Decks of Jay Schneider: Red Dawn
Jay Schneider

Greetings All,

Block is out and it's time for a new Type II. As usual the first deck I've built for the format is a Sligh deck. In Onslaught Type II, realizing you need to build and test a Goblin based Sligh deck isn't exactly rocket science. However, the details of the build (as always) are trickier than they initially appeared.

In magic as in most things, no one develops in a vacuum. I'm lucky enough to have steady testing, feedback and suggestion from friends across the US:

In Seattle: "The Seven Samurai - Food Court Warriors" and the Microsoft Cafeteria Magic Players. Thanks to Coyote Games in Redmond, Cardhaus games in Seattle and the Wizards Store in Crossroads Mall for providing Seattle players places to play.
In Oregon: Dave Meddish, Mons Johnson.
In CA & Elsewhere: Chris Cade, Sean Frackowiak, Rick Saunooke & Jeremy Pickett(Kelmage)
In Atlanta: Dave Leader, Sol Malka, Andy Wolf and the ever reclusive Paul Sligh (who just got married last week!)

"Deck building is in the details."

Red Dawn Jay Schneider
	Format: New_Type2
»Apprentice Format«
Read the article that included this deck HERE
Main Deck	Sideboard
4 Goblin Piledriver 3 Goblin Pyromancer 4 Goblin Raider 4 Goblin Sledder 4 Goblin Taskmaster 4 Grim Lavamancer 4 Sparksmith 1 Final Fortune 4 Firebolt 4 Lava Dart 4 Reckless Charge 4 Bloodstained Mire 1 Goblin Burrows 11 Mountain 4 Wooded Foothills	3 Fledgling Dragon 4 Meekstone 4 Pillage 4 Stone Rain
Total deck value: $
Notes:
NONE
Average Casting Cost (ACC)= 1.83 Spells by Color: Cards by Type: *Hover mouse cursor over pie chart to see ==>> COLOR/TYPE : # CARDS : % OF CARDS. *Left click on pie chart or legend to isolate a field.

O.K. this isn't really what most people expect out of the Goblin Deck from Onslaught. It's full of pricey rares, doesn't have a traditional mana curve and where's the Blistering Firecat!? I'll address all these issues and more in the following sections and present why I feel this is the best build for a Goblin/Sligh deck in Onslaught. I'll also address what parts of this deck are likely to change in the future.

First to address the decks name: "Red Dawn". The name is a reference to this deck being an early rising star in the long line of Red decks that will come out of Onslaught and possibly dominate the Type II environment. That and Red Dawn is the title of a classic Patrick Swayze movie.

Red Dawn is focused on speed and consistency. Red Dawn threatens a turn 4 or turn 5 kill almost every game. What's really nice is that Red Dawn does this even when opposed. It's also a very consistent deck drawing 2-3 lands in the first few turns and it top decks extremely well due to it's unusual mana build.

The mana curve in Red Dawn is worth examination. Instead of being based on the 20 something Mountain build (ignore the Goblin Burrows for the sake of this discussion), Red Dawn is based on 12 Mountains and 8 Fetch Lands. This has a noticeable and significant effect on the mana draws in Red Dawn. The mana curve in Red Dawn is comparatively flat after the initial hand is drawn.

So if we deal with expected draws, in the first hand you can expect to have 2 lands, this is true with either mana build. However, unlike the 20 mountain builds where in the next 8 draws (about the length of most games for Red Dawn) you'll expect to draw another 3 Mountains, in Red Dawn you'll only expect to draw 2 additional Mountains. This doesn't seem like a lot but it really is. It's a bonus action card in the early game and although not quite as good as drawing 8 cards in your initial hand it feels very similar. It's also important to understand that the longer the game goes, the greater the effects of deck thinning (assuming you continue to break additionally drawn fetch lands) compared to the 20 land build. So in the next 15 cards drawn you can expect to gain almost 2 action cards over the 20 land build, and you'll gain almost 5 cards worth of action cards in the final 30 cards.

