[First of all, DISCLAIMER: it's just a theoretical debate, nothing else. It's a topic of interest to me, so I'd like to collect feedback on it.]
It's not about the power level. We don't have a Human problem. I'm not worried if there are 5 Human decks per week (and there aren't), since they can be (and mostly are) all different enough, for intrinsic reasons. And, like I said, I think that if Human ceased to exist as a tribe, you could just turn the more power-hungry of those decks into 4-color Wizards, which we already see plenty of. So it's not about "solving" an issue. Or even fixing a flavor according to personal criteria.
What is it about then? It's exactly what RexDart wrote: it's a Melvin thing. The Human type is clearly an anomaly in the game. I won't be surprised if at some point, should one of the rule gurus feel like that, the Human sub-type would really be moved to super-type. Mechanically, nothing would change. It would only affect Legacy Tribal Wars games. (Which is why I don't see it happen anytime soon.) But it would be about putting things in order and correct the anomaly.
How I know it's an anomaly? Well, the numbers speak for themselves, as does the inherent definition of super-type (either an attribute or a larger group defining a characteristic that's shared by different, lesser groups of items.) But think about this: most Zombies should have the Human type. We see that Zombie isn't treated as a race, it's a state: they keep their race's type on the card. Jarad was an Elf, died, was raised from the dead, he's now a Zombie Elf. Same as Glissa. Balthor was a Dwarf, he's a Zombie Dwarf. Helldozer is a Zombie Giant. And so on. In MTG universe, zombified things remain what they were, only in zombified form. But what about Mikaeus? There's a version of him who actually has the Human type. Yet the Zombie version loses it. Same goes for Geth or Haakon: the official background stories for these characters tell us that they were Humans. I'd say 70%/80% of the Zombie cards actually depict Humans turned Zombies. Why they don't have the Human type then? What if they will at some point? Once upon a time, all legendary creatures were only "legends", without any other type. Then they fixed it.
And here's what Mark Gottlieb was writing during the Grand Creature Type Update, which is 6-year-old now: "Creatures that lacked races got a race. The vast majority of these are creatures that should be Human but were printed before Mirrodin. In fact, more than half of the cards involved in this update got 'Human' added to them."
The vast majority. More than half. They didn't do this to help the Human tribe as defined in Tribal Wars. I'm sure they didn't even think of Tribal Wars at the time. They did it because it felt right. Those were clearly humans on the cards. It's a fantasy game based (like most) on our folklore: of course the basic state of a character is humankind. We're freaking human ourselves, and so are all the WotC creatives (even MaRo!) So, what if at some point they'll do another update during which they'll realize Mikaeus and co. should get the Human type too? We'd have a tribe where you can accommodate a proper Zombie deck with 4 Dark Confidants inside the tribal base.
Given the current rate of growth, I think 10 blocks from now half the existing creatures in the game will have the Human type. Building them as a tribe it'll be halfway from building a "Creature Tribal deck" (currently, we're at one quarter of the way.) And way less ridiculous than building a "Legendary Creature Tribal deck" or an "Artifact Creature Tribal deck", since the former are only 490 and the latter 449. Hell, all the cards with the Artifact type, both creature and non-creature, are 1446, just a little more than the number of Humans. Now, that's ridiculous.
Yes Metamorph on Mage, and though I don't remember all the details, possibly naming Magus of the Unseen since Metamorph is still an artifact. Not too important in the scenario, though, as I'm pretty sure my Morph-Mage trades next turn when you attack.
One sure way to make sure I never associate with the apocalypse again is to ban the Humans tribe. Sure if there is something special going on by all means stick them in the no go section, otherwise this is a ridiculous idea.
Edit:
Perhaps I should clarify my objection.
I think that banning an entire tribe (particularly such a large one) is an alienating act. I don't think it solves anything flavorwise. People will still build against the spirit of the format if that is their will. Those who do so, don't do so because they are flaunting the idea but because they find certain decks win more consistently. If you want to calm the power level of humans down you can ban certain ones that seem like trouble spots (but I am not for this either.)
If it isn't as you say about the power level then you need to keep sculpting the scene to be more flavor friendly and rewarding creativity such as with the achievements system which is an excellent innovation.
