Magic Memories: Lion's Eye Diamond

[Oversoul]

ProsBloom
After initially deciding to start a Memories thread for LED, I was brainstorming and thought, "Well, it was in ProsBloom at one point." No, it wasn't. Not as far as I can tell, anyway. Not in tournaments. I was mistaken because I used it in ProsBloom in casual play, many years after ProsBloom was but a distant memory in tournament Magic.

I forget how much I've emphasized it, but I've never really been good at the whole deck-brewing thing. It's not that I've never done it or can't do it, but I'm certainly not prolific about it. I think my decks when I play Limited formats are probably fine, although I'm not an experienced drafter. And I don't get to play as much as I used to in the past, so I might be a bit rusty. But even at my hypothetical peak, whenever it was, making a new deck from scratch was just not my strong suit. This was where Al0ysiusHWWW and I made such a good team: he has a good sense for deckbuilding and once I have a deck in my hands to playtest, I can be very thorough about deck refinement, exhaustively examining lines of play, matchups, and fine-tuning both the decklist and my playstyle to optimize winning. Nick's true passion was for control decks, prison decks really, but he liked variety. He was into combo decks. The problem was I already had a leg up on that. I owned more of the really good combo cards than he did. Later, he'd catch up. But in the years while I was blowing games out with Tolarian Academy, High Tide, Pandemonium, Sliver Queen, and so on, he methodically sought out combo decks for which I didn't own the cards. Sometimes, he found something good, like when he mastered Krark-Clan Ironworks. Other times, he hit road blocks, like when he put in work on a Recurring Nightmare + Palinchron infinite combo deck only for me to inform him that the "free" spells from Urza's Block had been hit by power-level errata specifically because of the Recurring Nightmare + Great Whale combo that had overtaken Standard before he'd started playing Magic. And still other times, he focused on things that were just plain silly, like Lure + Basalt Golem. At some point, he noticed that he owned more Mirage Block cards than I did and when the topic of ProsBloom came up, we decided to build it for him. Lacking his deckbuilding imagination, I pretty much went with stock lists from the deck's tournament history. But this wasn't a Standard deck: it was a casual deck! Nick kept the important, core cards of the deck, but added things like Lotus Petal, Meditate, Stormseeker, and oh yeah I already talked about this. Well, he also threw in Lion's Eye Diamond, at a time when the card wasn't really recognized as good. We both played his Storm Seeker ProsBloom deck a lot and it was one of the most fun combo decks I've ever played. LED was not vital for this deck, but we found it practical in a lot of cases, especially when we could tutor up and draw into blue card-drawing spells. Along with Lotus Petal, it gave a source of blue mana to keep Prosperity chains going while Cadaverous Bloom gave us all the black and green mana we needed.

Perhaps more than anything else, this deck is what motivated me to blather on for so long about a simple mana-producing artifact. Our LED-using ProsBloom was built before Scourge enabled the card to take the world by Storm, but after its odd appearance in Broken Jar decks, so perhaps it wasn't coming completely out of nowhere. We didn't even rely on LED. But we found it to be practical. And it was considered a bad card. It was in the same set as other ProsBloom staples, such as Cadaverous Bloom and Infernal Contract, but the pros didn't elect to use it. A few years later, we did, and we liked it. Now, our deck had some advantages because we weren't constrained by the Standard card pool. This is an important distinction. Still, I can't help but wonder! Maybe the Standard card pool that gave rise to the original ProsBloom couldn't really exploit Lion's Eye Diamond. The pros certainly never saw it, and I'm sure they spent more time on this stuff than I ever did. I can't rule out the possibility that I'm just engaging in wishful thinking here, that the card didn't yet have a viable niche. Taking advantage of LED in an environment without Yawgmoth's Will, without Demonic Tutor, without Wheel of Fortune, without Memory Jar, and so on, the "discard your hand" clause really starts looking like a crippling drawback in a small cardpool. There are still options, things like Three Wishes, Vampiric Tutor, and Infernal Contract. But this takes some doing. It's not an automatic, obviously broken thing. I admit that this is starting to look rough.

