Spotlight on Sylvan Library

Sylvan Library is my favourite Magic card.

At a cost of one and one green Sylvan Library is easy to cast and easy to integrate into a deck that has some green in it. It is therefore a card to consider for a lot of people playing old school or eternal formats.

The value of shuffling

Most of you who played back then, and many who played it later, probably have heard of the Sylvan Library Shuffle as a “shuffle to see 3 new cards”. That seems true enough doesn’t it ? Yet it’s misleading in more than one way.  But first let’s define what we call a “Sylvan Library Shuffle” (SLS) :

  •  unless you payed life to draw additional cards or other stuff happened that displaced them, your last SL use left you with the knowledge of what are the top two cards of your library
  • regularly those two are unwanted as you kept the best of the 3+ you saw last turn and left cards that are deemed useless in the current situation, underpar compared to what you’d expect from two random cards from your library, or just not what you want to find to solve a problem that needs solving fast
  • shuffling your library would give you a chance to see two new, potentially great cards in their stead

The biggest mistake is to count the third card. Sure you’ll see three cards at least if your SL is still there by the draw step, but even if you didn’t have a shuffle effect, you’d still see a new card : the third one. So remove one from that saying.. and you’re still not there : you might see two new cards, and you might just shuffle your top two cards right in their initial place. Of course that would be a rare occasion, but what would greatly affect the novelty of the draw is how many copies of those two cards on the top of your library you play, and still have in your library.

Now let’s calculate the chances that at least one of those top two cards is seen again post-shuffle, and compare them with a more diversified, less redundant deck : if the first card you don’t want to see is in 3 instances in your library, the second in 2, then overall the chances to see at least one of them post shuffle is 19% for a 50 cards-library. Since those are reasonable numbers for a deck with many redundant cards, I’d say in this case you’d really only see 1.8 new cards with every shuffle. For a 30 cards library if one of the two cards is in two, and the other is the only one left, you’d get the same chances to see at least one of those cards back. Now if you had a very diversified deck, it would make sense to calculate the odds for a card in two and a singleton in a 50-card library : 11.7%, and the odds for two singletons in a 30-card library : 13% (all this computed thanks to the app at Seems like by diversifying your deck, you could make the Sylvan Library shuffle an almost two-new-cards effect, though more like a 1.9 one (and decreasing more abruptly than in the more redundant deck). The difference between those odds aren’t gigantic (around 5% each time you SLS, although over the course of many SLS turns the odds of getting something out of it would accrue) but they add up to the benefit of diversifying a deck when you have access to a lot of card filtering, which are enormous : in a deck deprived of cards filtering and card draw, maximizing copies of the good cards is the way to try to beat the inherent variance of the game, in a slow deck with a lot of card manipulation, diversity is key to making those seen cards matter, as the more diversity you have the more solution to threats or angles of threat you can find.

Such is the hidden charm of Sylvan Library : it improves with the diversity in your deck, and thereby gives incentive not to build excessively redundant decks, as we so often do. You’ll still want to play your best cards in multiple exemplars, and unlike in singleton formats like highlander or EDH you can build your deck with maximizing synergies and combos in mind, but I think you’ll see that, generally, successful decks using Sylvan Libraries as one of its main card are more diversified than the norm, and I’d say they tend to hit the sweet spot between the redundancy of the everyday competitive deck, and the difficulty of assembling combos in singleton formats.


Land Tax
What else ?

Which segues nicely into a deck we couldn’t think to omit when examining Sylvan Library : I mean ErhnamGeddon of course. ErhnamGeddon rose to fame in the second quarter of 1996, after Lestrée reached the final of the first Pro-Tour, for (again) a defeat against a clearly inferior deck. That’s the deck and era that put Sylvan Library in the spotlight for most of us “old-timers”. I don’t know how much the originators of that deck thought about it that way, or if the design just came into its own that way through a combination of trial, error and “natural” selection, but in the end, ErhnamGeddon is one of my favorite deck design for this very reason : it’s almost impossible to sideboard against (you’d have to have access to such an improbable card as Dystopia.. which, ahem, of course is exactly what

The name is proof of the conspiracy : that’s just retarded fake news !

they released just a few months after the deck rose to preeminence 😦 ), as it doesn’t push in any direction, instead opting for a delicate balance of all forms of threats and kinds of spells. A bit of excellent artifacts, creatures, enchantments, spells, a bit of green, a bit of white.. the diversification of which makes the deck so elegant and make the inner, deeper working of the library so congruent : looking at the decklist, ErhnamGeddon while certainly not a singleton deck, still plays many of its cards in limited numbers, and disposes of a large variety of effects. To me ErhnamGeddon is the initial if not ultimate deck of class and style. Up to that point each and every powerful decks I knew were unbalanced, narrowly focused, often almost maniacally so. Sure ErhnamGeddon played Land Tax in a deck that was set up to abuse it, and made it so broken that it would be restricted first chance. Now broken cards can be seen as vulgar, but the rest is all subtlety, the setup not all that self-evident, and Sylvan Library is magic elegance at its purest.

