General LuckyPaper Kaldheim Community Survey Results - See what the community thinks of Kaldheim!

[sigh]

Here is where I am at. I'm not good at writing things eloquently so here are some points.

• Multicolored cards are narrow
• If they aren't better than monocolored cards I'd rather not play them and instead run more cards from either color
• You can hint at what you want the color to do through density of effects in each color (maybe more suitable for more experienced drafters)

I agree with this, IF the multicolored cards are worth it (my second point). Here is what it takes for me to consider adding a gold card

• Random beaters? I feel they need to be exceptional to be worth a slot (Figure of Destiny as a scalable one drop)
• Unique effect that helps the color pair's weakness (or Showdown of the Skalds as card draw)
• Cool build around that you cannot get elsewhere or that spawns an archetype on it's own (Feather, the Redeemed)

One compelling argument I am ignoring is that the multicolored cards help guide your players through the draft. Depending on your playgroup this could be a big one. But you know your group better than anybody so adjust accordingly.
Another is that what I find to be an exciting build around might be different from you. So maybe the Swiftblade Vindicator is the nuts and I just see it as a random beater.
Hope this makes sense!

Monocolor cards have much more competition in drafting if they aren't very specific in function, which makes just slotting in a monocolor version of a card not nearly as good for a specific multicolor deck as providing the multicolor deck the card directly in it's colors. (See ravnic's post for the opposite and just as bad downside of narrow monocolor cards).

Adorned pouncer is attractive to

• WB weenies
• RW double strike
• GW +1/+1 counters
• UW tempo

And that's only direct color pairs. I want the adorned pouncer effect available directly to RW specifically. Pouncer can't do that. So, like you state, you have to run densities of effects to make up for it. But then am I running 4 adorned pouncers? Why, when I can run the multicolor card that reserves itself perfectly for the slot I'm trying to fill in the deck I want it to be in. Including the pouncer is still a good idea because it's good to have >1 of that effect in the deck and/or to have that effect available to other decks, but they aren't directly comparable or shouldn't even really be discussed about supplanting each other IMO. I'm providing the RW deck an effect it needs. I'm ok with it being narrow. I want it to be narrow. I will actively provide the deck with less effects it needs if I instead seek out a "cooler" or "higher power" card just for being cooler/higher power. Showdown of the Skalds is so widely good it's as likely to get splashed for in a midrange deck that wants the draw than it is actually finding it to the RW deck. Is that really what I want? I checked the high power box, but why?
If I had a heroic deck, I'd have no qualms with including Feather. She would be perfect for the deck in that instance. But I don't, and trying to run her because she's cool would just require me to muddy my entire boros section to get her to work.

Edit: I still may run the Showdown. The card is very good, and if the deck shows it needs later game fueling, it's a great choice. I'm trialing Akiri and/or Winota as a mid-late game way to fuel the deck currently, and haven't seen a strong need to move to the Showdown yet.

For my other slot, I'm strongly considering

as a card that can provide a little broader archetype support (I'd probably splash red to get her in my UW flier deck, for instance), while still providing lots of things that the RW deck is looking for: a body that holds equipment well, not just one but two pump effects in lieu of or in addition to equipment, and some evergreen keywords as garnish on top.

 

[Nanonox]

I understand where you are coming from, but I have a different view I guess.

If the Pouncer in your example is valuable to all of the white color pairs then:
1. Your cube is well built, I love it!
2. All those color pairs probably want more than 1.

With Swiftblade, you guarantee that Boros gets at least one, I get it. I am saying that I don't care that the Boros deck doesn't get there this draft, it happens. If that means that the GW counter deck did (using your other white double striker instead of Swiftblade) that's still a win.

Take a random example (100% not random, I cheery picked the hell out of it ):

Pick 1: Bonesplitter
Pick 2: Adorned Pouncer
Pick 3: Solemn Recruit (instead of Swiftblade Vindicator)
Pick 4: Now you see Krenko, Tin Street Kingpin or Rishkar, Peema Renegade

Now you have a choice! Do you branch into red or green. Red wants equipment or Reckless Charge to boost power, green you go counters.
I'd argue that this is a more interesting draft than if your 3rd pick is Swiftblade Vindicator since it puts you in a lane and you wouldn't even consider the Rishkar.

Now replace Krenko with Akiri, Fearless Voyager. This, for me is a better gold card than Swiftblade (if it is powerful enough in your environment). It's narrow since only the Boros equipments want it and it will put in good work there, but that's ok. There isn't any demand for the equipment payoff elsewhere. Plus it signals better what the guild is doing.