From testing Red Dawn the advantages to using Fetch Lands for deck thinning are more significant than those presented mathematically. This is because everyone cheats at shuffling in magic - I have never seen a player who can comply with Section 21 of the DCI Floor Rules where it states "Regardless of the method used to shuffle, players' decks must be sufficiently randomized." and " By presenting their decks to their opponents, players are stating that their decks are sufficiently randomized." I would agree that many if not most players comply with the other clause of " a reasonable effort to sufficiently randomize his or her deck." The problem is even with the best of intentions, I'm not sure that it's possible for the average player to sufficiently randomize their deck in the time limits provided.

<-------------Begin digression into shuffling------------->

To digress, in an attempt to forestall the mail I will likely receive about this - the statement that a deck is "sufficiently randomized" after 7 riffle shuffles is a myth. This comes from the work of Trefethen, L.N., and L.M. Trefethen. 2000. How many shuffles to randomize a deck of cards? Proceedings of the Royal Society, London A 456(Oct. 8):2561.

and their work was based on the work of

Bayer, D., and P. Diaconis. 1992. Trailing the dovetail shuffle to its lair. Annals of Applied Probability 2:294.

The works in question do show that a computer or mechanical randomizing device can riffle shuffle somewhere between 5-7 times and sufficiently randomize a deck. However, many popular science publications make an error of the type conclusion validity, specifically the error of taking (and reporting) as a conclusion of the work something that is a researchers opinion instead of a research finding. In this case the error is assuming that the average person can sufficiently randomize a deck in 5-7 riffle shuffles when the papers prove that a computer or a mechanical device can do so.

The evidence actually points against this being the case, as the algorithm used and results of a riffle shuffle by a computer or mechanical device is quite different from the algorithm and results of a human performing riffle shuffle. Many experts believe the number of riffle shuffles needed to sufficiently randomize for an average person is in the thousands. I wouldn't be surprised if this were true.

Of course all of this depends on the definition of sufficiently randomized. As the DCI doesn't define sufficiently randomized, I define sufficiently randomized for the purposes of DCI card games as shuffling a deck so that each card has an approximately equal probability of being in any place in the deck. And any patterns of cards or card groupings have an approximately equal probability of occurring. And the placement and pattern of the cards in the deck is unknown to the shuffler.

<-------------End digression into shuffling------------->

So assuming that human shuffling is not a sufficiently randomizing activity, the question arises is there a benefit to looking through your deck and shuffling it? I believe there is a benefit and this is where another advantages of using Fetch Lands for deck thinning occurs. This can be seen by comparing the results of draws while playing comparent to the expected results (see there was a point to that discussion about shuffling.) What happens is when shuffling, players randomize their deck favorably to themselves. As mentioned I don't believe players do this intentionally or consciously, but it occurs in things like making the decision of when to stop riffle shuffling or how many piles to pile shuffle on or similar shuffling decisions.

Anecdotally this is most evident in M.O.S.S. (either Magic Online Shuffling Syndrome or Magic Online Shuffling Sucks), when players suddenly lose this advantage and they suffer through an unusually high % of bad draws from a deck. They're actually experiencing "true" sufficient randomization for the first time, as opposed to "favorable" sufficient randomization.

So what increases the odds of "favorable" sufficient randomization? The primary factor is deck knowledge. Breaking a Fetch Land gives you deck knowledge and the shuffle that occurs based on this knowledge is likely to make the decks randomization more favorable than its previous state.

So am I saying you should play Fetch Lands because it helps you unconsciously cheat? No. I'm saying that you should play Fetch Lands because it thins your deck. A side effect of Fetch Lands is that if you're not a very precise shuffler it will be to your advantage. If you're interested in a method to prevent yourself from cheating see the discussion on shuffling (the method I use for shuffling) from my previous article on [Salmon Snack]

Now this deck thinning does come at a price. The obvious cost is the cost in life to activate the Fetch Lands. In a hyper-aggressive deck like Red Dawn this cost is insignificant. As with Sligh decks it's not your life that's relevant, only the other opponent's life total. This even holds in the mirror. In the normal case where you might break 2 Fetch Lands at a cost of 2 life, you gain a card which will be either a Goblin that requires the opponent to Shock (netting you the life back again) or a Burn spell that will deal the 2 points of damage to the opponent (or kill one of their dorks) also netting you back the 2 life.