Winds of Change does shuffle, though this time I wouldn’t consider playing it in combination with Chains of Mephistopheles nor Underworld Dreams ! And the card might be unworthy without those..
I never realized Shahrazad made people shuffle ! Just doing a search for “shuffle” isn’t actually enough, as playing a magic subgame means that after the subgame is over, people will shuffle the cards from the subgame into their library !

Fork you.


In 93/94, or old school Type I you find cards known to be excellent in their own right that do shuffle your deck : Timetwister, Transmute Artifact and last but not least Demonic Tutor.

Timewalk to ’95 and we find some more interesting shuffling cards in Ice Age. Portent is a nice and underrated shuffle effect plastered on a cantrip. Nature’s Lore I’ve already talk about : that card’s probably better than you imagine. From the often ridiculed Homelands set, Merchant Scroll is a great Type I shuffling effect, adding an Ancestral Recall or Mana Drain to your hand on top of the shuffling. Going further down the line, and outside of the Eudogames format into Old School 96 : from Alliances, Diminishing Returns too can be a very powerful shuffling effect.

Magic: the Shuffling

But it is Thawing Glaciers that actually may have the longest history of successful collaboration with the Library. The Glaciers are already great in themselves, fixing your mana in both quantity and colors while generating card advantage, but as they do so you quickly end up with two of those which mean you can shuffle every turn ! That’s a much more reliable shuffling engine than Land Tax, albeit a much less broken one, which ain’t a bad thing too.

No Shuffle.

Voila. 2 Blacker Lotus for your 3 Moxes.

We talked about ErhnamGeddon, but possibly the first top deck that used the Sylvan Library “extensively” was Adam Maysonet’s Rack-Balance deck (MRB). Bertrand Lestrée’s zoo deck from the first world championships final played it, but didn’t abuse it and it only played one. I already talked about MRB when examining the Balance’s “case” in 93/94 so I won’t add the list here : it clearly shows that one of the “engines” of the deck is the Sylvan Library + Bazaar of Baghdad combo. While second in that deck to the Bazaar + Library of Leng one, it is nevertheless the most consistent since the SL is a good card on its lonesome and the Bazaar, while not great without a card to combine with, can be put to use now and then, unlike the lone LoL which is well, lol. Bazaar will get those two unwanted cards out of the way, at the price of not having a hand, which some decks aren’t uncomfortable with, since they’re fast or want to put stuff in the graveyard for reanimation anyways. In that case, it does indeed remove them, unlike a shuffle effect that keeps them in the library.
Another effect that does just that is Millstone’s ability. It may seem strange to mill

My first Magic crush.

yourself in a mill deck : you win by having a bigger library than your opponent, so milling yourself in that case isn’t like the mill you can get by Bazaar as that one is typically played in relatively aggressive deck, or at least non-turtle speed decks like ctrl-millstone. But to win in a control deck, filtering cards that way results in virtual card-advantage as for instance useless lands that would have gotten drawn are replaced by useful spells, sculpting your hand to be ready for everything. Also you might just be under pressure to find a solution to an urgent problem, and having the SL-Millstone engine to filter your card every turn is far from little help, it’s technically filtering two cards away, seeing three cards every turn instead of one : you get to your solution card three times faster. An old school deck that’s set up to use that interaction is the “Conceding Dreams” one.
You don’t necessarily have to mill the cards though, Demonic Consultation, already one of the most influential card in the old school formats it is legal in, let you try to get the card you want at the price of exiling at least your top 6 cards, which means a significant part of that drawback works to your benefit if you have a SL with two unwanted cards on top.

Orcish Librarian : nope. That one does it all by himself, he does his own shuffle, and he sort of let you choose among 4 and then less, so it pretty much has the SLS all included in one card and would make the Sylvan Library quite redundant : the librarian is awesome, and is the secret magical overpowered core of Sligh. Books are good ! Jayemdae Tome, Library of Alexandria, Sylvan Library, Browse and the Librarian : all awesome cards in old school formats.

Orcish Librarian.jpg
You’re just too good to be true..