But here's the thing: the old ProsBloom was already pretty bad. Somewhere, I remember seeing a play-by-play of one of the matches with Mike Long piloting the deck, and he did nothing to affect his opponent for the first seven turns. Seven. That's how slow the environment was. Don't get me wrong: Mirage Block did contain some very cool cards. Lion's Eye Diamond was one of them. But the environment it created was agonizingly slow. Combo has come a long way since then. Magic as a whole has come a long way. You can't expect to sit around doing nothing and go for a turn eight kill. It doesn't work. I don't think that I can prove that LED would have been an improvement, but I strongly suspect that it would have been.

 

[Psarketos]

Very interesting to get a context for the pace of Magic back then, which is also about the time I came into the game. I have a challenge for you as a deck refiner: optimize the Chronarch deck that may be getting posted as part of an article responding to your theory feedback into a Tier 1 deck without sacrificing its main function

 

[Oversoul]

Some of the most powerful tools to use with Lion's Eye Diamond were well-known before the card existed
I mentioned this before. The most important examples are ones that would go on to be used with LED once people actually bothered with the card: Wheel of Fortune, Timetwister, and Demonic Tutor. To a lesser extent, this applies to Braingeyser and Balance. There were other cards that fell out of favor by the time that deck construction caught up to LED, but they might have been used in interactions with the card if anyone had thought to do so at the time. These include Recall, Regrowth, Diminishing Returns, Demonic Consultation, and Animate Dead.

In particular, it's notable that the powerful Infernal Tutor + Lion's Eye Diamond synergy lets Infernal Tutor act like Demonic Tutor, but the original is still strictly better and could just as easily be used alongside LED.

It's just as good as Infernal Tutor when used with LED, and can also be used in the same way while holding onto your hand, for greater flexibility. And yet, people weren't doing it. This reminds me of something I once read...

So I have no idea if this is true, if anyone actually ever did this, but it makes for a great story anyway. I hope it's true but I've seen no documentation of it. The story goes, there was this old psychology experiment conducted on children. You put the subject in a room with an open door leading into a second room, which itself has another open door at the other end. The subject is instructed to cross the room, to make it from the one doorway to the other, without any part of his or her body touching the floor of the room in-between the two doorways. Oh, and also it is made clear that the task is being timed, that faster is better. In the first room, near the starting position, you leave the following items: two wooden boards and two lengths of rope. Supposedly, the result is that pretty much all of the children successfully complete the task, and in pretty much the same way: by wrapping one length of rope under each board, grasping the ends of the ropes in the hands, placing one foot on each board, and dragging the boards along the floor, sort of like a skiing motion. The especially bright ones figure out the method sooner than the others and the particularly athletic ones complete the crossing faster ones they've set the boards up, so if you got enough data you'd get a bell curve of completion times, but overall the kids complete the task successfully, figure it out on their own, and all use the same method.

Now, we run a slightly different version of the same experiment: children of the same age, with the same room and the same instructions, but this time they're only given one board and one length of rope. The "skiing" method isn't available, so instead they use a different method, and again pretty much all of the kids arrive at the same one: wrapping the length of rope around the front portion of the board, grasping the ends of the rope in the hands, mounting both feet toward the back of the board, pulling up on the rope to slightly lift the front of the board and executing a series of small heel-toe hops, propelling themselves across the room as the board slides forward with each little hop. As in the first version of the experiment, you get a bell curve: some of the children are quicker to arrive at the solution and some of them are more adept at the physical execution of the task. But here's the thing: this method is faster than the other one, and not just a little bit. It's so much faster that the slower children using the one-board method are faster than the faster children using the two-board method. In other words, the second board (and rope too) is superfluous. If the children knew of both methods in advance and had seen them in action, they'd ignore the extra materials and use the one-board method. But left to their own devices, they devise the inferior method when given the opportunity to do so.

