What do you guys all think about the episode, and the information presented? Did it change your mind about anything? Were there any lightbulbs that went off for you?
For me, I only had one surprise. I think Superfriends and planeswalkers in general are really weak. I still think that the stats are wrong on this one due to low sample size, but that was my surprise.
For the Sol Ring, I don't always put it in my decks. I know. Sacrilege. It is a good rock, and just as powerful as everyone says, but I never though that it actually helped people win. Multiplayer corrects for known power, and the person who takes a strong lead first is less likely to win. Strong counter attacks are much more likely to win and accumulated value.
Turn order makes sense in general, but I think that this is potentially variable between meta. cEDH it would make a huge difference, casual but strong EDH it would make a pretty strong difference, but for decks that are not seriously tuned, it will make less impact than the random junk people play. If you are reading this, your deck is probably tuned, so are your opponents' decks, and the more you should account for first turnn advantage in your play. Player 1 is top dog in any game, unless you know that they aren't. Everyone should read Jusstice's primer on table position, consider this data, think about their play style and meta, and then adjust their threat assessment to increase the value of the advantage that they see player 1 having.
The available lands stat is interesting, but I think I want to see some correction for available mana of other players AT THE TIME THAT THE LOSER WAS PUT OUT. It is understandable that the last player standing will have more land available, the winner likely had more turns for land drops, and because other players were put out sooner. What about looking at the ratio of lands that players have at the time a player is put out, and then stats on how much ramp they each play? Every player will make every land drop available, and most will ramp when they can, so how we play is unlikely informed at all by this info, it is more useful as a deck building clue than anything. The ratio at the time someone is put out and the number of lands and ramp in the deck would correct some for the extra turns the winner gets, and could give us actionable ideas, like players of more land ramp win more, players with more lands in their deck win more, etc... The whims of fate that give one player more land and they just win because of it are not actionable. The knowledge of how to build a deck with more ways to get lands, while hopefully not sacrificing other essential deck elements, that is actionable. As it stands, I am not sure we gain any information from this, because the statistic is not complete for what we think it says, and we are left with no idea of what it means for deck building.
The deck type discussion was much less useful than the rest. This is not about archetype, and the only impressive number in there was significantly under represented in the data set, so we can draw no reliable conclusions. I would like to see p values for these stats, even knowing that p value is not the perfect reflection of the reliability of a stat. Without it, we only have representation within the sample size, and a data set with only 4 data points representing Superfriennds is not convincing, unless it had won every game. As it stands, with a 44% win rate, I think that they are probably misunderstanding or misrepresenting the stats. 44% looks like 4 of 9, so there would have to be 9 superfriends decks, not 4, or the stats would have to be 25%, 50%, 75%, or 100%. So a deck type that showed up 9/313 of the time, and only won 4/313 of the time, that is not enough of an impact on the data set to draw any conclusions.
I look forward to further stats and discussions, this is a good direction to go for the game. I just think, like every good statistical analysis, the numbers presented need to be questioned, considered, and study limitations, confounding, and bias brought up for discussion. As it stands, I am unable to draw conclusions that will change my deck building or play style. I reviewed these stats with my EDH playing kids, and it was good for them to hear why I already play and build the way I do, but my play and deck construction style are unlikely to change as a result of these numbers.
What about everyone else? Will these numbers affect your game at all? Will metas change any due to these numbers? Are there hidden gems that we can take advantage of here? I look forward to a good discussion.
I was pleasantly surprised to see them attempt to talk in depth about statistics, possible advantages and so on, because, while i like them, i usually feel they are too much into good stuff for my taste ("hurr, durr, Ancient Tomb and Vedalken Orrery go into every deck, hurr, durr").
As you said, there are only few things to take away from their analysis:
- I wasn't surprised at all, that turn 1 Sol Rings actually reduce ones chance of winning. For one, said player will be overrun in fear of an overwhelming advantage. Plus, the other opponents are more likely to develop critical advantages while everyone is dealing with the fast mana guy.
- Yes, the player that will go first will have a certain advantage and price on their head, until (s)he is overtaken in terms of mana availability and resources to spend them on - most likely cards in hand. This can only be news for anyone who hasn't went any further than EDH 101. Targeting the first player when in doubt is common knowledge in my playgroup and usually the third criteria in line when it comes to deciding who to focus on, with mana available being the first and cards in hand (or available in any other zone) a very close second.
- I can only speak for my meta, but mana/lands available isn't too good of an indicator of who will loose, while being a great one for who will win. Whenever i see someone being mana/color screwed, i'll focus on another opponent, who's more likely to be a challenger for the W. There are plenty of rounds i can roughly recall where the runner up and the first one to loose had a similiar amount of lands, with the big difference, that the player to loose first had them several rounds before the runner up did. Being the bigger threat he was taken down first, as the one behind was an easier player to handle, even if he might've catched up.
- As for the deciding factor of winning, combat damage is without a doubt the definite go to route. Noncombat damage second seems realistic, but after that it's where they statistics don't match our meta. With the likes of Rafiq of the Many, Atarka, World Render, Prossh, Skyraider of Kher and others there are plenty of players getting eliminiated by commander damage. Infect does have it's place as well, with Tainted Strike being a short route for Voltron decks and Triumph of the Hordes the short route for aggro decks. We have a ton more infect Ws than through mill, which usually only happens by "accident", say The Scarab God focusing a player hard, in the hope of finding juicy reanimation targets.
- I am unsure about card type as a criteria, as i can recall different decks that would fulfill several quotas - say 25 creatures + 25 instants/sorceries. Of course instant and sorcery heavy decks have an advantage over permanent heavy decks, because they don't telegraph their current state as much. Yet, a strong permanent heavy deck will have it easier to outvalue an instant and sorcery heavy deck, if the quality of those permanents isn't rubbish. I still think there aren't too many great enchantress commanders out there, yet that cardtype is very likely to change the outcome of games oftenly, as the top notch enchantments are hard to interact with for certain color( combination)s. E.g. while not being an enchantress deck, my Brago, King Eternal is very enchantment heavy and wins it's fair share of games simply by being hard to be disrupted. Superfriends can be a strong archetype, yet i don't find them too strong, if you prepare accordingly. If i can tell a deck is superfriends i will cling onto my Disenchant stuff just until Doubling Season drops, because that's the actual problematic card, planeswalkers are fine to interact with other than that. They mentioned that superfriends outvalues other decks when you leave too many of them sticking around... but hey, that happens with other permanents as well, only difference is, they might not have a toolbox stapled on them as much.