Banning humans because they are ubiquitous would just flip those who abuse them back to gobbos or pointy-eared treehuggers or some other tribe that lends itself to this kind of abuse.
I think it would be wonderful instead if you had a humans only day with a top 10 ban list involving the top most played humans. And then put them on no go list for special days.
Off the top of my head, one of the most recent Human-only creatures is Seller of Songbirds. One of the most powerful is Peacekeeper. You can still build a nice deck with Human-only creatures. Maybe it can be an achievement at some point.
That is an interesting post.In relation from it, H&R Block admitted to making an error in the filing of 600,000 returns seeking education tax credits. That delay could affect the federal financial aid of some of those taxpayers as well. A short term loan can help you pay for the things you need while waiting on a return.
"I give you Crabs" had me chuckling more than it deserved, ha.
On the topic of Humans as a tribe...
When I first got into this format a couple years ago, I didn't consider Humans a "real" tribe. I have played them as a tribe a few times, sometimes with decks that had alot of flavor or actual use for the Human creature type (i.e. Mayor of Avabruck, the Deranged Outcast, etc). But sometimes it's just the most flexible tribe to use when adapting some legacy goodstuff deck to tribal. Our combo-oriented players sometimes use them as the easiest way to bring their legacy combo decks into tribal as well. I don't know if that's really the "point" of tribal, but that's an argument that never goes anywhere but in circles.
Personally, despite having played them a few times, if I were the grand exalted ruler of the Apocalypse I would eliminate Humans as a legal tribe. It adds an additional split between the Apocalypse and WotC's legacy tribal wars rules, but the players seem to be okay with the extensive splits we already have, so that objection doesn't go too far. Humans are perhaps a bit too flexible and lack character, and I agree with Kuma that they feel more like a "supertype" than a real tribe. I do concede they have one actual limitation despite the size of the tribe, which is the lack of larger creatures, but that is also true of many other tribes and is hardly that distinctive. I would say give 'em the boot, but I am totally upfront about the fact that this is just my personal preference and is for aesthetic and flavor reasons as much as anything.
I would ask first, however, whether anybody knows if there are creatures with ONLY Human as a creature type? Creatures like that would be effectively stranded if we do this.
An ORC posted a link to a beta client survey. I lost the link, but they can probably find it if you want to fill it out. So far, the biggest issue with the beta client is crashing.
This is exactly the question that I have about the data. I didn't really address it, because I simply don't have an answer for this, and it is much more complicated than what I can predict statistically.
I should mention, this kind of approach doesn't actually invalidate the data. In many ways, I suspect that it actually reinforces the idea that popularity is very bad at predicting how much a deck is over or under drafted.
The problem is that I haven't figured a way to figure out how this works statistically. Looking at the three different queues is an interesting proposition, but I worry that there may be more variables changing between the different queues. Do some queues favor decks that are easy to play? Does that change the way the results play out? I'll have to think about a way to run these kinds of numbers, but this is exactly the kind of question that I appreciate so much on my columns; the kind that take a thoughtful and multi-faceted look at the statistics.
Great Article Mikey. More than any other time, I really feel like there is a correlation between your writing of this primer and the meta blowing up with the deck.
Another possible explanation for popular decks continuing to do well could be to theorize "the community is mostly fairly good about jumping out of an archetype if it isn't open enough from their seat". The data can't distinguish between a draft where five people pursued a guild in pack one but three of them switched out, versus a draft where two people went for it and stayed there.
In some sense the guild that gets tried but abandoned a lot is more popular than one where this doesn't happen often. But the correct winnowing down to a number of drafters the carpool can support keeps us from seeing the dilution of win rate we would expect if 4-5 drafters stuck it out to the bitter end with any frequency.
It might be telling to compare results between 8-4s, 4-3-2-2s, and Swiss drafts to see if overdrafting has more impact in some queue types. I will say anecdotally I have the impression that core set Swiss drafts will more often see a color severely over or under drafted than I see In other types of queues. This is just an impressing, though, I haven't applied any statistical rigor to the question.