I am a bit of a science nerd and feel that it is important to stress that I have seen no evidence that this experiment was actually done, no published paper or first-hand accounts. I read about it in an old book, I forget where, but the context was someone recounting this to make a point (like I'm doing now), so it could quite possibly be made-up. I tell the story anyway because the result just sounds so utterly believable and it's such a strong allegory for the way the human brain works. And I suspect that this sort of lesson is relevant to Magic deck construction. After Lion's Eye Diamond came out, anyone could have used Demonic Tutor and sacrificed multiple Diamonds for mana, then used that mana to pay for the card that was searched for by Demonic Tutor. The exact potency of such an interaction in 1996 relative to what it might be used for in later years isn't immediately clear, but there were some strong possibilities. But, to my knowledge, it simply wasn't done. A decade later, Infernal Tutor was released and players noted, "Hey, I can use LED to turn on Hellbent for this and use the mana to pay for the card that I tutor up." And from there, it became obvious that Demonic Tutor and Lion's Eye Diamond work well together. Technically, Demonic Tutor is better suited to the task (like I said, it's more flexible).

 

[Psarketos]

I think we have to be careful when analyzing a method for 'inferior,' particularly because there are often multiple facets that a person might reasonably use to make a judgment. As one example regarding the ski boards, perhaps that method was more comfortable to propel, or even just more fun. Despite being given notice that time elapsed was a factor in other peoples judgment, a person might still reasonably make a tradeoff between what they think is important (like comfort and fun) and some variable like time elapsed that they are considering for someone else, or for some other reason. The "fun" to "win more" ratio particularly strikes me as relevant to both Magic generally and CPA more specifically In terms of historic context, when you have little pressure to affect your opponent before turn 7 in some games, being more mathematically efficient might have had a lot less utility than something like, "This combo piece is both a cool card and always starts a fun conversation with my opponents!"

 

[Oversoul]

I think we have to be careful when analyzing a method for 'inferior,' particularly because there are often multiple facets that a person might reasonably use to make a judgment. As one example regarding the ski boards, perhaps that method was more comfortable to propel, or even just more fun. Despite being given notice that time elapsed was a factor in other peoples judgment, a person might still reasonably make a tradeoff between what they think is important (like comfort and fun) and some variable like time elapsed that they are considering for someone else, or for some other reason.

While I do like the story, I suspect that it is apocryphal. But even if it isn't, my reason for bringing it up was to illustrate the concept that giving humans more to work with sometimes blinds them to options that they would notice if they were actually given less, options which they might be more inclined to use once cognizant of them. The idea in the story isn't that the children choose the two-board method because it is more comfortable or more fun. The idea in the story is that they choose the two-board method because they are completely and utterly oblivious to the existence of a one-board option. They fail to notice it. They are given a task and their brains set to work looking for solutions to the puzzle, and the set of circumstances, two boards, two ropes, two feet that need to be kept off the floor, lead them to the intuitive answer of rigging up this ski-like method. But when that intuitive answer is taken away, they have to look for something else, so they come up with one-board hopping, which is faster. This is, I suspect, analogous to a lot of other things in life: our brains have certain natural puzzle-solving pathways and sometimes the only way we initially find a different answer is when something is taken away. You know, "necessity is the mother of invention" and all that. So that was the point of the story. You could come up with a different version: the children do notice the one-board method and, when given a second board, use the two-board method anyway because they want to buck the system, because they don't care that they were told it was timed and they just find the two-board method more comfortable or fun. But that wasn't the scenario I was describing! For my purposes, that the children using the two-board method don't even notice the one-board method, specifically because they have two boards, is not just a feature of the story but the entire point of it.

It's not a direct analogy, but I think that the same general principle applies to Demonic Tutor and Infernal Tutor. Before Infernal Tutor existed, there was nothing outright stopping anyone from casting Demonic Tutor, activating one or more copies of Lion's Eye Diamond in response, grabbing some big bomb of a card with the Tutor, and then using the mana to pay or to help pay for that big bomb of a card. And yet, it generally wasn't done. Infernal Tutor shows up and suddenly people have a tool that can be used in this way and cannot be used in the more traditional "Demonic Tutor and keep the rest of my hand too" manner. So they adapt and discover that Infernal Tutor + Lion's Eye Diamond is actually quite good. Demonic Tutor + Lion's Eye Diamond offers the exact same thing and additionally comes with the bonus that Demonic Tutor is more versatile than Infernal Tutor, but it took the demonstration of the success of the more constrained Infernal Tutor to drive the point home. I think so, anyway.