Bottom line: Meta calls are still a thing, the amount of info was too small and sometimes akwardly defined. The discussion is very interesting and it could be worth it already just to pitch the topic to players who aren't as "evolved" as my fellow ones are.
It was an interesting episode, but it was way too small a ample size, especially for some things (like only four superfriends decks).
Also, I feel some things were misinterpreted. For example, if going first gives a huge advantage, shouldn't going second still give an advantage over going third, which in turn gives an advantage over going fourth? After all, the if the player in second place wins, they had one more turn than third and fourth place. They showed second, third, and fourth as being all equal, but that makes no sense.
It was an interesting episode, but it was way too small a ample size, especially for some things (like only four superfriends decks).
Also, I feel some things were misinterpreted. For example, if going first gives a huge advantage, shouldn't going second still give an advantage over going third, which in turn gives an advantage over going fourth? After all, the if the player in second place wins, they had one more turn than third and fourth place. They showed second, third, and fourth as being all equal, but that makes no sense.
Not sure how that an example of misinterpreting the data. If you look at that data, how do you interpret it?
Stats via Command Zone
For me, I only had one surprise. I think Superfriends and planeswalkers in general are really weak. I still think that the stats are wrong on this one due to low sample size, but that was my surprise.
Superfriends can be fairly fragile, but what Josh said was right, once the deck lands a second or third walker, the rest of the table has to focus removal or attacks. And if the table doesn't have the removal, or trample to get through blockers, the deck snowballs hard.
For the discussion, the Sol Ring thing confirmed my suspicions. As they said, EDH isn't about the first person to drop their 4-5 drop. And ramping usually paints a target on your back.
I haven't finished the vid, but I do wonder about how fair their methods are. If they're only using YouTube games, I suspect those are less likely to be boring, one-sided games because those aren't interesting enough to upload. You know, like games where someone successfully snowballs a sol ring into an early win. Which means the only t1 ring/crypt games are the ones where that fails.
To say t1 ring is a liability, but going first is a boon because mana...seems like a contradiction.
I haven't finished the vid, but I do wonder about how fair their methods are. If they're only using YouTube games, I suspect those are less likely to be boring, one-sided games because those aren't interesting enough to upload. You know, like games where someone successfully snowballs a sol ring into an early win. Which means the only t1 ring/crypt games are the ones where that fails.
To say t1 ring is a liability, but going first is a boon because mana...seems like a contradiction.
I don't view it as a contradiction:
T1 Sol Ring puts people really far ahead, so people target them.
The people who have the most consistent resources tend to win.
Also, did you watch the most recent Game Knights episode? That was entirely one sided. I can't remember if the winner had a T1 Sol Ring though.
How often is a sol ring destroyed in the first 3 turns?
How often is a sol ring destroyed in turn 4 or more?
If a sol ring survives for 1 turn, what is it used to cast?
If a sol ring survives for more than 1 turn, what type of spell(s) is it used to cast?
This evidence is not provided in the video.
Since the question posed in the video was not "What is Sol Ring's Impact on the Format," but rather "We Examined 300 Games and These are the Observations we Made" the data and observations they shared were fine.
How often is a sol ring destroyed in the first 3 turns?
How often is a sol ring destroyed in turn 4 or more?
If a sol ring survives for 1 turn, what is it used to cast?
If a sol ring survives for more than 1 turn, what type of spell(s) is it used to cast?
This evidence is not provided in the video.
Since the question posed in the video was not "What is Sol Ring's Impact on the Format," but rather "We Examined 300 Games and These are the Observations we Made" the data and observations they shared were fine.
The title is "Exactly how good is sol ring in the format"? Then the facts provided are only "% winrate w/ not-early ring" and "% winrate w/ early ring". That only provides an incomplete view of the power of sol ring in the format. Followed up by sampled data based on what people think the impact of sol ring is.
I can't be the only one curious what the percentage is for using their rings for. Like for example how often were equipment, commanders, other ramp, etc were cast/activated with the assistance of a ring. As that type of information is so valuable whether its for deckbuilding, monetary purchases, threat assessment, or just ban/unban.
It was an interesting episode, but it was way too small a ample size, especially for some things (like only four superfriends decks).
Also, I feel some things were misinterpreted. For example, if going first gives a huge advantage, shouldn't going second still give an advantage over going third, which in turn gives an advantage over going fourth? After all, the if the player in second place wins, they had one more turn than third and fourth place. They showed second, third, and fourth as being all equal, but that makes no sense.
Not sure how that an example of misinterpreting the data. If you look at that data, how do you interpret it?
Haven't seen the data yet (I thought they weren't posting it until next week). But the perfectly even split from players 2-4 looks like they only paid attention to how often Player One won and then divided the remaining percentage evenly among the other players. If they had tracked how often players 2-4 won, I would expect some variation, particularly in favor of the lower number players.
It was an interesting episode, but it was way too small a ample size, especially for some things (like only four superfriends decks).
Also, I feel some things were misinterpreted. For example, if going first gives a huge advantage, shouldn't going second still give an advantage over going third, which in turn gives an advantage over going fourth? After all, the if the player in second place wins, they had one more turn than third and fourth place. They showed second, third, and fourth as being all equal, but that makes no sense.
Not sure how that an example of misinterpreting the data. If you look at that data, how do you interpret it?
Haven't seen the data yet (I thought they weren't posting it until next week). But the perfectly even split from players 2-4 looks like they only paid attention to how often Player One won and then divided the remaining percentage evenly among the other players. If they had tracked how often players 2-4 won, I would expect some variation, particularly in favor of the lower number players.
I could be wrong but I thought hey did actaully track those wins and the lack of deviation was was the data showed.
The best takeaway from that video is a cautionary tale. It stands as a clear example of how 'scientific' studies can be misrepresented and misused by someone with an agenda and a weak understanding of methodology. In this case the agenda is to create 'surprising' results that fly against the expectations of the typical commander playerbase. The small sample size, blatant ignorance displayed near the start of the video of how statistics work, and lack of a control of any sort should all be red flags.
All they've done is produce clickbait: "Study finds 1st-turn Sol Ring actually horrible play!" or "Study shows Superfriends is best commander archetype!" for example. Now those 'facts' will be repeated ad nauseam in forum discussions about Sol Ring, deck archetypes, winning strategies, and so on. Feeding back into their show when people want to reference the source.
To be fair, they mention repeatedly that this data should not be relied upon. But since the majority of the video is them immediately attempting to justify the results that they get, they end up sending mixed messages. Ultimately the impression that will be left with most viewers is what they spoke most emphatically about; that being "this unintuitive and shocking conclusion is true and here's why!".