Great article as always. Good news on the Savannah and I'm happy for another you make the card as I remember those from when I was a casual player. The trading bug is just horrible.
On another matter: I don't know where to say this so I'll say it here, my article wasn't published yesterday so I expected it to be today and since only 2 were published today I don't see any reason why it shouldn't. I didn't get it back either so it was probably just a mistake.
I'm not sure that matters. In the end you had a finite number of decks between the two columns, and in MTGO the average win rate of a closed set of decks (in the sense they only play against other decks within the set) is 50%, so the two columns should still average to 50%. The only thing I can think of to explain the difference if you are including wins for an archetype when it played against a deck that is not in the archetypes (non-guild decks, which would make the chart not a closed set).
Of course, had my Duress been a Misstep, I would have won regardless (assuming he attacked as he did). As A.J. indicated, he had some narrow escapes, and I got screwed/then flooded in subsequent games, which is partly deck design error I think, but my draws certainly could have been more forgiving. In g.2, A.J., your Misstep stopped my turn 1 kill which was otherwise a weak draw. Your Mages were fairly decent at hitting outs, but were ineffective inasmuch I didn't draw them. I was thinking your g.2 Mage should have named Metamorph over Dismember, and sure enough I even then very luckily drew the Metamorph (for which, luckily for you, you had the Force to make the question moot), but I suppose you couldn't have assumed I would have boarded it in. Both were 1-ofs but Metamorph was more tutorable, and more dangerous. I sent you my g.2 draws afterwards for the match for your notes. My g.3 draws were I think 9 lands/mana sources, one spell; I could have hardcast a Colossus on turn 8 (though Magus of the Unseen would have had a field day had I done so).
I would not have attacked with the Magus had I not drawn the Misstep. (At least, I know I spent some time thinking and talking about that.) 1 damage wouldn't be worth the potential to disrupt a combo.
You could well be right. I don't think there's a whole lot in it but Figure might be marginally better, particularly for dodging bolts and helix as you mentioned.
Yes, Zach, I was going to play Vault, Key, activate all in one turn - had the combo (with Enlightened Tutor played during upkeep) and the mana (with Workshop in hand) when he was tapped out. The Misstep on the exact turn he needed to draw it won. To be fair, he may not have attacked with the Magus had he not drawn the Misstep.
Actually, the explanation is pretty easy. You made the incorrect assumption that both sides have an equal number of total decks.
The columns represent an equal number of total archetypes. Those archetypes have different popularity ratios. There are more decks in column one than in column two. This accounts for the discrepancy.
I generally agree with and understand everything you are getting at here but there is one thing bothering me which has some small implications.
In the first chart, as far as the explanation goes, it would seem every deck is in consideration { one side being (0,1) the other [1,2) } but the win rates do not add to 1. There are no draws in MTGO so that wouldn't be the distortion, so the chart seems to imply ~1.5% of the wins just up and vanished.
I see Bombardment more of a combo piece that I'd like to avoid. It works well with Tombstone Stairwell and Gravecrawler, that's about it. If it synergized better with the rest of the deck I'd run it, like I do in my Sek'Kuar deck. Vengeful Dead is even more nasty with Stairwell while synergizing with the entire deck very well.
Gravecrawler is something I'll have to try out sometime. From what I can tell, it synergizes with Corpse Harvester and Grim-Grin. I don't know if that itself warrants inclusion but at least a test. Skullclamp has very few synergies, just Gravecrawler and Corpse Harvester, Goblin Bombardment if I run it too. If I had multiple 1 toughness creatures I'd run it in a heartbeat. I just don't think it's the right deck.
About 600 decks. Which is about 700 games. It takes about that much in order to get enough statistically relevant data. I mean, I could technically find out a lot of useful data off of closer to 400 decks, but I'd be missing out on data for the less drafted archetypes. Specifically, I need 600 decks to get enough data points to cover the splashing data.
I also need to clarify (further), then.
[First of all, DISCLAIMER: it's just a theoretical debate, nothing else. It's a topic of interest to me, so I'd like to collect feedback on it.]