Anti-Discard Tool

Funny enough, just like in the MRB, the Sylvan Library is second to Library of Leng too here as an anti-discard measure, since the LoL will protect your entire hand, while SL can only protect the cards you see with its effect, and up to a limited point (if they pile up, you’ll tragically have to draw one of those great cards, poor you). Maybe what I really should be doing is an article about LoL ? Anyways, SL is a quite interesting anti-discard measure. That tech was used a lot during the heyday of the ErhnamGeddon/Necro clash for Type II dominance (that’s how Standard was called back then) : you would keep the critical card on the top of your library, so that the expected discard (Hymn to Tourach mostly, though sometimes a Hypnotic Specter would connect) couldn’t get it. You’d often keep a disenchant, so that the Necro player couldn’t make you discard it and then play their Nevinyrral’s Disk safely. Another card that you would keep on top is the Balance. Since ErhnamGeddon doesn’t play any counterspell, it would often see its hand destroyed. That’s when delaying the Balance draw pays off : all that discard and any advantage accrued on board in terms of creatures and lands is then canceled by that tactic. Of course this would sometimes come at a cost, if you have a card you need to keep on top, then you lose the opportunity to use the SLS for profit. And you can’t say that that’s the normal order of things, that you don’t shuffle when you have great stuff in your top two cards. The normal order of things with SL is that you draw the great cards and try to shuffle away the lesser ones. Losing that option (if you could have shuffled AND you’d have wanted to) is no joke and needs to be taken into account when considering Sylvan Library as an anti-discard measure of sorts. Still, as an added bonus to all it can do for you, know that : Sylvan Library can save your life in the face of discard. It can be quite skill-testing to play that way, as you have to measure the costs of not shuffling versus the benefit of keeping the potentially crucial card away from discard measures, which in my book is a definite plus !

4 life per card

And after all those way to benefit from the library, we still haven’t talked about the card-advantage aspect ! 4 life per cards is a poor rate compared to Necropotence for sure, but Sylvan Library’s elegance is also in its moderation : the eternally great card that just won’t pass the threshold of brokenness or restriction-worthiness. The first clear way to do that is just to go ahead and spend your life points. Since you only need one to win, and some decks won’t hurt you (or not until very late in the game), there’s that way to punish those slow decks (and hence another reason why Lestrée was the clear favorite in that Pro Tour I final, as ctrl-Millstone is exactly the kind of deck you’ll feel comfortable paying life against). You can therefore make some reasonable assumption if you see your opponent hasten to counter or destroy it : mainly that their deck doesn’t deal much damage, if any, at least not early on. In other words, even when it gets neutralized there are chances you end up with the upper-hand (since you might have derived valuable information about the style of your opponent’s deck from the situation) : dog how I love that card ! Still that’s only a fraction of the spectrum, and today’s dominating control decks in 93/94 have a significant array of cards to hurt, damage-wise. The means to gain life are in short supply in those old school formats, although things get a bit better with Ice Age and Zuran Orb. Ivory Tower is probably the first card that comes to mind when we think of life gain in Old School. The problem with it is that it limits you to win 3 life per turn, which means the combo isn’t exactly enough to let you draw two cards per turn. You could try to add another effect that would let you get to 8 cards during your upkeep. Such effects are usually excellent card-advantage effects in their own right, and might just pile-up and make the SL secondary (Library of Alexandria, Jayemdae Tome), or they might combine wonderfully with it (Land Tax, Thawing Glaciers), or none of the above and then some less (Library of Leng, again ? that card sticks !).

Library of Leng
I’d pay you to get rid of it.

But if you don’t have that, you could still draw three additional cards out of every 4 turn without hurting yourself with a full hand and an Ivory Tower.

Another famous way to gain life is Mirror Universe, that card has so much potential you often build your deck around it, as in the Mirrorball-Sylvan Mirror deck :

Deck by Martin Jordö
Mirrorball : I wouldn’t underrate that thing, were I you.
Spirit Link
That way to go potty !

What’s less talked about and I think a bit underrated is Spirit Link . There are some major cons to it, especially when using the modern rules. Let’s forget about the fact that the linked creature from your opponent can still block and therefore threaten to kill some of your attacking creatures, and let’s assume you’ve got that covered one way or another. Still, if you put a Spirit Link on your opponent’s Whirling Dervish it won’t be ideal because the Dervish trigger will happen, and it will grow, which means you could end up at the mercy of a Disenchant, but at least you’re safe as long as the Spirit Link is on the creature, right ? Nope, unless you use the Bazaar of Moxen rules which like the old ones states that you don’t die until the end of a phase, you’re in a losing position anyways. Even if you have still 20 life and nothing else happens, or whatever a player does, the other neutralizes, you’ll eventually lose to a 20/20 Whirling Dervish because the Spirit Link effect never gets to resolve, as you’re immediately state-based-checked for having 0 or less life right after it goes on the stack. In a more realistic setting, if you put a Spirit Link on your opponent’s Erhnam Djinn, that Erhnie still represents a temporary 4 life points loss, which of course might end up killing you. Since Spirit Link doesn’t prevent damage, it won’t prevent Hypnotic Specters from making you discard either, which is another terrible con. Still, in a format poor in life-gaining tool, I think that card can and should be considered in some decks where it would be a three-pronged weapon : a creature removal of sort, a life gaining tool, and a part of a card advantage engine.