I don't know how to explain it, but for me the example of Swiftblade Vindicator is the wrong type of narrow for your guild slot (similar to what Dom was saying about Lightning Helix).

Please tell me to stop if I am misunderstanding what you are saying!

 

[sigh]

unfortunately I'm not seeing any difference between the Vindicator and Akiri examples, especially since they are custom tailored to make the point, which is fine!. It's just that I could as easily swap the Vindicator to pick 4 so you have the two double strikers, but are now weighing strengthening the double strike theme even further or going to a counter/double strike GW deck. Like so:

Pick 1: Bonesplitter
Pick 2: Adorned Pouncer
Pick 3: Solemn Recruit
Pick 4: Now you see Swiftblade Vindicator or Rishkar, Peema Renegade

I would be happy with my drafter going either of these routes, and both seem good. Akiri fits here too, but I'm leaning away from her after testing, to something with a little more oomph like Aurelia or Winota.

Or we switch it up to something like:

Pick 1: Bonesplitter
Pick 2: swiftblade vindicator

Which puts us on the Boros double strike path, but doesn't lock us into it. It's only the second pick! If I see a bomb pick 3 I could pivot with virtually no loss in draft performance. If I see the Solemn Recruit, I know I'm on a good path.

And these contrast with the more-likely scenario of it coming up picks 16-45, and at that point it's a perfect fit for the equipment deck that is well under way, but missed out on the Adorned Pouncer to a WB deck a couple seats down.

But we could make little custom scenarios all day, and I don't think further attempts will progress things much. I understand your position and hope you understand mine. I want people to consider allowing themselves to pick solid signpost cards even if slightly unassuming, if they have decks with a clear desire to have that effect and running the card gives that deck ready access to the effect. Basically: In my opinion, we shouldn't be so dead set on finding all-star standouts if the decks themselves aren't asking for those standouts.

 

[Nanonox]

unfortunately I'm not seeing any difference between the Vindicator and Akiri examples

The Vindicator effect can be found in mono-colored but not the Akiri, that's it.

I do understand your position and will take some time to reflect on the ramifications and see if I can adjust my view, which is basically:
Don't use a gold slot for an effect you can get in your mono-colored sections unless you cannot get the density you need.

Enjoyed the talk, thanks!

Random thought: maybe this connects to broad vs more detailed archetype support?

 

[sigh]

The Vindicator effect can be found in mono-colored but not the Akiri, that's it.

Maybe this is the case, but I'm not so sure. The example of Adorned Pouncer has been easy to discuss due to the casting cost similarity, but Double Strike + Trample + Vigilance is a lot better than plain double strike, especially when pushing damage past tokens or when racing an aggro mirror. It's also a combination of abilities that is only mirrored by one monocolored card (Zetalpa, Primal Dawn, which does not compete for the same slot at all)

Don't use a gold slot for an effect you can get in your mono-colored sections unless you cannot get the density you need.

That's one way to word it, but I think of it more with inverse wording:

"Be open to running an effect in a guild slot if a deck wants to ensure it has that effect, unless you have a way to ensure you can provide that effect reliably in monocolored slots."

Random thought: maybe this connects to broad vs more detailed archetype support?

Possibly. In this case I probably wouldn't stick to finessing one slot, as most multicolored sections have at least 2, often more, cards per guild. Some of those can be focused, and some benefiting broader spectrums of decks. Again Aurelia as my example second slot alongside Vindicator.

 

[japahn]

I am actually about to increase my mana fixing options because the drafts done on my cube (I've been getting a few since it's featured) indicate that players are not ending up with enough lands

Don't base your decisions about fixing on CubeCobra bots. They take nonbasic lands very, very highly. Take a look at the pick-by-pick breakdown of bot seats. There are other ways CubeCobra bots are pretty skewed, but this is the most glaring one. Perhaps the current weights work well for high power cubes, but they don't scale realistically for lower power.

 

[Onderzeeboot]

Don't base your decisions about fixing on CubeCobra bots. They take nonbasic lands very, very highly. Take a look at the pick-by-pick breakdown of bot seats. There are other ways CubeCobra bots are pretty skewed, but this is the most glaring one. Perhaps the current weights work well for high power cubes, but they don't scale realistically for lower power.

Well, everybody should be taking fixing high in my cube, I think. It is a multicolor cube after all.

 

[TrainmasterGT]

Don't base your decisions about fixing on CubeCobra bots. They take nonbasic lands very, very highly. Take a look at the pick-by-pick breakdown of bot seats. There are other ways CubeCobra bots are pretty skewed, but this is the most glaring one. Perhaps the current weights work well for high power cubes, but they don't scale realistically for lower power.