Some of the conclusions they draw from it are hilariously nonsensical, even going so far as to suggest that a pet card, Vedalken Orrery, somehow mitigates the disadvantage of not going first. Their sample size for the superfriends archetype winrate is four. That's not statistics. Realistically that's closer to an anecdote about your friend Steve's totally super-rad Atraxa superfriends deck that always wins when he plays it and therefore proves the archetype is top-tier.
... None of that is to say that a study of this sort can't or shouldn't be done. It's simply that I have tremendous misgivings about their methodology. It is difficult and very much expensive to get a large enough data set in a given activity, particularly when there are so many factors involved, when that activity typically takes an hour or more, and when the goal of those involved is ill-defined (since many of these players involved are not necessarily trying to win commander games, but rather create shows that are fun to watch for their youtube viewers).
Now those 'facts' will be repeated ad nauseam in forum discussions about Sol Ring, deck archetypes, winning strategies, and so on.
This is my concern. I'm not looking forward to seeing any discussion about whether or not Sol Ring and other fast mana cards should be banned because inevitably some rube is going to bring up this video and completely derail the discussion. I suspect this video will do more harm than good in the long run.
Incidentally, this thread on Reddit suggests the exact opposite of their conclusions about both Sol Ring and going first.
Okay - the data's out, and I looked at it to answer my own question regarding turn order and win rates.
1st Player: 95 wins or 30.25%
2nd Player: 72 wins or 22.93%
3rd Player: 74 wins or 23.57%
4th Player: 73 wins or 23.25%
I went back and looked at the graphic they had used to present this info, and the bar graph actually does show a matching variation, which I hadn't noticed before because I was focused on the numbers, which they listed as 30%, 23%, 23%, and 23%. I should have seen that it didn't add up to 100% (they ignored significant digits and rounded 0.57% down, even while rounding 0.93% up). I wish they had included the actual win numbers on the graph, in addition to the percentage.
I still have to ask why. If Player 1 enjoys such a huge benefit from having one more turn than his opponents, why does player 2 have the lowest win rate, despite having more turns than Players 3 and 4? (assuming most players tend to win on their own turn, and ignoring extra turn effects).
Interestingly, I looked at which player killed the most opponents, and it was a little closer to what I expected.
Player 1: 288 knockouts
Player 2: 228 knockouts
Player 3: 216 knockouts
Player 4: 218 knockouts
Is does show Player 1 killing the most, followed by Player 2. Players 3 and 4 again defy this trend.
These could be off by a little, in part because multiple games list everyone as dead or list two players as having won.
It's cool to see, but the sample size is really small. I'd love to see this done again with a much, much larger sampling, but I appreciate how difficult that would be to do.
Correlating cost with win% is immediately a problem. I have by far the largest, most expensive collection at my LGS among the frequent players, but that's because I've also been playing longer, and certainly more intensely, than anyone else there. Having better cards helps, sure, but also just being more experienced at the game and better at playing it, frankly. Both of those things are caused by me having played a lot, so it's more correlation than causation. Also obviously having a high-paying job helps.
The other issue, not to keep harping on this, is the data. I don't watch EDH games on youtube, but my intuition is that they probably don't have one guy with a 10K deck and one guy with draft chaff because, unless someone is playing Phelddagrif #ReadMyPrimer, the game is probably going to be very uninteresting to watch. The tiny gap in price, as well as the low price overall (sub $600? Yeah that's not going to pay for very many ABU duals) leads me to believe that most of the games they're watching are of moderate budget decks of similar prices fighting each other, which also makes a lot of sense as youtube content. But it's hardly representative of the actual question people want to know, which is "how much more likely is the 10K guy to win than the draft chaff guy?" And the answer to that question is most likely "a lot, both because pickup games aren't always reasonably balanced like youtube games, and because the 10K guy has probably been playing a lot longer if he's willing to invest that much money into his deck, so he's also probably a better player."
I'm seriously confused by their "winningest cards" section. Sol ring is #1. How is that even possible? Nearly every deck plays sol ring, so even if none of the decks that DIDN'T play sol ring EVER won, it still wouldn't be that "winning" - especially if, as seems the case, most of the games where people weren't playing sol ring were because it was banned, which makes it impossible for it to win.
Obviously the stat you want to know is how many times decks containing X card won, divided by the number of games featuring decks containing card X. So if card X only appeared in one game, and it won, then it would look amazing, which is obviously a problem so you'd probably need a lower limit on the number of cases to qualify, but that's clearly not what's happening here. It kinda seems like they're just taking how many times decks containing X card won...and then totally ignoring how often that card appeared in the sample. Which is just grade-school level stats, I mean seriously. Wtf. I hope I'm wrong about this because that would be absolutely pathetic.
Correlating cost with win% is immediately a problem. I have by far the largest, most expensive collection at my LGS among the frequent players, but that's because I've also been playing longer, and certainly more intensely, than anyone else there. Having better cards helps, sure, but also just being more experienced at the game and better at playing it, frankly. Both of those things are caused by me having played a lot, so it's more correlation than causation.
Excellent point. I've been playing MTG for almost 16 years straight, and I've played EDH/Commander for almost 9 years. Having played that long, I like to think that I am better than a noob, regardless of deck value. But that experience has also affected the value of my decks. I tend to avoid paying lots of money for cards, yet I own cards that are easily worth 5-10x what I paid for them years ago. Cards fluctuate in value, and over the years I've been fortunate to take advantage of that, both by acquiring when they are low and by trading away when they spike. I'd probably balk at buying nearly all of my decks for face value (not that mine are even high compared to a lot of other players), and I don't expect many noobs to buy-in at that level immediately. So, yeah - $$$-to-win% is a very slippery slope.
To address the OP, RE: what changes will you make:
Even besides the issues I have with their sources of data and the conclusions they're drawing, I don't think I'll be making any changes. Most people, to a certain extent, are trying to make decks that will be fun, and winning is only part of that. And the players that are only looking to win won't care about most of these results, even if they were justified - the specifics are way more important than the generalizations. How well a Circu deck does doesn't make Zur worse, even if they're sharing the same colors. And most of the results aren't in a cEDH meta anyway, rendering them about as relevant as draft results to a standard format.
to go through the results one by one:
going first: board state is board state. Threat assess correctly, which means early the first player is probably most threatening, but that can change. There's nothing inherent about going first that makes it better, except that you're some fraction of a card and a land drop ahead of the table. If you miss a land drop and discard a card, then you may as well have gone last. So just threat assess correctly. Duh.
