Knowledge of the masses: Innistrad Sealed I

Written by derflippi on October 06, 2011




Today I write about Innistrad Sealed, applying an unusual approach of a limited article:

1. The use of deck analysis in limited
2. Sealed events on magic-league
3. Sealed deckbuilding process in ISD
4. A few individual cards
5. Summary, a look into the future of limited analysis

1. The use of deck analysis
Competitive players always want to optimize their chancesat winning. One key decision in Constructed formats is the deckchoice. Deckchoices can never be done in a vacuum. RDW is barely the best deck in Standard(T2) when everyone plays Timely Reinforcements, but it is a valid option when everyone plays Ramp. Basically, it's a bad idea to chose rock in a field of paper.

An idea that has always fascinated me is to analyze metagames and metagame trends. In Standard(T2) or other constructed formats, similiar decks can be categorized into archetypes and then you look at the average results of the archetype to find out about its strength.

In limited formats: Sealed and Draft, analyzing archetypes is more difficult.

When the pool a deck is built of is random, the outcom can rarely be classified as archetype. One approach can be to look at decks by their colors instead. But then a green-black deck can be completely different to another green-black deck. Also, it's not very useful information knowing a specific combination of 30+ cards is awesome in limited when it's unlikely to receive even 50% of the same combination ever again.

Therefore I believe at this point it's a better idea to look at the individual strength of cards rather than combinations of cards.

2. Sealed events on magic-league
For that I minimize my own subjective (yet competitive) view on the powerlevel of the cards and concentrate on an empiric analysis of what the average player on magic-league thinks about playing or not playing a certain card in Sealed. Due to the nature of the tournaments on magic-league, Draft decks are not saved anywhere here.

As of October 2nd, 107 Innistrad Sealed events have finished on magic-league. I extracted the total numbers of played cards and sideboarded cards from the magic-league database. The original intention was to find information about  the playability of nishe-cards like Naturalize, One-Eyed Scarecrow, Spare from Evil and Ancient Grudge. How many enchantments, spirits, humans and artifacts are actually played? If on average, players play 5-6 artifacts or enchantments, then Naturalize is certainly a maindeck option.
During the process, one specific, extraordinary number catched my eye, so I explain that one first, moving the nishe-cards to the end of the article.

3. Sealed deckbuilding process
Some people say rares win matches alone. However, even if that'd be true, one couldn't do anything against it. Everyone knows that planeswalkers are good in Sealed. Some of the players with planeswalkers still lose though, because the rest of their deck is a total mess. They cut a complete color with all its removal just to play this one Sorin Markov?
I've moved to building a deck's core with commons and uncommons, and only then see if any rare is worth playing or getting splashed. This explains why I'm not putting the emphasis about how good or bad Devil's Play and Bloodline Keeper are in Sealed. I concentrate on commons and uncommons,.. and basic lands.

A first look at the quota of a card being played brought a big surprise. I expected top commons of each color in the general top: Victim of Night, Brimstone Volley, Claustrophobia, Avacynian Pries and Ambush Viper. Eight out of the Top10 cards are white is the information

Rarity Color Cardname MD SB quota
C A Blazing Torch 411 116 0,78
C R Brimstone Volley 360 191 0,65
C W Avacynian Priest 342 197 0,63
C W Rebuke 301 184 0,62
C W Voiceless Spirit 320 211 0,6
C W Bonds of Faith 285 243 0,54
C W Smite the Monstrous 287 246 0,54
C W Chapel Geist 287 247 0,54
C W Thraben Sentry 322 284 0,53
C W Elder Cathar 301 272 0,53



The same story continues in the uncommon slot where the Top4 are white too, even surpassing artifacts, that fit into almost any deck: Silver-Inlaid Dagger, Galvanic Juggernaut and Geistcatcher's Rig. The majority of players is more likely to build a deck using Mausoleum Guard than with Silver-Inlaid Dagger.

While this is only a small piece of information: It does not tell exactly whether Thraben Sentry or Dead Weight is better, it clearly outlines that white is, on average, superior to all other colors in Sealed.

That being said, let's look at the other colors. The top played cards are

Red Blue Green Black
Brimstone Volley 0.65 Moon Heron 0.4 Ambush Viper 0.42 Dead Weight 0.51
Harvest Pyre 0.42 Claustrophobia 0.39 Villagers of Estwald 0.41 Vampire Interlopers 0.43
Geistflame 0.4 Stitched Drake 0.39 Festerthide Boar 0.4 Victim of Night 0.42


Blue and green seem weak. The best green cards are creatures, one being a mere 2/3 vanilla until turn five or so, and then still not winning against Thraben Purebloods.