It's not about the power level. We don't have a Human problem. I'm not worried if there are 5 Human decks per week (and there aren't), since they can be (and mostly are) all different enough, for intrinsic reasons. And, like I said, I think that if Human ceased to exist as a tribe, you could just turn the more power-hungry of those decks into 4-color Wizards, which we already see plenty of. So it's not about "solving" an issue. Or even fixing a flavor according to personal criteria.
What is it about then? It's exactly what RexDart wrote: it's a Melvin thing. The Human type is clearly an anomaly in the game. I won't be surprised if at some point, should one of the rule gurus feel like that, the Human sub-type would really be moved to super-type. Mechanically, nothing would change. It would only affect Legacy Tribal Wars games. (Which is why I don't see it happen anytime soon.) But it would be about putting things in order and correct the anomaly.
How I know it's an anomaly? Well, the numbers speak for themselves, as does the inherent definition of super-type (either an attribute or a larger group defining a characteristic that's shared by different, lesser groups of items.) But think about this: most Zombies should have the Human type. We see that Zombie isn't treated as a race, it's a state: they keep their race's type on the card. Jarad was an Elf, died, was raised from the dead, he's now a Zombie Elf. Same as Glissa. Balthor was a Dwarf, he's a Zombie Dwarf. Helldozer is a Zombie Giant. And so on. In MTG universe, zombified things remain what they were, only in zombified form. But what about Mikaeus? There's a version of him who actually has the Human type. Yet the Zombie version loses it. Same goes for Geth or Haakon: the official background stories for these characters tell us that they were Humans. I'd say 70%/80% of the Zombie cards actually depict Humans turned Zombies. Why they don't have the Human type then? What if they will at some point? Once upon a time, all legendary creatures were only "legends", without any other type. Then they fixed it.
And here's what Mark Gottlieb was writing during the Grand Creature Type Update, which is 6-year-old now: "Creatures that lacked races got a race. The vast majority of these are creatures that should be Human but were printed before Mirrodin. In fact, more than half of the cards involved in this update got 'Human' added to them."
The vast majority. More than half. They didn't do this to help the Human tribe as defined in Tribal Wars. I'm sure they didn't even think of Tribal Wars at the time. They did it because it felt right. Those were clearly humans on the cards. It's a fantasy game based (like most) on our folklore: of course the basic state of a character is humankind. We're freaking human ourselves, and so are all the WotC creatives (even MaRo!) So, what if at some point they'll do another update during which they'll realize Mikaeus and co. should get the Human type too? We'd have a tribe where you can accommodate a proper Zombie deck with 4 Dark Confidants inside the tribal base.
Given the current rate of growth, I think 10 blocks from now half the existing creatures in the game will have the Human type. Building them as a tribe it'll be halfway from building a "Creature Tribal deck" (currently, we're at one quarter of the way.) And way less ridiculous than building a "Legendary Creature Tribal deck" or an "Artifact Creature Tribal deck", since the former are only 490 and the latter 449. Hell, all the cards with the Artifact type, both creature and non-creature, are 1446, just a little more than the number of Humans. Now, that's ridiculous.
Yes Metamorph on Mage, and though I don't remember all the details, possibly naming Magus of the Unseen since Metamorph is still an artifact. Not too important in the scenario, though, as I'm pretty sure my Morph-Mage trades next turn when you attack.
One sure way to make sure I never associate with the apocalypse again is to ban the Humans tribe. Sure if there is something special going on by all means stick them in the no go section, otherwise this is a ridiculous idea.
Edit:
Perhaps I should clarify my objection.
I think that banning an entire tribe (particularly such a large one) is an alienating act. I don't think it solves anything flavorwise. People will still build against the spirit of the format if that is their will. Those who do so, don't do so because they are flaunting the idea but because they find certain decks win more consistently. If you want to calm the power level of humans down you can ban certain ones that seem like trouble spots (but I am not for this either.)
If it isn't as you say about the power level then you need to keep sculpting the scene to be more flavor friendly and rewarding creativity such as with the achievements system which is an excellent innovation.
Banning humans because they are ubiquitous would just flip those who abuse them back to gobbos or pointy-eared treehuggers or some other tribe that lends itself to this kind of abuse.