Sol ring/mana crypt: imo the results are too tainted to justify any change in terms of playing sol ring, although sandbagging if you have no follow-up play is probably fine advice. Otherwise, good threat assessment is good. duh.
lamds: unclear how they're comparing this, they totally handwave players being eliminated before the game ends, which is super weak. Even then, it's not much new info. Obviously being ahead in some metric means you're more likely to be ahead in general. Ramp is generally good, this is known. Nothing new here, really, even if the data was trustable.
types of decks: too small of sample size to be useful imo. Even if it was, see original argument. Specifics > trends, when you're talking about commander. tibor and lumia sucking doesn't make jeleva suck, and vice versa.
how are players eliminated: this one is actually sort of interesting, although it's not clear to me how they're determining some of these categories. is mill combo mill? What's infinite turns, is that combo or is it damage since the damage isn't really infinite? Either way, though, your meta is probably going to have totally different stats, and in any case there's not a ton of "to-dos" based on this data imo.
budget: see my earlier post. The question is problematic on a number of a levels - budget can be correlated without being the cause of the wins, the data is going to be skewed because of the source, the budgets don't seem to be particularly high in any case...none of this is good data analysis imo. And the end result is too vague to really be actionable imo.
Colors: all the color info is pretty useless for the reasons described in my initial argument, but also because of the really pathetically low sample sizes for specific multicolor combos. Specifics > trends. Data could be interesting at least as cocktail facts if it had sufficient sample size and better sources, but as-is I think it's useless.
cards: wuuuut is this. See previous post, but either they're doing some really arcane analysis or they're doing some really stupid analysis. I'd expect to see the "best card" as something like food chain because, if it shows up, it's probably OP as hell. Sol ring makes zero sense because nearly every deck is playing it, therefore its mere presence in the deck cannot be what makes a deck win, since it's even across the board. They seem to have correctly properly for this with color, and then didn't correct at all for this for cards - probably because otherwise the result would be something dumb and boring like food chain, which practically no one uses unless they're a butt. In practice, though, unless you actually recorded which games cards were drawn or played in, instead of just existed in the deck - the numbers are most likely going to either be some specific combo piece, or they're going to be a card that's indicative of the deck as a whole, and it's not the specific card itself that made a major difference. Case in point, enlightened tutor - if you're playing white and aren't running it, then there's a good chance you're running "garbage white". But merely adding the tutor wouldn't make your deck good. The problem is your whole deck. The existence or lack thereof of enlightened tutor is just a symptom.
tl;dr - cute idea, ruined by the execution and ultimately pretty useless either way.
In general I do agree having access to resources first tend to give you an advantage.
But without a doubt most wins come from players who utilize the most resources, aka maximizing your turns/mana/spells compared to your peers. This is also why white tends to suffer the most.
"Oh the person who plays Sol Ring gets targeted" is some real absurd thinking, it is an argument people will make..but when does it ever actually happen? Just tonight I saw a Muldrotha deck, start with "Fetch, Sol Ring.." and Player 2 cast Natralize on Player 3's Hardened Scales instead of the Sol Ring which lead us to the fun of "Strip Mine, Muldrotha with Swiftfoot Boots."
That's the issue with Sol Ring. Even though it's always a good play to remove it, unless you do it in the first couple of turns there will probably be a better target.
I tend to think it's rarely worth trading 1:1 with ramp, tutors, or draw, at least in 4+ player games. I'd rather wait for the actual threat and answer that. If you're too proactive with your answers you miss the opportunity for that player to have weak follow-up and be a non-issue, for someone else to answer it for you, or for it all to become moot in a board wipe.
Varies by meta ofc. If you know it's a cEDH deck at your casual table, then fire and brimstone all the way.
I tend to think it's rarely worth trading 1:1 with ramp, tutors, or draw, at least in 4+ player games. I'd rather wait for the actual threat and answer that. If you're too proactive with your answers you miss the opportunity for that player to have weak follow-up and be a non-issue, for someone else to answer it for you, or for it all to become moot in a board wipe.
Varies by meta ofc. If you know it's a cEDH deck at your casual table, then fire and brimstone all the way.
This needs to be brought up more often. Player skill in threat-assessment, play sequencing, etc. is not talk about enough. Whereas it's the first thing that should be brought up.
It's just difficult to quantify skills in play and deck building. Scapegoating a particular card, budget, or gathering "stats" on colors/turn order is just too simple. I've had opponents be mad at me for having a foiled deck at the end of games.
Private Mod Note
():
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
What do you guys all think about the episode, and the information presented? Did it change your mind about anything? Were there any lightbulbs that went off for you?
For me, I only had one surprise. I think Superfriends and planeswalkers in general are really weak. I still think that the stats are wrong on this one due to low sample size, but that was my surprise.
For the Sol Ring, I don't always put it in my decks. I know. Sacrilege. It is a good rock, and just as powerful as everyone says, but I never though that it actually helped people win. Multiplayer corrects for known power, and the person who takes a strong lead first is less likely to win. Strong counter attacks are much more likely to win and accumulated value.
Turn order makes sense in general, but I think that this is potentially variable between meta. cEDH it would make a huge difference, casual but strong EDH it would make a pretty strong difference, but for decks that are not seriously tuned, it will make less impact than the random junk people play. If you are reading this, your deck is probably tuned, so are your opponents' decks, and the more you should account for first turnn advantage in your play. Player 1 is top dog in any game, unless you know that they aren't. Everyone should read Jusstice's primer on table position, consider this data, think about their play style and meta, and then adjust their threat assessment to increase the value of the advantage that they see player 1 having.
The available lands stat is interesting, but I think I want to see some correction for available mana of other players AT THE TIME THAT THE LOSER WAS PUT OUT. It is understandable that the last player standing will have more land available, the winner likely had more turns for land drops, and because other players were put out sooner. What about looking at the ratio of lands that players have at the time a player is put out, and then stats on how much ramp they each play? Every player will make every land drop available, and most will ramp when they can, so how we play is unlikely informed at all by this info, it is more useful as a deck building clue than anything. The ratio at the time someone is put out and the number of lands and ramp in the deck would correct some for the extra turns the winner gets, and could give us actionable ideas, like players of more land ramp win more, players with more lands in their deck win more, etc... The whims of fate that give one player more land and they just win because of it are not actionable. The knowledge of how to build a deck with more ways to get lands, while hopefully not sacrificing other essential deck elements, that is actionable. As it stands, I am not sure we gain any information from this, because the statistic is not complete for what we think it says, and we are left with no idea of what it means for deck building.