Blue takes only 3 spots in the Top40 common cards, meaning its commons don't make one play the color often. The best uncommon being on place 16 of all uncommons seals the destiny of blue.
A quick look at the total baisc lands being played further evidences my thesis:

Name total
Plains 4023
Island 2640
Forest 2852

On average, blue and green pools are bad.

I showed my approach should work for players who build their decks bottom-up. If deckbuilding is concentrated on the base first, white is the strongest color, followed by black and red.
Now it's also possible that a player sees a rare/mythic bomb and then builds his deck around it. How much is the chance of a player playing a given rare or mythic?

Starting the Sealed deckbuilding process with rares, the rare which people are most likely to put it in their decks is a card probably noone expected here: Manor Gargoyle. Lumengrid Gargoyle upgraded is a lock-in in every deck. It's not the best card of the set, yet people are never sad opening it. Following shortly in descending order: Angelic Overseer; Mikaeus, the Lunarch; Olivia Voldaren; Bloodline Keeper; Devil's Play; Geist-Honored Monk; Mentor of the Meek; Angel of Flight Alabaster; two green rares; Reaper from the Abyss; Balefire Dragon; Army of the Damned; Instigator Gang; Liliana of the Veil; Elite Inquisitor; Divine Reckoning, Bloodgift Demon and Falkenrath Marauders.
Within the Top20 rare and mythic rares, that's every rare that gets played by at least 60% if it's in a player's pool, there are seven white, six red, five black and only two green rares.
I conclude a player who builds his deck top-down will rarely end in green, and probably never in blue. Seeing Elite Inquisitor, a good yet not really gamebreaking card has a higher chance to end up in a players deck than every blue card (including C/UC) and with the exception of Mayor of Avabruck and Kessig Cagebreakers every green card(including C/UC) is a clear sign that barely anyone trusts into green or blue.

The knowledge of the average player, the majority of magic-league and any Grand Prix says the same, regardless on their approach of deckbuilding. The average white deck is certainly better than the average blue/green deck. If no color is spectacular, I suggest to build white-black or white-red deck, possibly splashing the thirft, non-green non-blue part.

That much about knowledge of the masses on the topic of color valueing in Sealed.

4. A few individual cards
Let me now add a backed up reasoning to play or not play a few special cards.

To create cardadvantage off Geistflame, I'm looking for creatures with toughness one. Of corse, Geistflame does more than killing X/1 creatures, but using it as Crushing Pain is not the best way to go. The list of worthy targets is quickly assembled, backing up what most intuitionally believe anyways: Voiceless Spirit, Vampire Interloper, Avacynian's Pilgrim, Markov Patrician and Kessig Wolf. Yes, Geistflame is indeed great.

Valueing Ghoulraiser does not make much sense here as it doesn't depend on the incomplete information of your opponent's decks but depends on your own pool, of which you have complete information.

About Wooden Stake: Although the majority plays W/B/R, which means several Vampires, many of which are regularily played too, it doesn't do anything to them. Either they're evasive; or they have a toughness of one or two only. This is not a maindeck card.

Back to arguing with empirical data, One-Eyed Scarecrow is up for many discussions. I check how many "targets" are there and how relevant the -1/-0 malus is to them. We know most decks use white now, One-Eyed Scarecrow shuts down all spirit tokens and weakens the regularily played Voiceless Spirit, Chapel Geist, Abbey Griffin and Vampire Interlopers significantly. Although only a relatively small amount: 15% (of players that have it in their pool) maindeck this card: the magic-league community rates it as bad, it should have a place in more maindecks I think.

Naturalize is a green card. For whom this doesn't suffice, the most played artiffacts are Blazing Torch and Wanderer's Twig which are being sacrificed quick anyways. As for the uncommons: the average player uses one artifact per deck. An old thumb rule in limited is to play one silver-bullet (cards with very specialized use only) per three cards the opponent uses of that type. The quota of played artifacts and enchantments being so low makes Naturalize a bad choice. Aside it's green.

The same logic applies to Ancient Grudge, except that card is not green...

5. Summary, look into the future of limited analysis
Summarizing my points, I say that when in doubt, you should dismiss green and blue in Sealed. If possible, open white rares and white common cards, add black and splash red.

The latest 6-0 ISD Sealed Trial deck was built bottom up: first, the black-white base was found, then Brimstone Volley and Olivia Voldaren splashed into it. I believe the deck would've worked without Olivia Voldaren too, if it were replaced by the average Abbey Griffin or simliar.

After GP Milan, probably 250 ISD Sealed tournaments, including a Master will have taken place. That's enough data to go one step further from average player to the tournament top players. How is green/blue valued among the top and do they play white as often as the average user?

This time the analysis did not prove what's good. So far I just displayed what everyone thinks is good.

...Until next time, become a better player by playing on magic-league.com.

