Hitting em with all you got

[kamman13]

I've messed around with my share of quickburn allyless mage decks. That is, decks that rely solely on DD spells for damage. The idea behind these decks is that you can retain enough board control and do enough damage with your spells to finish off your opponent before he/she finishes off you. The goal of this deck is to maximize the damage you do per resource spent (actually per turn, but if you use all of your resources each turn (which you should be doing), this equates to damage per resource). So the question for this sort of deck is, what cards maximize damage per resource?

I've gone into these decks assuming that the DD spell cards are the best way to maximize damage per resource. Putting an ally in play is an unreliable source of damage, since if you do not have board control the ally might not even do damage. But I've recently realized portal changes this picture. Portal allows allies to attack the turn they enter play, which allows them to do damage even if they die the next turn to your opponent's board control. The consequence is that allies coupled with portal can be as effective damage sources per resource as DD spell cards.

What is the resource cost of portal per damage done? This is tricky, because although portal costs 4 to cast, you bring in multiple allies per portal cast. Low cost allies make portal more cost effective, high cost allies less so. This is because the cost of portal is effectively split over all the allies you are able to bring into play while portal is active. And you can simple bring more low cost allies into play. So a portal on turn 4-5, gives you 5-7 resources over the next two turns to use for allies. Over two turns, let's say you have 12 resources on average. To figure out how much portal adds to each creature's casting cost (cc), take cc/12 * 4. This is the number of creatures of that casting cost that could be added into play while portal was active (cc/4) times the cost of portal. So if you use portal to bring two 6 cc creatures, portal effectively added 2 cc to the cost of each creature. Here's the breakdown by casting cost:

1cc- extra 1/3 of a resource
2cc- extra 2/3 of a resource
3cc- extra 1 resource
4cc- extra 4/3 of a resource
5cc- extra 5/3 of a resource
6cc- extra 2 resources

So, now let's break down the damage/cost of the spells or allies, to figure out which cards are truly the best cards to put in a deck from a solely damage perspective. The number after each card is the damage/resource.

Let's start with the mage type DD spells. I'm going to count the damage that can be inflicted against the hero, not the total damage of the spell.

Fireball - 1.3
Nova - 1
Lightning strike - 0.75
Special delivery (human only) - 0.75
Arcane burst - 0.5

Now let's move to human allies. I'm already adding in the extra resource cost due to portal for these cards.

Blake - 1.13
Jasmine - 0.75
Kris - 0.75
Puwen - 0.75
Aldon - 0.75 *
Erika - 0.6 *
Earthen - 0.6
Lily - 0.56
Dirk - 0.5
Raven - 0.45
Sandra - 0.38
Marshland - 0.38
Tainted - 0.38
Aeon - 0.38
Sandworm - 0.3
birgitte - 0

* These cards are likely undervalued by this measure. Aldon also adds value to other cards in play. In this case, there's probably only one other card in play, which would give him a damage ratio of 1. Erika is more difficult to kill, and so may actually persist for several turns. If you keep her in play for two turns, for example, her damage ratio bumps to 1.2.

Now the shadow allies:

Minotaur - 0.9
Thaddeus - 0.75
Belladona - 0.75
Chimera w/ pump - 0.75
Medusa - 0.75
Bad wolf - 0.75
Snake - 0.75
Helsteed - 0.75
Carnivoar - 0.75
Flayer - 0.75
Shadow Knight - 0.75
Plasma w/ bolt - 0.68
Keldor - 0.6
Deathbane - 0.5
Gargoyle - 0.5
Plasma - 0.45



Also of interest:
Extra Sharp - 0.75

Even though extra sharp is a 1 for 1 ratio, if played during a portal it is using up the resources that might otherwise be spent on a 2/2 creature, thus giving the 0.75 ratio.

Bloodlust - 1, with 2 allies in play, 1.5 with 3.

By a similar logic, if bloodlust is cast when portal is active, it clogs up resources and has a higher cost. It only becomes cost effective as a boost to low cost allies.



Some additional notes:
Kris is probably a better card than we are giving it credit for. This is because he feels the effect of portal whether or not portal is in play. So he's actually a 1 for 1 card that can be played for damage before you get around to portal.

While some low damage ratio cards have high survivability, we are not interested in board control in these games, so their effect in sticking around is irrelevant. Often, they would have to stick around for 3 or 4 turns to be cost effective, which is plenty of time for an opponent to kill them before they become cost effective. Plus, these games are meant to be rapid fire, and a 3-4 turn effect is too slow.

Poison and flames only become cost effective if in play for 4 turns. Since they can't be played until turn 4 and games are usually over by turn 8-9, they just border on being effective. but since only one can be used at a time, they can't be spammed like fireball or lightning, so IMO aren't worth including in this sort of deck.