The deck type discussion was much less useful than the rest. This is not about archetype, and the only impressive number in there was significantly under represented in the data set, so we can draw no reliable conclusions. I would like to see p values for these stats, even knowing that p value is not the perfect reflection of the reliability of a stat. Without it, we only have representation within the sample size, and a data set with only 4 data points representing Superfriennds is not convincing, unless it had won every game. As it stands, with a 44% win rate, I think that they are probably misunderstanding or misrepresenting the stats. 44% looks like 4 of 9, so there would have to be 9 superfriends decks, not 4, or the stats would have to be 25%, 50%, 75%, or 100%. So a deck type that showed up 9/313 of the time, and only won 4/313 of the time, that is not enough of an impact on the data set to draw any conclusions.
I look forward to further stats and discussions, this is a good direction to go for the game. I just think, like every good statistical analysis, the numbers presented need to be questioned, considered, and study limitations, confounding, and bias brought up for discussion. As it stands, I am unable to draw conclusions that will change my deck building or play style. I reviewed these stats with my EDH playing kids, and it was good for them to hear why I already play and build the way I do, but my play and deck construction style are unlikely to change as a result of these numbers.
What about everyone else? Will these numbers affect your game at all? Will metas change any due to these numbers? Are there hidden gems that we can take advantage of here? I look forward to a good discussion.
WUBSente: The Politics and Metaphor of Stones
My Vampire Hunter Kit Innistrad Themed Cube!
As you said, there are only few things to take away from their analysis:
- I wasn't surprised at all, that turn 1 Sol Rings actually reduce ones chance of winning. For one, said player will be overrun in fear of an overwhelming advantage. Plus, the other opponents are more likely to develop critical advantages while everyone is dealing with the fast mana guy.
- Yes, the player that will go first will have a certain advantage and price on their head, until (s)he is overtaken in terms of mana availability and resources to spend them on - most likely cards in hand. This can only be news for anyone who hasn't went any further than EDH 101. Targeting the first player when in doubt is common knowledge in my playgroup and usually the third criteria in line when it comes to deciding who to focus on, with mana available being the first and cards in hand (or available in any other zone) a very close second.
- I can only speak for my meta, but mana/lands available isn't too good of an indicator of who will loose, while being a great one for who will win. Whenever i see someone being mana/color screwed, i'll focus on another opponent, who's more likely to be a challenger for the W. There are plenty of rounds i can roughly recall where the runner up and the first one to loose had a similiar amount of lands, with the big difference, that the player to loose first had them several rounds before the runner up did. Being the bigger threat he was taken down first, as the one behind was an easier player to handle, even if he might've catched up.
- As for the deciding factor of winning, combat damage is without a doubt the definite go to route. Noncombat damage second seems realistic, but after that it's where they statistics don't match our meta. With the likes of Rafiq of the Many, Atarka, World Render, Prossh, Skyraider of Kher and others there are plenty of players getting eliminiated by commander damage. Infect does have it's place as well, with Tainted Strike being a short route for Voltron decks and Triumph of the Hordes the short route for aggro decks. We have a ton more infect Ws than through mill, which usually only happens by "accident", say The Scarab God focusing a player hard, in the hope of finding juicy reanimation targets.
- I am unsure about card type as a criteria, as i can recall different decks that would fulfill several quotas - say 25 creatures + 25 instants/sorceries. Of course instant and sorcery heavy decks have an advantage over permanent heavy decks, because they don't telegraph their current state as much. Yet, a strong permanent heavy deck will have it easier to outvalue an instant and sorcery heavy deck, if the quality of those permanents isn't rubbish. I still think there aren't too many great enchantress commanders out there, yet that cardtype is very likely to change the outcome of games oftenly, as the top notch enchantments are hard to interact with for certain color( combination)s. E.g. while not being an enchantress deck, my Brago, King Eternal is very enchantment heavy and wins it's fair share of games simply by being hard to be disrupted. Superfriends can be a strong archetype, yet i don't find them too strong, if you prepare accordingly. If i can tell a deck is superfriends i will cling onto my Disenchant stuff just until Doubling Season drops, because that's the actual problematic card, planeswalkers are fine to interact with other than that. They mentioned that superfriends outvalues other decks when you leave too many of them sticking around... but hey, that happens with other permanents as well, only difference is, they might not have a toolbox stapled on them as much.
Bottom line: Meta calls are still a thing, the amount of info was too small and sometimes akwardly defined. The discussion is very interesting and it could be worth it already just to pitch the topic to players who aren't as "evolved" as my fellow ones are. Do you happen to have a link for that? I am too clumsy to find it.
Also, I feel some things were misinterpreted. For example, if going first gives a huge advantage, shouldn't going second still give an advantage over going third, which in turn gives an advantage over going fourth? After all, the if the player in second place wins, they had one more turn than third and fourth place. They showed second, third, and fourth as being all equal, but that makes no sense.
2023 Average Peasant Cube|and Discussion
Because I have more decks than fit in a signature
Useful Resources:
MTGSalvation tags
EDHREC
ManabaseCrafter
BChainer, Dementia Master(Big Mana/Reanimator)
BRRakdos, The Showstopper (Mass Life Loss/Ramp)
BUThe Scarab God (Zombie Tribal/Control)
BWKarlov of the Ghost Council (Life Gain)
BGJarad, Golgari Lich Lord (Stompy/Dredge)
BRGProssh, Skyraider of Kher (Tokens/Non-infinite Combo)
Superfriends can be fairly fragile, but what Josh said was right, once the deck lands a second or third walker, the rest of the table has to focus removal or attacks. And if the table doesn't have the removal, or trample to get through blockers, the deck snowballs hard.
For the discussion, the Sol Ring thing confirmed my suspicions. As they said, EDH isn't about the first person to drop their 4-5 drop. And ramping usually paints a target on your back.
To say t1 ring is a liability, but going first is a boon because mana...seems like a contradiction.