Back to Magic: the Gathering Articles

Comments:
by derflippi on 2011-10-06 16:57 CET

The complete data table can be looked at at http://tinyurl.com/65m3kvw and http://tinyurl.com/6h43aye


by sc4rs on 2011-10-06 18:42 CET

"This time the analysis did not prove what's good. So far I just displayed what everyone thinks is good."

I think this should be emphasized. Unless you're only looking at the color data for players that go 2-1 or better, "what most people play" doesn't necessarily mean "best" at all. Telling people "dismiss green and blue" when, on average, they already are seems shortsighted - tell them to re-evaluate their green and blue because they may be relying on borderline cards like Elder Cathar to signal them to go into white when they have better cards in other colors.

I say this because blue in this sealed format is actually pretty powerful when running past the normal "play every good removal spell and see what colors that means you're in" sealed strategy. All it takes is opening one or two Curse of the Bloody Tome and/or Selholf Occultist, and you have an extremely powerful all-common mill core.

Plus, red/white/black tend to be overvalued in the early goings of any sealed format because that is where the majority of the removal lies. That usually persists for a while, and isn't always "wrong," but is, again, another confounding variable to consider when asking "does this data actually mean white is the best sealed color?"


by P_P4E on 2011-10-06 19:03 CET

there are too many playable in this format, and differentiating them by power level is also hard. Hard meaning challenging I guess, but I also think they could have included a couple more useless cards to suss out the colors better and give you less of a problem deckbuilding. I've played 2 sealed IRL and a million online and every time I get at least 4 colors that have 4-5 solid playables. Maybe this isn't the same for everyone else, But I eventually just build 2 color plus splash and include the most removal I can. There is too much removal in this set.


by GunsAndDope on 2011-10-06 21:53 CET

you need to weigh the card choices with how well a deck did in a tournament versus how many rounds the tournament was.


by pascal3000 on 2011-10-06 22:02 CET

The most shocking statistic out of this article:
22% of Magic-Leaguers dont maindeck their Blazing Torches (well, probably a little more even, because its 22% of blazing torches, but the good players are likely to be more active players and play a bigger role in the statistic)...


by Eldohir on 2011-10-06 22:04 CET

I won my prerelease with a UG deck... am I bad ???? :p

nice article anyway ;)


by Shooter on 2011-10-07 00:23 CET

Can't see me ever letting Torch on my sb.


by pugz on 2011-10-07 00:27 CET

no eldohir but the other players may have been, or you got a nice tempo deck =)


by Diogenes5 on 2011-10-07 01:53 CET

The analysis seems spot on and is similar to SOM sealed only in SOM sealed you could be lucky and open a sick poison pool.

White has basically all the removal in the format and at common level. Red has like 3: geist, brimstone, and harvestpyle. It's afar cry from its removal in SOM where turn to slag and galvanic blast were amazing and shatter was defacto removal.

Blue has one strict removal (Claustrophobia) and Green has one as well (the fight card).

Creature-quality wise, white has nice creatures that fit the curve with 2 good common fliers and a nice 5/4 beater. It is not surprising that white is often the color chosen.


by mishimakaz on 2011-10-07 02:46 CET

Immediately made a O_o face when I saw only 78% MLers maindeck Blazing torch.


by Pushover on 2011-10-09 09:27 CET

I kind of feel like Naturalize's part about hitting enchantments was overlooked. Claustrophobia for anyone in blue is going to be at least 1-2. The white pacifism is probably most commonly played as a removal spell, given how slow the format is most of the time. The black curse that makes your creatures suck can also take a seat.


by Trivial on 2011-10-10 00:51 CET

Excellent analysis.


by Trivial on 2011-10-18 03:43 CET

Any correlation between number of lands played and record?


by Conkisstador on 2011-10-26 03:32 CET

Awesome article.
BUT! Why are there 30% more Thraben Sentries in pools than Rebukes? O YEAH, you're sealed generator is broken. Fix this problem so we stop seeing an abnormal number of Villagers of Estwald and Thraben Sentry.
Also, your conclusions are off. You forgot the anomaly that White is indeed SO good that you don't need many cards in another color and thus U/G often play the best support role. At both top8s at recent NewEngland PTQs, 6/8 of the sealed decks were WG or WU


by ShrimpHouse on 2011-10-31 07:01 CET

True I think there are some problems with the sealed generator that should be ironed out. That kind of takes away from these findings. Still good methodology however and I look forward to reading more from you in the future.


by limited on 2011-12-12 14:30 CET

i like green a lot in this format. blue has looked good to me on paper but i think i lose quite often when i go blue.


All content on this page may not be reproduced without written consent of Magic-League Directors.
Magic the Gathering is TM and copyright Wizards of the Coast, Inc, a subsidiary of Hasbro, Inc. All rights reserved.


Contact Us | Privacy Policy
Join Swagbucks!