Casting portal on a later turn increases the cost efficiency of the allies relative to the spells, since you can get more allies in for a given portal cost. This maxes at AT/cc for an ally, which hovers around 1 for most allies. But note that you probably won't be playing more than 1 portal per game, maybe two, since this is all for a quick "race to the bottom" type of game.

Allies also give an added bonus, not taken into account in an allyless deck calculation. Allies will do repeated damage to your hero, unless they specifically attack and kill your allies. This in turn prevents enemy allies from attacking your hero, shielding you from damage you might end up absorbing if you ran an allyless deck. If your opponent ignores your allies and attacks your hero, you are almost guaranteed to beat your opponent, as you are maximizing damage per resource, and so should be doing more damage to them than they to you. Of course, in an allyless deck you'd use cards like retreat and rain delay to prevent this damage, followed by effective board wipes. It is difficult to assess the relative advantage of using allies as "delays" rather than retreats or rain delays.

[kamman13]

Deck selection:

As you can see, the shadow deck has a lot more offensive allies than the human deck (no surprise really). The humans, though, contain the most cost effective ally- blake. Since you'll be stocking your deck with only 4-5 different types of allies, the shadow's aggressive advantage really doesn't amount for much, since there are easily 4-5 aggressive human allies to choose. There are really only 6 in my opinion (the top 6 on the list), so your choices are somewhat limited compared to the top 11 to choose from on the shadow list (I'm considering anything lower than 0.75 inefficient, since then you'd opt for lightning and special delivery).

What should be included in every deck:
4 fireballs
2-3 novas
4 LS
4 bazaars (you can probably get away without bazaar in a Majiya or GB deck, by just using their natural draw and 4 bellas, though this will change with the new hero abilities in 1.28)
3-4 portals

Fill in the rest of the deck with cost-efficient allies of your choice.

Fireballs are key here, they are currently the most cost effective damage source in the game. NEVER SACRIFICE A FIREBALL. when I play allyless decks, I tend to save up 2-3 fireballs in my hand. This may give you opponent a false sense of security, as any point after turn 6 you can hit them with an easy 8 damage.

The cards you should feel safe sacrificing are your extra bazaars, extra portals, Lightning strikes, or some of your less cost-efficient allies.

Playing strategy:

So in this sort of game, what is the correct way to play an ally damage burn deck? This depends a bit on your hero and your opponent and your draw. If you are playing a human deck, I would spend the first four turns spamming kris and puwen to retain board control, and then at turn 4-5 leading off with portal. The next couple turn are spent relentlessly hammering the hero with allies, as your opponent probably wastes his time killing your allies. When portal dies, you clean up the mess with a nova, eliminating your opponents board control.

Another nice strategy is to build up a favorable ally list in your hand, before laying it all down after a turn 4-5 portal. There's two ways to play this, depending on when you play nova.

Style 1:
T1 Sac
T2 Bazaar
T3 Fireball
T4 Portal
T5 allies
T6 allies
T7 nova + low cost ally
T8 Finish with FB/LS, or cast another portal if they have too much life left

Style 2:
T1 Sac
T2 Bazaar
T3 Fireball
T4 LS
T5 Nova
T6 Portal
T7 Allies
T8 Allies
T9 finish with FB/LS

Which style you play depends in part on the board situation by turn 4. If your opponent has many allies in play by turn 4, you may want to wait until your turn 5 nova before casting portal. Otherwise, your opponent may have enough strength to easily knock down your allies and do some damage to you each turn. You want him to have to work at killing your allies, wasting as many resources as he's got to kill those allies. If your opponents deck doesn't seem too ally heavy, go right into portal at turn 4.

Another consideration is if you have the cards to most effectively use your resource pool on turn 4. This means having a 5-cost ally, or a 2 cost and 3 cost ally. If you don't, might be best to use the more cost effective damage of a Lightning and Nova, waiting till you get more allies for a turn 6-8 portal.

Special notes- for Nish, you may not have to wait for nova on turn 5. You can thrown down a portal on turn 4, use his ability at the beginning of your turn 5 to wipe the board, then start spawning allies.

Why should the game be over by turn 8-9?

Well, if you used your resources correctly, you should have done more than enough damage to your opponent to win. By turn 3, you are supposedly spending every resource you've got to do damage. Resources convert to damage at let's say a ratio of 0.75 (true for most allies and LS, but doesn't take into account FB or the nova, so a conservative estimate). The amount of resources you have to spend is 3+4+5+6+7+8, or 33 resources by turn 8. This converts to 24.75 damage, nearly enough to kill most heroes, definitely enough if you consider FB and Nova to add a more effective damage to that list. By turn 9, you've spent 42 resources, equaling 31.5. This takes down everything but a few healing heroes, but again your FB and Nova should take care of even them.