EDH Primers
Phelddagrif - Zirilan
EDH
Thrasios+Bruse - Pang - Sasaya - Wydwen - Feather - Rona - Toshiro - Sylvia+Khorvath - Geth - QMarchesa - Firesong - Athreos - Arixmethes - Isperia - Etali - Silas+Sidar - Saskia - Virtus+Gorm - Kynaios - Naban - Aryel - Mizzix - Kazuul - Tymna+Kraum - Sidar+Tymna - Ayli - Gwendlyn - Phelddagrif 4 - Liliana - Kaervek - Phelddagrif 3 - Mairsil - Scarab - Child - Phenax - Shirei - Thada - Depala - Circu - Kytheon - GrenzoHR - Phelddagrif - Reyhan+Kraum - Toshiro - Varolz - Nin - Ojutai - Tasigur - Zedruu - Uril - Edric - Wort - Zurgo - Nahiri - Grenzo - Kozilek - Yisan - Ink-Treader - Yisan - Brago - Sidisi - Toshiro - Alexi - Sygg - Brimaz - Sek'Kuar - Marchesa - Vish Kal - Iroas - Phelddagrif - Ephara - Derevi - Glissa - Wanderer - Saffi - Melek - Xiahou Dun - Lazav - Lin Sivvi - Zirilan - Glissa
PDH - Drake - Graverobber - Izzet GM - Tallowisp - Symbiote Brawl - Feather - Ugin - Jace - Scarab - Angrath - Vraska - Kumena Oathbreaker - Wrenn&6
Elaboration
This evidence is not provided in the video.
I don't view it as a contradiction:
T1 Sol Ring puts people really far ahead, so people target them.
The people who have the most consistent resources tend to win.
Also, did you watch the most recent Game Knights episode? That was entirely one sided. I can't remember if the winner had a T1 Sol Ring though.
Since the question posed in the video was not "What is Sol Ring's Impact on the Format," but rather "We Examined 300 Games and These are the Observations we Made" the data and observations they shared were fine.
I can't be the only one curious what the percentage is for using their rings for. Like for example how often were equipment, commanders, other ramp, etc were cast/activated with the assistance of a ring. As that type of information is so valuable whether its for deckbuilding, monetary purchases, threat assessment, or just ban/unban.
2023 Average Peasant Cube|and Discussion
Because I have more decks than fit in a signature
Useful Resources:
MTGSalvation tags
EDHREC
ManabaseCrafter
EDH Primers
Phelddagrif - Zirilan
EDH
Thrasios+Bruse - Pang - Sasaya - Wydwen - Feather - Rona - Toshiro - Sylvia+Khorvath - Geth - QMarchesa - Firesong - Athreos - Arixmethes - Isperia - Etali - Silas+Sidar - Saskia - Virtus+Gorm - Kynaios - Naban - Aryel - Mizzix - Kazuul - Tymna+Kraum - Sidar+Tymna - Ayli - Gwendlyn - Phelddagrif 4 - Liliana - Kaervek - Phelddagrif 3 - Mairsil - Scarab - Child - Phenax - Shirei - Thada - Depala - Circu - Kytheon - GrenzoHR - Phelddagrif - Reyhan+Kraum - Toshiro - Varolz - Nin - Ojutai - Tasigur - Zedruu - Uril - Edric - Wort - Zurgo - Nahiri - Grenzo - Kozilek - Yisan - Ink-Treader - Yisan - Brago - Sidisi - Toshiro - Alexi - Sygg - Brimaz - Sek'Kuar - Marchesa - Vish Kal - Iroas - Phelddagrif - Ephara - Derevi - Glissa - Wanderer - Saffi - Melek - Xiahou Dun - Lazav - Lin Sivvi - Zirilan - Glissa
PDH - Drake - Graverobber - Izzet GM - Tallowisp - Symbiote Brawl - Feather - Ugin - Jace - Scarab - Angrath - Vraska - Kumena Oathbreaker - Wrenn&6
BChainer, Dementia Master(Big Mana/Reanimator)
BRRakdos, The Showstopper (Mass Life Loss/Ramp)
BUThe Scarab God (Zombie Tribal/Control)
BWKarlov of the Ghost Council (Life Gain)
BGJarad, Golgari Lich Lord (Stompy/Dredge)
BRGProssh, Skyraider of Kher (Tokens/Non-infinite Combo)
All they've done is produce clickbait: "Study finds 1st-turn Sol Ring actually horrible play!" or "Study shows Superfriends is best commander archetype!" for example. Now those 'facts' will be repeated ad nauseam in forum discussions about Sol Ring, deck archetypes, winning strategies, and so on. Feeding back into their show when people want to reference the source.
To be fair, they mention repeatedly that this data should not be relied upon. But since the majority of the video is them immediately attempting to justify the results that they get, they end up sending mixed messages. Ultimately the impression that will be left with most viewers is what they spoke most emphatically about; that being "this unintuitive and shocking conclusion is true and here's why!".
Some of the conclusions they draw from it are hilariously nonsensical, even going so far as to suggest that a pet card, Vedalken Orrery, somehow mitigates the disadvantage of not going first. Their sample size for the superfriends archetype winrate is four. That's not statistics. Realistically that's closer to an anecdote about your friend Steve's totally super-rad Atraxa superfriends deck that always wins when he plays it and therefore proves the archetype is top-tier.
... None of that is to say that a study of this sort can't or shouldn't be done. It's simply that I have tremendous misgivings about their methodology. It is difficult and very much expensive to get a large enough data set in a given activity, particularly when there are so many factors involved, when that activity typically takes an hour or more, and when the goal of those involved is ill-defined (since many of these players involved are not necessarily trying to win commander games, but rather create shows that are fun to watch for their youtube viewers).
- Rabid Wombat
This is my concern. I'm not looking forward to seeing any discussion about whether or not Sol Ring and other fast mana cards should be banned because inevitably some rube is going to bring up this video and completely derail the discussion. I suspect this video will do more harm than good in the long run.
Incidentally, this thread on Reddit suggests the exact opposite of their conclusions about both Sol Ring and going first.
[Primer] Erebos, God of the Dead
HONK HONK
1st Player: 95 wins or 30.25%
2nd Player: 72 wins or 22.93%
3rd Player: 74 wins or 23.57%
4th Player: 73 wins or 23.25%
I went back and looked at the graphic they had used to present this info, and the bar graph actually does show a matching variation, which I hadn't noticed before because I was focused on the numbers, which they listed as 30%, 23%, 23%, and 23%. I should have seen that it didn't add up to 100% (they ignored significant digits and rounded 0.57% down, even while rounding 0.93% up). I wish they had included the actual win numbers on the graph, in addition to the percentage.
I still have to ask why. If Player 1 enjoys such a huge benefit from having one more turn than his opponents, why does player 2 have the lowest win rate, despite having more turns than Players 3 and 4? (assuming most players tend to win on their own turn, and ignoring extra turn effects).
Interestingly, I looked at which player killed the most opponents, and it was a little closer to what I expected.