I think it would be wonderful instead if you had a humans only day with a top 10 ban list involving the top most played humans. And then put them on no go list for special days.
Off the top of my head, one of the most recent Human-only creatures is Seller of Songbirds. One of the most powerful is Peacekeeper. You can still build a nice deck with Human-only creatures. Maybe it can be an achievement at some point.
That is an interesting post.In relation from it, H&R Block admitted to making an error in the filing of 600,000 returns seeking education tax credits. That delay could affect the federal financial aid of some of those taxpayers as well. A short term loan can help you pay for the things you need while waiting on a return.
There are a few. Ali From Cairo and Angry Mob. Mostly, if they're only human, they're in older sets like Homelands.
"I give you Crabs" had me chuckling more than it deserved, ha.
On the topic of Humans as a tribe...
When I first got into this format a couple years ago, I didn't consider Humans a "real" tribe. I have played them as a tribe a few times, sometimes with decks that had alot of flavor or actual use for the Human creature type (i.e. Mayor of Avabruck, the Deranged Outcast, etc). But sometimes it's just the most flexible tribe to use when adapting some legacy goodstuff deck to tribal. Our combo-oriented players sometimes use them as the easiest way to bring their legacy combo decks into tribal as well. I don't know if that's really the "point" of tribal, but that's an argument that never goes anywhere but in circles.
Personally, despite having played them a few times, if I were the grand exalted ruler of the Apocalypse I would eliminate Humans as a legal tribe. It adds an additional split between the Apocalypse and WotC's legacy tribal wars rules, but the players seem to be okay with the extensive splits we already have, so that objection doesn't go too far. Humans are perhaps a bit too flexible and lack character, and I agree with Kuma that they feel more like a "supertype" than a real tribe. I do concede they have one actual limitation despite the size of the tribe, which is the lack of larger creatures, but that is also true of many other tribes and is hardly that distinctive. I would say give 'em the boot, but I am totally upfront about the fact that this is just my personal preference and is for aesthetic and flavor reasons as much as anything.
I would ask first, however, whether anybody knows if there are creatures with ONLY Human as a creature type? Creatures like that would be effectively stranded if we do this.
An ORC posted a link to a beta client survey. I lost the link, but they can probably find it if you want to fill it out. So far, the biggest issue with the beta client is crashing.
What would you have named with Metamorph? I assumed FoW.
This is exactly the question that I have about the data. I didn't really address it, because I simply don't have an answer for this, and it is much more complicated than what I can predict statistically.
I should mention, this kind of approach doesn't actually invalidate the data. In many ways, I suspect that it actually reinforces the idea that popularity is very bad at predicting how much a deck is over or under drafted.
The problem is that I haven't figured a way to figure out how this works statistically. Looking at the three different queues is an interesting proposition, but I worry that there may be more variables changing between the different queues. Do some queues favor decks that are easy to play? Does that change the way the results play out? I'll have to think about a way to run these kinds of numbers, but this is exactly the kind of question that I appreciate so much on my columns; the kind that take a thoughtful and multi-faceted look at the statistics.
I know Joshua is busy, and I dumped my article on him at the very last minute (okay, several hours after the last minute.) Hopefully he'll get to it.
Great Article Mikey. More than any other time, I really feel like there is a correlation between your writing of this primer and the meta blowing up with the deck.
Another possible explanation for popular decks continuing to do well could be to theorize "the community is mostly fairly good about jumping out of an archetype if it isn't open enough from their seat". The data can't distinguish between a draft where five people pursued a guild in pack one but three of them switched out, versus a draft where two people went for it and stayed there.
In some sense the guild that gets tried but abandoned a lot is more popular than one where this doesn't happen often. But the correct winnowing down to a number of drafters the carpool can support keeps us from seeing the dilution of win rate we would expect if 4-5 drafters stuck it out to the bitter end with any frequency.
It might be telling to compare results between 8-4s, 4-3-2-2s, and Swiss drafts to see if overdrafting has more impact in some queue types. I will say anecdotally I have the impression that core set Swiss drafts will more often see a color severely over or under drafted than I see In other types of queues. This is just an impressing, though, I haven't applied any statistical rigor to the question.