Swords to Plowshares is the best creature removal of all time. Unless you’re playing against a deck that can transform life into cards. While Necro is much more potent at that exchange, ErhnamGeddon is usually more interested in playing big “swordable” creatures. Anyways, a StP on an Erhnam means that SL will translate that life gain into a card. In other words SL will regularly make StP a poor card ! The card is so prevalent, that if you play “fatties” alongside Sylvan Library, they almost become life-gaining, self-replacing cards.

There are other life gaining cards, but at the risk of underrating some, I’ll assume they’re generally not played. Beside, it would take you nine mana to get a single card advantage with Stream of Life + SL !

Anti Millstone

Well that’s quite simple : if Milling your top two cards is usually great, then your Sylvan Library make your opponent’s Millstone very frustrated. Add that to the “Lestrée should never have lost that final” pile.

Indirect CA

There are as far as I know three playable cards that combine with the Sylvan Library to generate card advantage by exploiting the knowledge it gives you of your top card  : Sindbad, Petra Sphinx and Vexing Arcanix.

Sindbad the Sailor.. No islandwalk ?

The Sindbad + Sylvan Library I’ve already explained here. Petra Sphinx, while clearly hard to cast, has something of importance to note about :  it can be one of the very few card advantage engine that doesn’t draw cards and is therefore unaffected by Underwold Dreams. You’d still suffer three life points every turn to make it work under the UD, but if you could, then at least a Sphinx wouldn’t hurt you any further. That’s actually the case of our last card too. That vexing card can actually make for a decent wincon too ! Petra Sphinx could pass for a hopeful half-Millstone effect but that’s not much to talk about, though to be fair none of those would work with an opposing SL, an active Orcish Librarian and, more dangerously since it could surprise you : Brainstorm.

Draw replacement

First off, the rules for managing Sylvan Library with replacement effects can be tricky, and has been the subject of its own page on the official magic judge blog.

Island Sanctuary.jpg
Another combo. Can you see it ?

Island Sanctuary isn’t all that tricky, since the interesting way you could make it combo with Sylvan Library isn’t legal anymore, and hasn’t been for a long time. That’s a shame if you ask me as that combo is quite interesting and powerful in the type II formats of ’96 and ’97. How it went was like that : draw phase you’d activate your Sylvan Library. You’d then replace one of the draw by the Sanctuary’s effect, gaining your one-sided Moat effect and the knowledge of your top card. You’d then put your card back and not pay 4 life. How ? Well, you haven’t drawn a card this turn other than the one you’ve drawn from the SL, so you can’t choose two cards to replace, only one, and you’ve replaced it. Then you may activate your Thawing Glaciers to shuffle away the unwanted card, and then at the time of your choosing you can draw your card for the turn. That interaction was made clear at the time of the “5th Edition rules”, that is when Mirage came out (cause those rules weren’t confusing enough, so they had to happen before their namesake !), since the rules made it explicit that you’d draw your card for the turn “in the middle” of your draw phase, people started to wonder if they could abuse Mangara’s Tome with it (proof that you can). If, knowing about the 5th edition rules, you think that only came to be with those rules you’d be mostly wrong. The only way that you’d be right is that by making that fact clear and available to a lot of people, those combo started to materialize. Which doesn’t mean that they weren’t before ! Yes, in fact you could have done the Sylvan Library + Island Sanctuary trick right from its inclusion in type II (in type I, today called Vintage, you’d want to play a Moat, simply, I imagine) ! That’s why it’s a bit frustrating for me to see that being impossible in Magic 95/96, it’s not like I’d have a deck that would beat ErhnamGeddon and Necro were that legal, but still, I really like a Moat effect, and tricky combos too 🙂

ErhnamGeddon was far from being Sylvan Library’s last cry, in fact Sylvan Library was to be a key Type II card in the years to follow (see for instance Alan Comer’s Goofy Gaea’s deck), with often one Island Sanctuary maindeck to exploit just that trick.