Player 1: 288 knockouts
Player 2: 228 knockouts
Player 3: 216 knockouts
Player 4: 218 knockouts
Is does show Player 1 killing the most, followed by Player 2. Players 3 and 4 again defy this trend.
These could be off by a little, in part because multiple games list everyone as dead or list two players as having won.
It's cool to see, but the sample size is really small. I'd love to see this done again with a much, much larger sampling, but I appreciate how difficult that would be to do.
2023 Average Peasant Cube|and Discussion
Because I have more decks than fit in a signature
Useful Resources:
MTGSalvation tags
EDHREC
ManabaseCrafter
Correlating cost with win% is immediately a problem. I have by far the largest, most expensive collection at my LGS among the frequent players, but that's because I've also been playing longer, and certainly more intensely, than anyone else there. Having better cards helps, sure, but also just being more experienced at the game and better at playing it, frankly. Both of those things are caused by me having played a lot, so it's more correlation than causation. Also obviously having a high-paying job helps.
The other issue, not to keep harping on this, is the data. I don't watch EDH games on youtube, but my intuition is that they probably don't have one guy with a 10K deck and one guy with draft chaff because, unless someone is playing Phelddagrif #ReadMyPrimer, the game is probably going to be very uninteresting to watch. The tiny gap in price, as well as the low price overall (sub $600? Yeah that's not going to pay for very many ABU duals) leads me to believe that most of the games they're watching are of moderate budget decks of similar prices fighting each other, which also makes a lot of sense as youtube content. But it's hardly representative of the actual question people want to know, which is "how much more likely is the 10K guy to win than the draft chaff guy?" And the answer to that question is most likely "a lot, both because pickup games aren't always reasonably balanced like youtube games, and because the 10K guy has probably been playing a lot longer if he's willing to invest that much money into his deck, so he's also probably a better player."
I'm seriously confused by their "winningest cards" section. Sol ring is #1. How is that even possible? Nearly every deck plays sol ring, so even if none of the decks that DIDN'T play sol ring EVER won, it still wouldn't be that "winning" - especially if, as seems the case, most of the games where people weren't playing sol ring were because it was banned, which makes it impossible for it to win.
Obviously the stat you want to know is how many times decks containing X card won, divided by the number of games featuring decks containing card X. So if card X only appeared in one game, and it won, then it would look amazing, which is obviously a problem so you'd probably need a lower limit on the number of cases to qualify, but that's clearly not what's happening here. It kinda seems like they're just taking how many times decks containing X card won...and then totally ignoring how often that card appeared in the sample. Which is just grade-school level stats, I mean seriously. Wtf. I hope I'm wrong about this because that would be absolutely pathetic.
More opinions as I keep watching.
EDH Primers
Phelddagrif - Zirilan
EDH
Thrasios+Bruse - Pang - Sasaya - Wydwen - Feather - Rona - Toshiro - Sylvia+Khorvath - Geth - QMarchesa - Firesong - Athreos - Arixmethes - Isperia - Etali - Silas+Sidar - Saskia - Virtus+Gorm - Kynaios - Naban - Aryel - Mizzix - Kazuul - Tymna+Kraum - Sidar+Tymna - Ayli - Gwendlyn - Phelddagrif 4 - Liliana - Kaervek - Phelddagrif 3 - Mairsil - Scarab - Child - Phenax - Shirei - Thada - Depala - Circu - Kytheon - GrenzoHR - Phelddagrif - Reyhan+Kraum - Toshiro - Varolz - Nin - Ojutai - Tasigur - Zedruu - Uril - Edric - Wort - Zurgo - Nahiri - Grenzo - Kozilek - Yisan - Ink-Treader - Yisan - Brago - Sidisi - Toshiro - Alexi - Sygg - Brimaz - Sek'Kuar - Marchesa - Vish Kal - Iroas - Phelddagrif - Ephara - Derevi - Glissa - Wanderer - Saffi - Melek - Xiahou Dun - Lazav - Lin Sivvi - Zirilan - Glissa
PDH - Drake - Graverobber - Izzet GM - Tallowisp - Symbiote Brawl - Feather - Ugin - Jace - Scarab - Angrath - Vraska - Kumena Oathbreaker - Wrenn&6
2023 Average Peasant Cube|and Discussion
Because I have more decks than fit in a signature
Useful Resources:
MTGSalvation tags
EDHREC
ManabaseCrafter
Even besides the issues I have with their sources of data and the conclusions they're drawing, I don't think I'll be making any changes. Most people, to a certain extent, are trying to make decks that will be fun, and winning is only part of that. And the players that are only looking to win won't care about most of these results, even if they were justified - the specifics are way more important than the generalizations. How well a Circu deck does doesn't make Zur worse, even if they're sharing the same colors. And most of the results aren't in a cEDH meta anyway, rendering them about as relevant as draft results to a standard format.
to go through the results one by one:
going first: board state is board state. Threat assess correctly, which means early the first player is probably most threatening, but that can change. There's nothing inherent about going first that makes it better, except that you're some fraction of a card and a land drop ahead of the table. If you miss a land drop and discard a card, then you may as well have gone last. So just threat assess correctly. Duh.
Sol ring/mana crypt: imo the results are too tainted to justify any change in terms of playing sol ring, although sandbagging if you have no follow-up play is probably fine advice. Otherwise, good threat assessment is good. duh.
lamds: unclear how they're comparing this, they totally handwave players being eliminated before the game ends, which is super weak. Even then, it's not much new info. Obviously being ahead in some metric means you're more likely to be ahead in general. Ramp is generally good, this is known. Nothing new here, really, even if the data was trustable.
types of decks: too small of sample size to be useful imo. Even if it was, see original argument. Specifics > trends, when you're talking about commander. tibor and lumia sucking doesn't make jeleva suck, and vice versa.
how are players eliminated: this one is actually sort of interesting, although it's not clear to me how they're determining some of these categories. is mill combo mill? What's infinite turns, is that combo or is it damage since the damage isn't really infinite? Either way, though, your meta is probably going to have totally different stats, and in any case there's not a ton of "to-dos" based on this data imo.
budget: see my earlier post. The question is problematic on a number of a levels - budget can be correlated without being the cause of the wins, the data is going to be skewed because of the source, the budgets don't seem to be particularly high in any case...none of this is good data analysis imo. And the end result is too vague to really be actionable imo.