[kentuequi]

Wow, that's a wall... I didn't took the time to read all you have write, but for Erika who would be difficult to kill, i disagree, because when she come on the board she is killed as soon as you end your turn. MC, lightnin strike, fireball, zaladar ability, crippling blow, plasma blast, shield bash, sentinel ability, special delivery, captured prey, rabid bite, smite, are always waiting a good card to be effective.

[He-Man]

I like how you are approaching this through a thorough analysis.

My main difficulty however with this, is its applicability. A major issue is that the math only works if you actually have all those cards in your hand--so to realistically achieve the numbers for 1cc allies for instance, you will need a hand that consists of a portal on turn 4, and 4 1-cost allies on turn 5, which is not very realistic to happen.

The damage/cost values (assuming a single shot with each card) are somewhat more relevant to the live situation, imo, but those only apply with a portal out and/or assuming the opponent has nothing on the board. (If he takes out your new ally right away, the cost effectiveness drops to zero to this regard.)

I would say, build the deck that would be dictated by your model as giving you the best damage yield, and let us know how it does.

Note also that using an ally as a retreat or rain delay does not work, as with the latter two you are sure of your targets and with the former you are not. (If you are almost dead, the opponent will not bother to attack you newly played ally for instance.)

All in all, thank you for sharing some useful information with the community.

[kentuequi]

After reading all,

Priest will destroy your bazaar and/or portal, and a priest with board control + KP = you lose
Warriors have enrage and you will hurt yourself with super nova helping them to kill your hero T8/T9 = you lose
Hunters will heal with your cheap allies, bazaar will give them all they need, you don't run a snow saphire = you lose
Shadows will destroy your portal.
Also, while you drop a super nova and hit your hero for 5 damages you'll give healing to elementals (soul reaper)

[Gondorian]

I think you've set a great example for others to follow here. It's certainly well thought through, and the logic and theory is sound on paper/iPad.

But, as others have said, it may not work in practise, so you will need to tweak your way to a high performance deck, and work out which bonus ally skills pay off for you beyond the attack values.

Also remember: each ally can soak up damage for you, so you can confidently supernova without dying. Maybe factor the health values into your calculations too?

[kamman13]

Kentequi,

You raise some good points, but I don't think they're game changers to the strategy, only weaknesses (which every strategy has). The only way to know I guess is to test and tell. I'm going to update here my wins and losses with this strategy, along with the opponent, who went first, the turn the game ended on, and the winner's remaining life (an important concern with a race to the bottom strategy). This is a thought exercise for me as much as for anyone else, so I'm curious how it will work.

LOSS vs. Jericho, 2nd, 8th turn, 2 life
WIN vs. Jericho, 2nd, 8th turn, 21 life
LOSS vs. Logan, 1st, 10th turn, 13 life
WIN vs. Boris, 2nd, discon but I was well ahead to win by turn 8
LOSS vs. Gravebone, 1st, 8th turn, 5 life
LOSS vs. Boris, 2nd, 8th turn, 12 life


This is with a 40 card Nish deck. I found card draw is more important than I expected it to be. I won 2 of the 3 games that I drew a bazaar by turn 2, and lost every one I didn't. Adding bad santa didn't help much- it was hard to find the time to put resources towards it late game, and it wasn't useful early game. Going to try a 30 card deck tomorrow.

Anyone else want to try it too?

[kamman13]

New deck, 34 card nish deck- 4 bazaars and 2 bad santas.

WIN vs. Majiya, 2nd, 8th turn, 14 life
LOSS vs. Majiya, 1st, 6th turn, 9 life (she ignored my blakes for the win, I had 10 life)
WIN vs. Boris, 1st, 7th turn, 22 life
WIN vs. Nish, 1st, 8th turn, 20 life
LOSS vs. Nish, 2nd, 9th turn, 5 life
LOSS vs. Majiya, 2nd, 8th turn, 7 life
LOSS vs. Gwen, 1st, 11th turn, 3 life

Well, clearly not a winning deck, at least in Nish's hands. A lot of the losses were one turn away from a victory, but obviously that's still not good enough.

[Asteredo]

since card drawing is essential, will it be better if the hero is mijaya?

[kamman13]

Ha, funny you asked, I figured no one was reading this thread anymore. I tried running Majiya, but my results were pretty similar. I actually like the human allies a lot more for this strategy, they're strong, cheaper and have more complimentary abilities, and Nish's ability is a nice bonus- board clear without the health loss. So the card draw was balanced by a few more negatives.

I run a Majiya deck that stacks both offensive and defensive allies, allowing to switch between a controlling strategy or death race type strategy described here. It's not as effective at death-racing as a deck stacked specifically for offense as described here, but it's versatility more than compensates for this fact. I don't want to post this deck just yet, but feel free to shoot me a PM if you want some details.