In this case, there's a huge confounding variable in that Demonic Tutor has usually been banned or restricted in tournaments and that Lion's Eye Diamond was restricted in Vintage, so reliably getting both cards at once was unlikely. So was it a card availability thing that ruled out the interaction or was it a "discarding my hand sounds bad" thing that caused people to shy away from what was actually a strong combo? I contend that both situations were simultaneously true.

The "fun" to "win more" ratio particularly strikes me as relevant to both Magic generally and CPA more specifically In terms of historic context, when you have little pressure to affect your opponent before turn 7 in some games, being more mathematically efficient might have had a lot less utility than something like, "This combo piece is both a cool card and always starts a fun conversation with my opponents!"

Yes, but this was the Pro Tour. I'm not objecting to slow decks or even to slow formats. My point about that Mike Long game was that Mirage Block Standard was such a slow environment that a combo deck didn't need to be as fast as we usually think of them being these days. And the reason I was bringing that up was my thesis that Lion's Eye Diamond was a viable card from day one. Looking back through the thread, I admit up-front that the card was widely written off as unplayable for tournament purposes and then became a key component in a dominant Vintage deck and was restricted in 2003, seven years after it was introduced to the game. Now it's a sought-after card, easily worth over $100 on the secondary market and universally recognized as powerful. One could argue that it was the cards printed in 2003 and since then that have made LED powerful, that it didn't have the appropriate tools to make it viable back in the days when players thought it was a bad card. In other words, one could argue that players were generally right about it and that it was the environment that changed, that new printings made LED go from a bad card to a good one, or at least from a casual novelty to a tournament staple. But my contention is that this is wrong, that Lion's Eye Diamond was strong from the start and that players, even the pros, missed this because it was too far afield from their prior experience. To support my conclusion, I try to show how the peculiarities of tournament format structure created what look like "near miss" scenarios, where a tournament deck that exploited the free mana from LED might have shown the Magic community how strong the card was, but set rotation happened right on schedule to rule such a deck out, or there just wasn't much time, or players were just oblivious.

I think the ProsBloom case is a very intriguing one because the deck is famous for breaking a kind of lull: combo decks existed from the beginning of Magic, but they relied heavily on powerful cards from the original core set and once tournaments used bans, restrictions, and the "Type 2" format rotation, combo decks practically went extinct up until Visions in 1997 combined with the other recent cards in Mirage gave rise to ProsBloom decks. Supposedly, one of the ProsBloom players (I remember Mark Rosewater talking about this, but I forget whether he said it was Mike Long or David Mills) demanded that WotC R&D confess that they built the deck into the sets on purpose, because Visions seemed to introduce exactly the cards needed to enable this combo deck. It all seemed to intricate, so intuitively engineered, the flow of gameplay from early hand-sculpting to the Squandered Resources + Natural Balance mana ramp into the Cadaverous Bloom engine, etc. And yet, they didn't use Lion's Eye Diamond, a card that is now associated with broken combo decks. Well, why not? My suspicion is that the pros just missed it, that the card seemed too extreme at the time, even for them. Another possibility, a counterpoint to my view, would be that LED wasn't optimal in the ProsBloom decks of the time, that they couldn't afford it. I don't know if mentioning the slow nature of the format does much to support my case on this one, but I decided to bring it up anyway.

But the important part is that this context was my reason for mentioning the slow format. It was, nevertheless, a tournament format. I wasn't making value judgments about format speed or the considerations of format speed in casual gameplay. That's an important topic, just not one that I was trying to cover here.