The more important and less obvious cost is the negative aspect of the flattened mana curve that the Fetch Lands provide. This makes the 3cc slot and 4cc slot very uncertain as you can't be confident of drawing the 3rd and 4th land in the very early game. That's why Red Dawn has no 3cc slotted creatures. However, playing mostly 1cc and 2cc spells plays into the strength of the Onslaught Goblins. Both the Pile Driver and Pyromancer gain significant benefits from large fast Goblin waves. The Goblin Sledder also enjoys lots of Goblins to "throw around."

So let's take a look at what's in Red Dawn. Goblin Sledder and Goblin Taskmaster are obvious choices and gems in their own right. So is the Goblin Pile Driver and Goblin Raider. The Grim Lavamancer is often mistakenly snubbed in these decks because he's "not a Goblin". That's a mistake and a big one. He competes with the Mogg Fanatic and Jackal Pup for the title of "best 1 drop ever". In a deck with Fetch Lands I think he might clinch the title. The last Goblin is the Sparksmith. The Sparksmith is a control creature but he owns the ground for a mere 2cc. He handles the opponents Mongrels and Brushhoppers (don't forget if the Brushhopper "hops" out of the way you don't take the damage) usually by himself but certainly with a minor amount of burn for assistance. Considering that Red Dawn will underrun any non-creature deck in record time, the Sparksmith is possibly the MVG.

The last creature is the Pyromancer. I just finished mentioning how the flattened mana curve makes it more difficult to get to your 4th mana, so what is Red Dawn doing with 3 4cc guys? It's because in Red Dawn, the Pyromancer is a 4cc Overrun and wins the game against anything but the mirror. You hate to draw 2 of him in a single hand and so I've strongly considered reducing the number to 2 in the deck. But if you have 4 mana (which you will about turn 6) he's the Goblin for the job. Don't forget that if he dies (like being sac'd by a Sledder) before the end of the turn, he's not around to cause the "Destroy all Goblins" effect. In the mirror match he can be used as a Tividar's Crusade.

In the burn department, Reckless Charge is obvious. Firebolt vs. Shock is a more difficult decision and I'm not really sure which is better. I have Firebolt in the deck right now but I've yet to ever cast it via Flashback. But then again, I've never been sitting on mana in Red Dawn where the Instant speed of a Shock would be relevant. The last burn Spell is Lava Dart. The ability to divide the damage to multiple targets has proven to be as good in Red Dawn as Fire has been in the past. The 0cc Dart sitting in the graveyard provides a really strong incentive to the opponent to not cast their Firecat unmorphed.

The last spell is Final Fortune. It's so good that I'm tempted to add another (in place of a Pyromancer as it has a similar effect.) But much like the Pyromancer it's a card you really don't want to see in multiples. Don't forget it's an Instant, so if you're uncertain whether to cast the Final Fortune the correct play is often to wait. If the opponent taps out to do something at the end of your turn, they'll often leave themselves in a helpless position when you Final Fortune after/in response to their end of turn effect.

The only other card that might make it into the deck is 1 more Goblin Burrows. It's not that great of a card in Red Dawn due to it's high activation cost but it's hard to argue against 1 and maybe 2 of them. Especially considering that all the 2cc creature drops have a colorless in the casting cost.

There are a few notable cards that aren't in Red Dawn. The first is Blistering Firecat. He's a good card, no question. But Red Dawn doesn't have much room for 4cc spells and you're not likely to cast them before turn 6. On turn 6, he's not as good as the Pyromancer as a finisher and the damage he will deal is much more vulnerable to disruption.

Other notable Goblins that are missing include the Raging Goblin and Skirik Prospector. Both of these cards suffer from a lack of quality when compared to the other 1 drops and there's no room (or really even need) for further 1 drops.

Lastly the sideboard. It's untested as there's no metagame to test it/optimize it against. With that said here's why the spells listed are in the board. Fledgling Dragon is board for the mirror. In that fight (again especially with the Fetch Lands) you get threshold rather quickly. The winner of the fight is the person with the last creature. Fledgling Dragin willl be the last creature they see. The Meekstone seems a good if possibly irrelevant sideboard choice against big green things. It doesn't keep any of your guys down but you don't want to leave them with Walls either - Flash of Defiance might be better. The land destruction sideboard is to replace the burn against creatureless decks (like Wake or Confinement.)

Enjoy the deck,

Jay Schneider

Questions? Comments? Feedback? Please check the FAQ to contact me.