Computer-Aided Speedbow Optimization

[Gorthos]

I wrote a script that simulates a game of a grabbix-style rush deck.  It just plays the cards appropriately (with some special logic for what to cast Immo on) and counts how many turns it takes to deal out 100 damage, if the opponent does nothing.  The cards I used in the simulation are unupped.  I can run the simulation 10000 times in a few seconds to see precisely how fast on average the deck works, with a fairly low margin of error (accurate to about 0.05 turns).

The next step was to randomly swap out cards and test again if the resulting deck is any faster than the one before.  If so, we keep the change and try again.  A few hundred random card swaps + autotests later and we are left with the very fastest possible unupped speedbow.

Hover over cards for details, click for permalink

Deck import code : [Select]

4vh 4vh 4vj 4vj 4vj 4vj 55u 590 590 590 5c1 5f2 5f9 5f9 5f9 5f9 5f9 5f9 5fa 5fa 5fa 5fc 5fc 5l9 5l9 5l9 5l9 5l9 5l9 5ur 8pm

According to my simulation, if the opponent does not respond, this deck can pull off a 5 turn win almost 50% of the time.  Average TTW in the simulation is about 5.6 turns.  In practice, 6 or 7 turns is typical, 5 occasional.

Note that this deck has no deflagrations.  2 deflagrations make it a lot more viable as a real deck.  I usually swap out one lycanthrope and the crimson dragon for the deflags.

I used a variant of this strategy to craft my tourney deck which I used to win the eastern tournament (the first tournament I entered!) on Sunday.  Basically just did the random swapping, with the condition that the deck must always have 6 Crimson Dragons.

[Pineapple]

I find the 2 lycans AND the vampire stiletto kinda...weird.
Have you tried the simulation with hammer, another graboid, or

[xdude]

I find the 2 lycans AND the vampire stiletto kinda...weird.
Have you tried the simulation with hammer, another graboid, or

or Arsenic.

[Gorthos]

In the simulation I pretended that vampire stiletto does 6 damage (to account for the 4 points of healing).  I guess the fact that you can get it out first turn very often counts for a lot.  Arsenic is apparently just too slow, since it would take many turns before the poison overtakes the greater direct damage for vampire stiletto.

[Pineapple]

so what do the speed results come out as if you count its damage as 4?

[Gorthos]

I don't know, Patch.  The script is not currently in a functional state, I was rewriting some things.  I doubt it would make much of a difference if I changed it to 4 - it might swap out the vampire stiletto, but swapping a single card is like 0.1 TTW difference at most.

I'm pretty sure the number of graboids is good.  If you get lava golems you'll be spending all your earth quanta growing them instead of casting graboids.

As to why it used 2 lycans... dunno.  Not having enough dark quanta is indeed a problem I've noticed often.  But the simulation says it's faster, so maybe lycans are just that good.  By coincidence they do come in handy if the enemy has maxwell's demon.

The achilles heel of this deck, like other speedbows, is early creature control, especially reverse time or scarabs/otyugh.  I've also been simply outdamaged by a mono-green adrenaline-healing rush in pvp, even with a good starting hand.

[Pineapple]

the 2 lycans is because it treats them as easy immo fodder.
and yes, 2 damage per turn is indeed a big difference..

[Gorthos]

Actually it only uses photons and phoenixes as immo fodder, in that order of preference (and only if it needs quanta).  6/6 after a turn to grow is just a good creature, I guess.

[ShiningSword]

Wow, this would be so helpful to deckbuilding when perfected. After all, nobody will ever get such a big testing sample as 10000 with real testing(because nobody will never play so many games to test a deck).

[deadlyd1001]

Wow, this would be so helpful to deckbuilding when perfected. After all, nobody will ever get such a big testing sample as 10000 with real testing(because nobody will never play so many games to test a deck).

the amount of time to get to 10,000 game with normal testing is insane

assuming against low level AI with a super fast deck (one minute per game)
10,000 games, 24 hours in a day, 60 minutes in a hour

you would have to play non-stop, no breaks,and no sleep for
6.95 days!
i can see how this can help with deckbuilding

[TStar]

2 Lycans means you have a greater chance of drawing once quicker and getting more damage out.  That's really the biggest reason for having 2, hoping to get that first one out in the first 4 turns.

[TheCrazyMango]

we can haz deck code? for those too lazy to make it manually coughmecough