 

[Oversoul]

I'm listening to the new episode of So Many Insane Plays, and there's a bunch of stuff about Doomsday, a fascinating card that I couldn't really do a Memories thread on because I hardly touched it myself. Stephen Menendian helpfully dug up the oldest Doomsday deck he could find. Even though I talked about the 2005-era Vintage Doomsday decks, I forgot to mention the original late 90's Type 1 Doomsday. I was new to the game at the time, but man, I wish I'd thought to look for these decks before. They might represent the oldest major tournament usage of Lion's Eye Diamond. I knew they existed, but I didn't really understand how they worked and apparently I forgot about them when I wrote that other post about Doomsday earlier in the thread. Well, here's the oldest Doomsday list Stephen Menendian found. It's from December of 1997, which I think represents the oldest competitive deck to use Lion's Eye Diamond that I've ever seen...

3 Doomsday
3 Arcane Denial
2 Abeyance
2 City of Solitude
2 Mystical Tutor
3 Vampiric Tutor
1 Demonic Tutor
2 Brainstorm
1 Ancestral Recall
1 Time Walk
1 Timetwister
1 Wheel of Fortune
1 Meditate
1 Braingeyser
1 Fastbond
1 Regrowth
2 Squandered Resources
1 Zuran Orb
3 Dark Ritual
3 Lion’s Eye Diamond
1 Black Lotus
1 Mox Emerald
1 Mox Jet
1 Mox Pearl
1 Mox Ruby
1 Mox Sapphire
1 Library of Alexandria
1 Strip Mine
4 City of Brass
2 Reflecting Pool
4 Gemstone Mine
4 Underground Sea
2 Bayou

Sideboard:
2 Abeyance
1 City of Solitude
1 Arcane Denial
4 Pyroblast
3 Hydroblast
4 Barbed Sextant

The goal was to Doomsday into Timetwister, Regrowth, Black Lotus, Lion's Eye Diamond, and Braingeyser. Exiling your graveyard and library meant that you'd only have those five cards plus Doomsday and whatever you had in your hand and on the battlefield. Ideally, you'd have a 2-card hand with Doomsday on the stack. Then Timetwister and Regrowth could loop, with Black Lotus and LED providing mana to make the loop infinite. Braingeyser with infinite mana available can usually kill your opponent.

 

[Oversoul]

As Stephen Menendian also noted in the podcast, that Doomsday archetype first showed up in late 1997 and was mostly active in 1998. It saw some refinement, but was superceded by newer combo decks. Then Doomsday itself was restricted in 1999 as part of the massive Type 1 restriction waves that happened in the wake of Tolarian Academy. Since "Doomsday pile component" was a severely confined niche for such a broad format, Lion's Eye Diamond essentially went dormant until Storm brought it back to tournament play.

So the Doomsday pile usage and the pre-erratum hack in Memory Jar decks do flatly contradict my narrative that the card had no tournament presence until 2003, but those are exotic cornercases. Notable, but in general it's still the case that the card was undervalued for years.

 

[Oversoul]

Reviving this thread yet again! For no particular reason, Lion's Eye Diamond is continuing to appeal to me as one of my favorite cards in the entire game. I find it fascinating because LED is a powerful mana accelerant that is essentially 100% locked in as a combo deck tool. Black Lotus could be used generically to cast Serra Angel faster or whatever. But Lion's Eye Diamond needs proper planning and synergy in order to function. And the discard aspect gives it additional functionality, beyond even what Black Lotus can provide, for a graveyard-fueled deck. I was going to refer to the Comboist Manifesto article I wrote about the graveyard as a game mechanic and resource, but then I realized Spiderman still hasn't posted that article!

Anyway, I'm perhaps slightly obsessed with this card. It's basically the best. Curiously, LED is currently propping combo up in Legacy while its usage trends downward in Vintage. In the thread for Dream Halls, I've been emphasizing that some cards are powerful when they can be used reliably in playsets, but don't really work well as singletons. While Lion's Eye Diamond is not subject to this problem as much as Dream Halls is, perhaps the same principle does apply to a lesser extent...