Colors: all the color info is pretty useless for the reasons described in my initial argument, but also because of the really pathetically low sample sizes for specific multicolor combos. Specifics > trends. Data could be interesting at least as cocktail facts if it had sufficient sample size and better sources, but as-is I think it's useless.
cards: wuuuut is this. See previous post, but either they're doing some really arcane analysis or they're doing some really stupid analysis. I'd expect to see the "best card" as something like food chain because, if it shows up, it's probably OP as hell. Sol ring makes zero sense because nearly every deck is playing it, therefore its mere presence in the deck cannot be what makes a deck win, since it's even across the board. They seem to have correctly properly for this with color, and then didn't correct at all for this for cards - probably because otherwise the result would be something dumb and boring like food chain, which practically no one uses unless they're a butt. In practice, though, unless you actually recorded which games cards were drawn or played in, instead of just existed in the deck - the numbers are most likely going to either be some specific combo piece, or they're going to be a card that's indicative of the deck as a whole, and it's not the specific card itself that made a major difference. Case in point, enlightened tutor - if you're playing white and aren't running it, then there's a good chance you're running "garbage white". But merely adding the tutor wouldn't make your deck good. The problem is your whole deck. The existence or lack thereof of enlightened tutor is just a symptom.
tl;dr - cute idea, ruined by the execution and ultimately pretty useless either way.
EDH Primers
Phelddagrif - Zirilan
EDH
Thrasios+Bruse - Pang - Sasaya - Wydwen - Feather - Rona - Toshiro - Sylvia+Khorvath - Geth - QMarchesa - Firesong - Athreos - Arixmethes - Isperia - Etali - Silas+Sidar - Saskia - Virtus+Gorm - Kynaios - Naban - Aryel - Mizzix - Kazuul - Tymna+Kraum - Sidar+Tymna - Ayli - Gwendlyn - Phelddagrif 4 - Liliana - Kaervek - Phelddagrif 3 - Mairsil - Scarab - Child - Phenax - Shirei - Thada - Depala - Circu - Kytheon - GrenzoHR - Phelddagrif - Reyhan+Kraum - Toshiro - Varolz - Nin - Ojutai - Tasigur - Zedruu - Uril - Edric - Wort - Zurgo - Nahiri - Grenzo - Kozilek - Yisan - Ink-Treader - Yisan - Brago - Sidisi - Toshiro - Alexi - Sygg - Brimaz - Sek'Kuar - Marchesa - Vish Kal - Iroas - Phelddagrif - Ephara - Derevi - Glissa - Wanderer - Saffi - Melek - Xiahou Dun - Lazav - Lin Sivvi - Zirilan - Glissa
PDH - Drake - Graverobber - Izzet GM - Tallowisp - Symbiote Brawl - Feather - Ugin - Jace - Scarab - Angrath - Vraska - Kumena Oathbreaker - Wrenn&6
In general I do agree having access to resources first tend to give you an advantage.
But without a doubt most wins come from players who utilize the most resources, aka maximizing your turns/mana/spells compared to your peers. This is also why white tends to suffer the most.
UR Melek, Izzet ParagonUR, B Shirei, Shizo's CaretakerB, R Jaya Ballard, Task MageR,RW Tajic, Blade of the LegionRW, UB Lazav, Dimir MastermindUB, UB Circu, Dimir LobotomistUB, RWU Zedruu the GreatheartedRWU, GUBThe MimeoplasmGUB, UGExperiment Kraj UG, WDarien, King of KjeldorW, BMarrow-GnawerB, WBGKarador, Ghost ChieftainWBG, UTeferi, Temporal ArchmageU, GWUDerevi, Empyrial TacticianGWU, RDaretti, Scrap SavantR, UTalrand, Sky SummonerU, GEzuri, Renegade LeaderG, WUBRGReaper KingWUBRG, RGXenagos, God of RevelsRG, CKozilek, Butcher of TruthC, WUBRGGeneral TazriWUBRG, GTitania, Protector of ArgothG
"Oh the person who plays Sol Ring gets targeted" is some real absurd thinking, it is an argument people will make..but when does it ever actually happen? Just tonight I saw a Muldrotha deck, start with "Fetch, Sol Ring.." and Player 2 cast Natralize on Player 3's Hardened Scales instead of the Sol Ring which lead us to the fun of "Strip Mine, Muldrotha with Swiftfoot Boots."
Dragons of Legend, Lead by Scion of the UR-Dragon
The Gitrog Monster
Gonti, Lord of Luxury
Shogun Saskia
Hive World
Atraxa hates fun
Abzan
Misc. EDH Stuff: Commander Cube | Zombies (Horde)
Resources:Commander Rulings FAQ | Commander Deckbuilding Guide
Follow me on Twitter! @cryogen_mtg
Varies by meta ofc. If you know it's a cEDH deck at your casual table, then fire and brimstone all the way.
EDH Primers
Phelddagrif - Zirilan
EDH
Thrasios+Bruse - Pang - Sasaya - Wydwen - Feather - Rona - Toshiro - Sylvia+Khorvath - Geth - QMarchesa - Firesong - Athreos - Arixmethes - Isperia - Etali - Silas+Sidar - Saskia - Virtus+Gorm - Kynaios - Naban - Aryel - Mizzix - Kazuul - Tymna+Kraum - Sidar+Tymna - Ayli - Gwendlyn - Phelddagrif 4 - Liliana - Kaervek - Phelddagrif 3 - Mairsil - Scarab - Child - Phenax - Shirei - Thada - Depala - Circu - Kytheon - GrenzoHR - Phelddagrif - Reyhan+Kraum - Toshiro - Varolz - Nin - Ojutai - Tasigur - Zedruu - Uril - Edric - Wort - Zurgo - Nahiri - Grenzo - Kozilek - Yisan - Ink-Treader - Yisan - Brago - Sidisi - Toshiro - Alexi - Sygg - Brimaz - Sek'Kuar - Marchesa - Vish Kal - Iroas - Phelddagrif - Ephara - Derevi - Glissa - Wanderer - Saffi - Melek - Xiahou Dun - Lazav - Lin Sivvi - Zirilan - Glissa
PDH - Drake - Graverobber - Izzet GM - Tallowisp - Symbiote Brawl - Feather - Ugin - Jace - Scarab - Angrath - Vraska - Kumena Oathbreaker - Wrenn&6
This needs to be brought up more often. Player skill in threat-assessment, play sequencing, etc. is not talk about enough. Whereas it's the first thing that should be brought up.
It's just difficult to quantify skills in play and deck building. Scapegoating a particular card, budget, or gathering "stats" on colors/turn order is just too simple. I've had opponents be mad at me for having a foiled deck at the end of games.