Adequate name.

Possibly the most famous and certainly the most abusing sort of old school draw replacement card to combo with SL is Abundance. That card got a lot of exposition thanks to how popular was Bob Maher and his display in the final of Pro Tour Chicago ’99  (the final doesn’t see that combo getting used, but as you can see there, it was part of the deck). I’m jealous I never got to try that (I wasn’t playing mtg at the time). In case that wasn’t clear, every Abundance you control let you get (and not draw) 2 additional cards without any penalty, after you’ve replaced every draw instance since the beginning of the turn with its effect, each of which is of your choosing between land and non-land, each of those choice done with the knowledge of what you got before that (each card draw is sequential, drawing two cards is doing two things sequentially, so each replacement effect happens before the next one there).

Leaving old school formats for more recent stuff, the dredge mechanic itself works as a draw replacement effect, which means that you could use it to dredge three cards per draw step without penalty. But apparently that’s not good enough for the competitive Dredge decks.. so let’s move on.

How Many Copies To Play

It’s no happenstance I deal with that matter right after introducing some interesting draw replacement effects. Those make it more valuable, and easier to include more Sylvan Library in your deck, since the card is so good as to be still played in such a high-level format as Legacy, and even in fact now and then in Vintage, it is usually desirable to draw it relatively fast. If you use the Island Sanctuary tech, the first one is almost entirely absorbed in the combo, so the following one isn’t redundant, and a third one would be as redundant as a second one if you didn’t have the sanctuary. And if you don’t have a Sanctuary or anything to absorb your first SL, the usual old school number of copies played was two. With a Mangara’s Tome, it’s already quite a lot to spend 6 mana to abuse your combo fully, so a second is generally about as unesuful as if you hadn’t. When considering Abundance then all bets are off. If you can reliably have a way to abuse the SL (the deck that played Abundance often also played Island Sanctuary), then my personal biased opinion is that you might as well play 4 !


Sylvan Library isn’t weak to Disenchant and Counterspell. Both of those cards cost two, so they’re at parity in terms of tempo and card count (although the second asks for two colored mana, and availability). Tranquility doesn’t see much play and would only be profitable if and when you’d have several SL in play, combine those two factors and it’s mostly not weak to it. While it would seem black with no way outside of Nevinyrral’s Disk to get rid of enchantments should be in the worst position to weaken Sylvan Library it is actually from its ranks that comes its worst enemies, namely Underworld Dreams and Chains of Mephistopheles.

We meet again, old friend.

At least you don’t have to draw the cards (SL’s trigger will go on the stack, but it gives you the choice of not drawing the cards), but then they’re much better than a disenchant if they stick in that case since a single one nullify each and every one of your SL copies, unless you’re ready to pay the price, which is enormous with the Chains, and usually not affordable for long (if at all) with Underworld Dreams. Other weak points that you might have gathered along the way is that not only you might not have any way to abuse it, like a Land Tax or a Bazaar, also even when you do there’ll be many times when you don’t want to as one of the top two cards will regularly be desirable since we’ll have to assume your deck is made of great cards, otherwise what are we talking about anyways ?

The Sylvan Library’s Legacy

Sylvan Library is the root of all card-filtering in Magic: the Gathering. To do justice to such an heritage isn’t something in my powers. Wizards made most likely more SL-inspired cards than I’d know. Brainstorm seem very clearly in that vein and has to be mentioned. But the more obvious examples possibly are in old magic Mirri’s Guile, more recently the broken Sensei’s Divining Top, and even currently the awesome Grim Flayer (wow, I actually alluded to a Standard card in an old school blog !).

Mirri's Guile
Reprint this, or else.

Unfortunately, after having made the case that the card is awesome in seemingly every respect, the sad fact is, Sylvan Library is very much unreprintable (outside of Commander sets and Judge promos of course). Some of the cards interaction, as explained with the combo with Sindbad are already prone to misunderstanding and doubts, but it goes over the top in some much more problematic cases, like the one that oddly involves one of its “progeny”, brainstorm : how do your opponent know what cards have been drawn and can be replaced this turn if a Brainstorm has been resolved before the SL ? The case is treated, so to speak, by the judging institution, but is still one of the most problematic rule interaction in the game, and more than enough incentive not to reprint it, for fear of any similar interaction with modern cards, along with general problematic enforcement, that asks of you to make a sub-zone in your hand, by putting the cards “drawn this turn” aside for use with the Library, despite the nonexistence of any such sub-zone in the rules.

Therefore I’m sad to say that Sylvan Library will remain the privilege of us old school and eternal formats players. Let’s all share a collective sob at that thought.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s