Great article as always. Good news on the Savannah and I'm happy for another you make the card as I remember those from when I was a casual player. The trading bug is just horrible.
On another matter: I don't know where to say this so I'll say it here, my article wasn't published yesterday so I expected it to be today and since only 2 were published today I don't see any reason why it shouldn't. I didn't get it back either so it was probably just a mistake.
I'm not sure that matters. In the end you had a finite number of decks between the two columns, and in MTGO the average win rate of a closed set of decks (in the sense they only play against other decks within the set) is 50%, so the two columns should still average to 50%. The only thing I can think of to explain the difference if you are including wins for an archetype when it played against a deck that is not in the archetypes (non-guild decks, which would make the chart not a closed set).
Of course, had my Duress been a Misstep, I would have won regardless (assuming he attacked as he did). As A.J. indicated, he had some narrow escapes, and I got screwed/then flooded in subsequent games, which is partly deck design error I think, but my draws certainly could have been more forgiving. In g.2, A.J., your Misstep stopped my turn 1 kill which was otherwise a weak draw. Your Mages were fairly decent at hitting outs, but were ineffective inasmuch I didn't draw them. I was thinking your g.2 Mage should have named Metamorph over Dismember, and sure enough I even then very luckily drew the Metamorph (for which, luckily for you, you had the Force to make the question moot), but I suppose you couldn't have assumed I would have boarded it in. Both were 1-ofs but Metamorph was more tutorable, and more dangerous. I sent you my g.2 draws afterwards for the match for your notes. My g.3 draws were I think 9 lands/mana sources, one spell; I could have hardcast a Colossus on turn 8 (though Magus of the Unseen would have had a field day had I done so).
I would not have attacked with the Magus had I not drawn the Misstep. (At least, I know I spent some time thinking and talking about that.) 1 damage wouldn't be worth the potential to disrupt a combo.
Thanks for the comment.
You could well be right. I don't think there's a whole lot in it but Figure might be marginally better, particularly for dodging bolts and helix as you mentioned.
Yes, Zach, I was going to play Vault, Key, activate all in one turn - had the combo (with Enlightened Tutor played during upkeep) and the mana (with Workshop in hand) when he was tapped out. The Misstep on the exact turn he needed to draw it won. To be fair, he may not have attacked with the Magus had he not drawn the Misstep.
Actually, the explanation is pretty easy. You made the incorrect assumption that both sides have an equal number of total decks.
The columns represent an equal number of total archetypes. Those archetypes have different popularity ratios. There are more decks in column one than in column two. This accounts for the discrepancy.
I generally agree with and understand everything you are getting at here but there is one thing bothering me which has some small implications.
In the first chart, as far as the explanation goes, it would seem every deck is in consideration { one side being (0,1) the other [1,2) } but the win rates do not add to 1. There are no draws in MTGO so that wouldn't be the distortion, so the chart seems to imply ~1.5% of the wins just up and vanished.
Figure of Destiny over Student of Warfare? Better at dodging Bolts/helix
Good point. Luckily I didn't really need it, but it's worth knowing that I had another way of getting around Kira's ability to protect her creatures.
I see Bombardment more of a combo piece that I'd like to avoid. It works well with Tombstone Stairwell and Gravecrawler, that's about it. If it synergized better with the rest of the deck I'd run it, like I do in my Sek'Kuar deck. Vengeful Dead is even more nasty with Stairwell while synergizing with the entire deck very well.
Gravecrawler is something I'll have to try out sometime. From what I can tell, it synergizes with Corpse Harvester and Grim-Grin. I don't know if that itself warrants inclusion but at least a test. Skullclamp has very few synergies, just Gravecrawler and Corpse Harvester, Goblin Bombardment if I run it too. If I had multiple 1 toughness creatures I'd run it in a heartbeat. I just don't think it's the right deck.
About 600 decks. Which is about 700 games. It takes about that much in order to get enough statistically relevant data. I mean, I could technically find out a lot of useful data off of closer to 400 decks, but I'd be missing out on data for the less drafted archetypes. Specifically, I need 600 decks to get enough data points to cover the splashing data.