Well, everybody should be taking fixing high in my cube, I think. It is a multicolor cube after all.

I think taking fixing highly is correct in most cubes, reasonably speaking. The average fail rate of having a card that is entirely not worth playing in a cube is far, far lower than in a retail limited environment. There are often going to be several cards during the draft that are good enough to splash but aren't in the main color(s) one is drafting. Taking a powerful piece of fixing early is usually a good move in the vast majority of environments.

The big issue with the CubeCobra bots is that they eat up lands for color combinations that they aren't reasonably playing. You might see a base U/W bot take a Bayou even though it's probably not going to play one of those two colors. I think this is mostly because the computer isn't sophisticated enough to understand the difference between a land that is absolutely vital to the function of the deck and a land that probably won't help it cast any of it's spells. There is also probably an Elo element involved, as a card like Bloodstained Mire is a high pick in even powered cubes because it's that useful of a card.

However, not seeing fetchlands ever passed in pack 1 or not generally seeing lands table is not a sign of the bots taking lands too highly. Rather, the machines are simply always trying to optimize a pick, and it's rare that the best card in a pack isn't the land.

Now, it is true that human drafters aren't always going to take a land out of a pack, either because they have a hard time recognizing when a powerful land is the correct pick, or because they recognize a power outlier or build around card that could contribute more to their deck or personal enjoyment than good fixing. I don't know if there is a way to solve for that variance on a computer without implementing some sort of randomization element that would purposefully make the bots ignore the lands and pick another card from the pack.

The point I'm trying to make here is that the CubeCobra bots taking lands highly is more of a feature than a bug, because fixing should be a top priority during a cube draft. Even in a low power cube, it's worthwhile to have a deck that can play and cast all of its best cards consistently as opposed to a deck that has a bunch of powerful stuff but runs into consistent mana issues. While the CubeCobra bots might take lands more highly than they should, not seeing enough fixing during a test draft is still a good indicator that the addition of more lands is warranted.

 

[Erik Twice]

Some time ago, Cube Cobra was awful for testing because you could take as many lands as you wanted. The average player ignores lands and, hence, the bots did the same.

Apparently the developers went hard and bumped the priority of lands artificially to try and solve this issue. However, it works a bit too well and now they take it too aggressively. It's better, but not perfect.

In general, I feel Cube Cobra bots have a very weak memory. The moment I don't draft an archetype or colour a couple times I start seeing stuff like last-pick Garruk Wildspeaker.

 

[Aston]

That's part of why I've always been sceptical of Lightning Helix even though it was the best of a bad bunch for Boros cards for a long time - it's not better enough than whichever Incinerate, Searing Spear, Lightning Strike etc is next on the list to make up for how much more narrow it is

This is true, but at the same time screaming "OH MY GOD IT'S SHOWDOWN OF THE SKALDS" when you topdeck it just doesn't hit the same

 

[Onderzeeboot]

Some time ago, Cube Cobra was awful for testing because you could take as many lands as you wanted. The average player ignores lands and, hence, the bots did the same.

Apparently the developers went hard and bumped the priority of lands artificially to try and solve this issue. However, it works a bit too well and now they take it too aggressively. It's better, but not perfect.

In general, I feel Cube Cobra bots have a very weak memory. The moment I don't draft an archetype or colour a couple times I start seeing stuff like last-pick Garruk Wildspeaker.

Someone's currently working on a series of articles explaining the Cube Cobra draft bots.

https://cubecobra.com/content/article/5f946f71e8f97310047fe794
https://cubecobra.com/content/article/6008b17565264010480a4a84

They're only 2 articles in of a planned 5, but the lands problem is actually discussed at the end of the second article. I just left some of my own thoughts, maybe you can add yours

 

[sigh]

I added some thoughts.

Agree fully that lands need to have additional considerations. I figured that the drafting bots looked at color identity and tried to stick to that, but it seems entirely not the case for lands, which is odd. ELO should handle rankings of ETB tapped vs. not, but it might still be worth it to have a knob to tweak in that regard.

I'm no programming whiz by any means, but from what I can see, one of the only ways to handle the A-B synergy problem is to train a bot on finished decks, not draft inputs, so the drafting bot can understand what cards end up in decks together and which don't. This is, after all, a large part of how us humans draft. We usually try and angle towards having a known deck at the end of the draft, not just a pile of cards that happen to be in a couple colors.