseline for unconditional removal in black for both card types; 3 CMC instant speed for 1. Even with how pushed threats have become in recent years, that still feels like a nice example to work off and I've tried to mostly keep it that way. I like offering choices to my drafters with additional riders or potential options. Like yeah, edicts aren't all that great nowadays, but finding the situations where edicts can be used profitably on a given boardstate is really fun for me in a 1v1 setting. Offering more decision points when it comes to pointing removal at threats has just led to more engaging play.

I do think that removal should be regulated in most environments overall. I think just slotting in the best options and oversaturating your offerings will actually lead to more gameplay and deckbuilding issues down the road. I've had drafts where my brain is telling me to take the removal because it should be a premium ala Limited, but that's likely to not be the reality at the end of the draft. People will just pass removal with the expectation that more will come around or not be valued as highly. I think this is especially an issue if you just play the best universal removal options up and down your list.

To that end, I like having the tension of a few options and more conditional spells like Fatal Push against large threats or stuff like Bloodchief's Thirst and Burst Lightning offering the kicker mode. Having a clear hierarchy allows for the puzzle of pointing interaction at threats to be more nuanced. It's part of why I'm not particularly interested in the new super Doom Blade in Midnight Hunt or Baleful Mastery from Strixhaven. They're absolutely super efficient, but I'm not sure if that's exactly what I want at that particular slot. Like I'll jam Baleful Mastery in all my black EDH decks due to additional political utility with the card draw, but sometimes I just want to make sure that that black unconditional removal will reach my grindy U/B drafter instead of getting gobbled up by any B/x deck. To that end, Vraska's Contempt checks off more boxes for me as an inclusion.

 

[blacksmithy]

I do think that removal should be regulated in most environments overall. I think just slotting in the best options and oversaturating your offerings will actually lead to more gameplay and deckbuilding issues down the road.

i agree with what you said about having varying degrees of conditional removal. stuff like Bloodchief’s Thirst and Fatal Push allows you to give players a reward for drafting riskier cards like Titania that could otherwise result in huge tempo losses, while still allowing early interaction.
however!
i actually think most cubes don’t have enough removal, and you can see this as several people have bemoaned “games coming down to who stuck a flier.” this is almost 100% a sign that you should consider going up in removal overall as a designer, particularly if you desire “interactive gameplay.” having a strong variety of versatile removal options, such as the aforementioned plus shocks/bolts, tempo counters, oring type effects, allows players to choose how to interact with the opponent as the game progresses and exert agency (!!) over where on the spectrum they want to spend precious mana, between building a strong board and demanding an answer, landing a single threat and keeping shields up, or shutting down the enemy’s plans completely. if you play a BO3 format, having a high density of removal also helps players shift between more proactive and reactive gameplans within a match, as they likely ended up with some removal in the board instead of having to scrap for and maindeck whatever they could grab.

 

[shamizy]

Yeah, that can definitely be an issue. Sometimes you just need a certain density of Murder effects to just give enough tools to deal with threats, especially if combat isn't particularly interactive. I think the issue might just end up being the composition of threats that must be answered in order to progress.

Like realistically that majority of us aren't running too many cards that would usually serve as filler in a Limited deck since those designs aren't very compelling. Unless it's a cube with specific restrictions, I'd say the vast majority of inclusions are just going to end up being rares and mythics because that's where we get the most complexity. This creates a scenario where we feel compelled to answer more threats than we would in a typical draft deck. There are different tiers of power between rares in the cube that will be set dependent, but I'd say the complexity is mostly on par.

I think the pitfall then ends up being that adding all of these complex interactions between cards that synergize or work in tandem can end up churning out scenarios where a card must be answered before it gets out of hand. May not necessarily be a card that warps the game completely around itself. But maybe your opponent is getting a wee bit too much value from that Lotus Cobra and being able to recur fetches. Normally it'd be eh whatever, a 1 mana jump is good but fine. But a Black Lotus worth of mana with proper sequencing? That's too much, I've got to kill it quick.

If you can't partially mitigate that through combat or applying pressure forcing blocks, then you require actual answers via removal in hand to deal with it outright. Pretty much if you're just going to be running a whole lot of cards that will require acknowledgement on the regular, you should probably be running a lot of interaction. Within that realm, however, you probably want to introduce certain restrictions instead of catch-alls, especially if your drafts have a smaller number of participants.

Early on in cube I can't count the number of times I've had less than 8 people drafts where players just ended up with a bevy of removal and had so many pieces just sitting around in the sideboard. Not the worst problem to have, but it feels kind of bad when removal doesn't command a premium and everyone would rather just be greedy in draft. It's all just a matter of finding that balance in a particular environment. Personally I kind of like reining back the density at my power level and making my drafters prioritize accordingly.

 

[sigh]

I haven't taken the time to manually count and/or filter my cube yet, but I feel like I am also on the slightly lean side, and I also heavily favor conditional removal that encourages creativity in usage and drafting.

 

[japahn]

i actually think most cubes don’t have enough removal, and you can see this as several people have bemoaned “games coming down to who stuck a flier.” this is almost 100% a sign that you should consider going up in removal overall as a designer, particularly if you desire “interactive gameplay.”

I don't think high removal density is an absolute need, but it gives you (deisgner) a lot more leeway to keep threats in check, and players more leeway to not prioritize it in the draft.

If you don't run high removal density, you basically need to make sure creatures will not run away with the game by themselves, which implies a much lower power level than has been printed in the last 10 years for constructed. Even at this low power level, something as mundane as a Dauthi Horror or Catapult Squad can run away with the game.

It's just much easier to have a high density of removal. Interesting to see the numbers in the thread are pretty consistent around 25% of nonlands.

 

[blacksmithy]

I don't think high removal density is an absolute need, but it gives you (deisgner) a lot more leeway to keep threats in check, and players more leeway to not prioritize it in the draft.

If you don't run high removal density, you basically need to make sure creatures will not run away with the game by themselves, which implies a much lower power level than has been printed in the last 10 years for constructed. Even at this low power level, something as mundane as a Dauthi Horror or Catapult Squad can run away with the game.

It's just much easier to have a high density of removal. Interesting to see the numbers in the thread are pretty consistent around 25% of nonlands.

it really has been interesting to see! i love threads like this that give everyone some real numbers on what is happening in our cubes. (and i think i’m still the highest for removals at 29+% haha)

 

[sigh]

Ok I did my count and I seem to be at about 21.1%. A little lean, as I suspected. And I did count bounce, but did not count counterspells. Plenty of conditionality in my suite: