Testing Stall cards

[ratcharmer]

Stall rating is a measure of how effective a given card or card combo is at slowing your opponent down.

Stall rating is expressed as a number, usually between 1 and 2. The higher the number the more a given card will slow your opponent down. This number is the ratio of the opponent's ttw versus a deck using the card in question (and only that card) and the opponent's ttw versus a deck of all pillars.

Thus a stall rating of 2 means that the card in question doubles the amount of turns the opponent takes to kill you, whereas a stall rating of 1 means the card in question does not slow down your opponent at all

Here's a spoiler for how to test stall rating on cards:
Okay, so thanks to Jmdt's ttw studies (link: http://elementscommunity.org/forum/index.php/topic,17903.0.html) we have a pretty good way to measure how fast different strategies are, but what about stall/control cards or strategies?

Unfortunately there is no method to quantify the effectiveness for different stalling strategies that's quite as clear cut as turns to win, but we can get close.

How it works
The trainer (link: http://www.elementsthegame.com/trainer) has a mode in which you can control not only your own deck, but the deck that the AI is running against you as well. We can use this to isolate the effect of just one card (or card combo) on the ttw of a given deck.

The first step is to pick the deck you'll be using to test different cards against. In many cases it will be important test a given card against multiple deck types (consider the effect black hole has on a rainbow versus it's effect on a mono deck).

For a my demonstration I'll be using USEM:

Code: [Select]

6rk 6rk 7ac 7ac 7ac 7ac 7ac 7ae 7ae 7ae 7ae 7ae 7ae 7ag 7ag 7ag 7ag 7ag 7ag 7ai 7ai 7ai 7ai 7ai 7ai 7bu 7bu 7bu 7bu 7bu
Once you've picked your deck the next step is to see what the ttw is if no one is trying to stall you. Using the trainer run your deck against a deck consisting of ALL TOWERS/PENDULUMS. Since most decks are very consistent when running unopposed you'll probably only need ~20 repeats to get an accurate unopposed ttw. If the deck you picked is not reliable when running unopposed then you should probably use a different deck, as it's hard to get accurate numbers if the deck isn't reliable.

Here's the codes form all tower/pendulum decks if you dan't want to look them up:
Aether
808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 808 81q 81q 81q 81q 81q 81q 8pu

Air
7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7oe 7oe 7oe 7oe 7oe 7oe 8pr

Darkness
7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7um 7um 7um 7um 7um 7um 8pt

Death
710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 710 72i 72i 72i 72i 72i 72i 8pk

Earth
778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 78q 78q 78q 78q 78q 78q 8pm

Entropy
6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ve 6ve 6ve 6ve 6ve 6ve 8pj

Fire
7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7dg 7f2 7f2 7f2 7f2 7f2 7f2 8po

Gravity
744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 744 75m 75m 75m 75m 75m 75m 8pl

Life
7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7bu 7bu 7bu 7bu 7bu 7bu 7bu 8pn

Light
7jo 7jo 7jo 7jo 7jo 7jo 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 8pq

Time
7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7ri 7ri 7ri 7ri 7ri 7ri 8ps

Water
7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7gk 7i6 7i6 7i6 7i6 7i6 7i6 8pp

After 20 games USEM gave me an average ttw of 5.9

Our next step is to give the AI a stall card and see how much the ttw changes. For our example, I removed 6 pillars/pendulums and replaced them with Discords:

by ratcharmer

Hover over cards for details, click for permalink

Deck import code : [Select]

6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6u5 6u5 6u5 6u5 6u5 6u5 8pj

After 20 games versus the above deck I had a ttw of 7.3.

While it would probably be better to test more games, I have a lot of cards to go through so I'm not going to run a huge number of games for each.

Once we have both the unopposed ttw and the ttw versus discord we can calculate the "stall rating" of discord.

(ttw versus card)/(ttw unopposed) = (stall rating)
in this case:
7.3/5.9= 1.24

What stall rating tells us
The stall rating of a given card tells you how much that card slowed down the strategy you were trying it against. A stall rating of 2 means that the card you're testing doubled the ttw of the deck you're running.

A stall rating less than 1 means using the card in question actually speeds up your opponent.

A few things to note
-This is not a perfect measurement, stall rating for a given card can vary a lot depending on what deck you're running against. Generally similar decks are vulnerable to similar sorts of stalling.

-Since the entire rest of the deck is pillars the AI will almost always have the quanta to play the card you're testing, so stall rating is largely independent of card cost. The exceptions to this rule are when testing multiple cards at once (i.e. to compare the effect of discord with black hole versus discord alone) or for very expensive cards.
Stall rating for cards costing 6 or more will be slightly lower than it should be, due to not being able to reliably play the card on the first turn (note: this does not apply to miracle since you wouldn't want to play that on your first turn anyway.)

-You CAN'T compare stall rating between cards that stack (i.e SoG) and those that don't (i.e. Ice Shield or Discord).

-Some cards simply can't be evaluated this way. Examples include Dissipation Shield, Stoneskin and Siphon Life (highly dependent on how much quanta is available).

-If the stalling aspect of a card doesn't change with an upgrade then the stall index shouldn't change either. For instance Discord should have the same stall rating both upgraded and non-upgraded. However, some cards (i.e. Vampire stilleto) should have different stall ratings when upgraded.

-Stall rating measures how well the AI can stall with a given card. Often times a player could use certain cards much more effectively.
--------------------------------------------------------------------------------------------------------------------
IF YOU WANT TO HELP:
I'd certainly love any help people can offer me. Here's what I think will work best:

-Pick a deck from Jmdt's ttw study. (link:http://elementscommunity.org/forum/index.php/topic,21865.0.html ) It's important to use a fast deck that's fairly common, hence why I'm picking from the ttw study. Don't pick USEM, since I'm already testing that one.

-Go to the trainer (link: http://www.elementsthegame.com/trainer ) and build that deck. Then click on the "Import" button (just above level 0) copy and paste one of the deck codes from my results (don't forget to add the mark back) into the box that appears.

-Play at least 20 games, recording your ttw each game. Post the ttw here, along with what deck you were using and what you were playing against. I'll add it to the results.
--------------------------------------------------------------------------------------------------------------------

The results so far
I'll update this part as I test more cards. I will move on to testing with other decks after I've finished testing with USEM.

--------------------------------------------------------------------------------------------------------------------
Testing with USEM:

Code: [Select]

6rk 6rk 7ac 7ac 7ac 7ac 7ac 7ae 7ae 7ae 7ae 7ae 7ae 7ag 7ag 7ag 7ag 7ag 7ag 7ai 7ai 7ai 7ai 7ai 7ai 7bu 7bu 7bu 7bu 7bu
Non-stacking stall cards:
Weapons:
Vs (Upgraded) Discord:
Stall Rating = 1.24

Code: [Select]

6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6u5 6u5 6u5 6u5 6u5 6u5Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.3
4- 0
5- 0
6- 6
7- 6
8- 4
9- 4
10- 0

Notes: It was surprisingly hard to EM against just discord. I frequently couldn't afford to heal if I wanted to finish my opponent that turn

Vs (Unupgraded) Druidic Staff
Stall Rating = 1.18

Code: [Select]

5c5 5c5 5c5 5c5 5c5 5c5 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7buTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 6.95
4- 0
5- 1
6- 6
7- 10
8- 2
9- 0
10- 0
11- 0
12- 1

Note: the 12 turn win was an abnormally bad draw (no creatures until turn 5) and may be affecting my numbers a little.

Vs (Unupped) Vampire Stilleto
Stall Rating = 1.17

Code: [Select]

5ur 5ur 5ur 5ur 5ur 5ur 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 6.90
4- 0
5- 1
6- 7
7- 8
8- 3
9- 0
10- 0
11- 1
12- 0

Note: the 11 turn win was an abnormally bad draw. The comparison between vamp stilleto and druid staff is kind of interesting since it shows the difference due to 1 hp/turn of healing.
Vs (Upped) Vampire Stilleto
Stall Rating = 1.19

Code: [Select]

7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7tb 7tb 7tb 7tb 7tb 7tbTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.00
4- 0
5- 0
6- 6
7- 10
8- 3
9- 0
10- 1
11- 0
12- 0

Note: The comparison between vamp stilleto and druid staff is kind of interesting since it shows the difference due to 1 hp/turn of healing.
Vs (Unupped) Owl's Eye
Stall Rating = 1.30

Code: [Select]

5ol 5ol 5ol 5ol 5ol 5ol 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7oeTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.65
4- 0
5- 1
6- 5
7- 7
8- 2
9- 2
10- 1
11- 0
12- 1
13- 1

Notes: The AI is pretty bad with this card, a human opponent could have done much better. The really long games are when I didn't draw a cockatrice.
*edit* I think the recent patch may have updated how the AI chooses it's target. This card was tested pre-1.27
Vs (Upgraded) Posiedon
Stall Rating = 1.01

Code: [Select]

778 778 778 778 778 778 778 778 778 778 7gs 7gs 7gs 7gs 7gs 7gs 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 5.95
4- 0
5- 6
6- 11
7- 1
8- 2
9- 0
10- 0

Notes: Wow . . . I feel like a bully  . Poseidon did keep me from EMing a couple times, but the effect on ttw is basically non-existent. For the first ten games I actually had better ttw against Poseidon than running unopposed. The difference in ttw due to Poseidon is literally the smallest one measurable.
This means that of the twenty games I played, Poseidon delayed me for a single turn in ONE of those games.
Since Poseidon uses two types of quanta I had to split between and , which led to a couple games in which the AI was not able to play Posiedon right away. Realistically I don't think it would have made any difference if it had been played first turn though.
Vs (Upgraded) Eternity:
Stall Rating = 1.23

Code: [Select]

7q8 7q8 7q8 7q8 7q8 7q8 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7riTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.25
4- 0
5- 3
6- 6
7- 5
8- 2
9- 2
10- 0
11- 1
12- 0
13- 0
14- 1
15- 0

Notes: Eternity doesn't seem to actually stall this deck very well, but it does seem to make a bad draw VERY hard to get out of. Note that most games were in the 5-8 turn range, with a couple of really long games.
A human opponent could have done slightly better than the AI, but not a huge difference.
I suspect this card would fare far better against a different deck (DEMA, PSNbow etc.).
Shields:
Vs (Upped) Improved Fog
Stall Rating = 1.37

Code: [Select]

7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7n0 7n0 7n0 7n0 7n0 7n0Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 8.1
4- 0
5- 0
6- 1
7- 7
8- 8
9- 0
10- 3
11- 0
12- 0
13- 1

Notes: This is the first shield I've tested so it's hard to say much about it's relative power. What I can say is that so far shields are better at stalling than weapons (you probably knew that).
Vs (Upped) Turtle Shield:
Stall Rating = 1.35

Code: [Select]

7q3 7q3 7q3 7q3 7q3 7q3 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7riTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.95
4- 0
5- 0
6- 2
7- 5
8- 8
9- 3
10- 1
11- 1
12- 0
13- 0
14- 0
15- 0

Notes: Second shield tested. Looks like Fog is in general better at stopping damage than Turtle shield.
This shield would fare better against a deck with no weapon or one that uses creature abilities.
Vs (Upped) Improved Dusk Mantle:
Stall Rating = 1.60

Code: [Select]

7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t8 7t8 7t8 7t8 7t8 7t8Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 9.45
4- 0
5- 0
6- 0
7- 1
8- 5
9- 6
10- 3
11- 3
12- 1
13- 1
14- 0
15- 0

Notes: The comparison between Fog and dusk is interesting. The additional 10% miss chance actually makes a big difference.
Vs (Unupped) Shield:
Stall Rating = 1.18

Code: [Select]

4tc 4tc 4tc 4tc 4tc 4tc 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 6.95
4- 0
5- 1
6- 6
7- 7
8- 5
9- 1
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: A 1-point damage reduction really doesn't do very much. Compare to tower shield (upped) and diamond shield (upped) to see the stall rating progression for 1, 2 and 3-point shields.
Vs (Upped) Tower Shield:
Stall Rating = 1.34

Code: [Select]

6rs 6rs 6rs 6rs 6rs 6rs 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.9
4- 0
5- 0
6- 0
7- 9
8- 8
9- 1
10- 1
11- 0
12- 1
13- 0
14- 0
15- 0

Notes: A 2-point shield. The difference between 1 and 2 points of damage reduction is fairly substantial.
Vs (Upped) Diamond Shield:
Stall Rating = 1.75

Code: [Select]

778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 77c 77c 77c 77c 77c 77cTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 10.3
4- 0
5- 0
6- 0
7- 0
8- 1
9- 3
10- 7
11- 6
12- 1
13- 1
14- 0
15- 0

Notes: A 3-point shield. The difference between 1, 2 and 3 points of damage reduction is fairly substantial.
This shield is expensive enough to play that there was one or two games wherein the AI had to delay playing the shield a turn in order to get enough quanta. This means that the stall rating of the card is actually slightly below where it should be.
3 points of damage reduction is huge.
Stacking stall cards:
Vs (Upped) Granite Skin, 6 towers:
Stall Rating = 1.26

Code: [Select]

4t6 4t6 4t6 4t6 4t6 4t6 4t7 4t7 4t7 4t7 4t7 4t7 4ta 4ta 4ta 4ta 4ta 4ta 77k 77k 77k 77k 77k 77k 78q 78q 78q 78q 78q 78qTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.45
4- 0
5- 0
6- 5
7- 6
8- 3
9- 3
10- 2
11- 0
12- 0
13- 0
14- 0
15-

Notes: Since Stoneskin is so dependent on quanta I tested this card multiple times, with varying amounts of towers in the deck.
Unupped shards were used as unplayable "filler cards"
Vs (Upped) Granite Skin, 12 towers:
Stall Rating = 1.49

Code: [Select]

4t6 4t6 4t6 4t6 4t6 4t6 4t7 4t7 4t7 4t7 4t7 4t7 778 778 778 778 778 778 77k 77k 77k 77k 77k 77k 78q 78q 78q 78q 78q 78qTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 8.8
4- 0
5- 0
6- 2
7- 5
8- 1
9- 7
10- 1
11- 1
12- 2
13- 1
14- 0
15- 0

Notes: Stoneskin with a moderate amount of quanta backing. Compare to stoneskin with 6 towers.
Unupped shards were used as filler cards.
Vs (Unupped) Purify:
Stall Rating = 1.13

Code: [Select]

5ia 5ia 5ia 5ia 5ia 5ia 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6Total games so far: 40

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 6.65
4- 0
5- 2
6- 17
7- 15
8- 5
9- 1
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Since USEM uses no poison, purify can be treated as 1 hp/turn healing. Notably it isn't all that effective on it's own.
The comparison between unupped purify, upped purify, sanctuary and SoG should provide a nice progression through regeneration amounts with stacking cards.
Vs (Upped) Purify:
Stall Rating = 1.20

Code: [Select]

7gq 7gq 7gq 7gq 7gq 7gq 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6 7i6Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.1
4- 0
5- 0
6- 8
7- 6
8- 4
9- 1
10- 0
11- 1
12- 0
13- 0
14- 0
15- 0

Notes: Since USEM uses no poison, purify can be treated as 2 hp/turn healing. Notably it isn't all that effective on it's own.
The comparison between unupped purify, upped purify, sanctuary and SoG should provide a nice progression through regeneration amounts with stacking cards.
I caught a typo in the deck code I used to test the upped card the first time. I think this explains the earlier irregularities in the data. The numbers now match much more to what is expected.
Vs (Upped) Sancuary:
Stall Rating = 1.29

Code: [Select]

7k6 7k6 7k6 7k6 7k6 7k6 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7laTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.6
4- 0
5- 1
6- 5
7- 5
8- 2
9- 4
10- 3
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: 4 hp/turn healing. Since USEM has no discord etc. the secondary effects of this card can be ignored.
It's interesting to note that with all of the healing/turn cards I tend to see a larger range in my ttw values.
This is because the longer a game runs, the more these cards will help you (each turn you're alive is another turn they heal you), so if a game starts slow a healing over time effect will drag things out, but after a fast start healing over time will have little effect since the game will not be long enough for the healing to build up.

Vs (Upped) Shard of Gratitude:
Stall Rating = 1.61

Code: [Select]

6rn 6rn 6rn 6rn 6rn 6rn 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7laTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.6
4- 0
5- 0
6- 3
7- 1
8- 5
9- 3
10- 2
11- 1
12- 1
13- 3
14- 0
15- 0
16- 1

Notes: 5 hp/turn healing. It's interesting to note that with all of the healing/turn cards I tend to see a larger range in my ttw values.
This is because the longer a game runs, the more these cards will help you (each turn you're alive is another turn they heal you), so if a game starts slow a healing over time effect will drag things out, but after a fast start healing over time will have little effect since the game will not be long enough for the healing to build up.
Technically I should have used quantum tower's to test SoG, but I wanted the best side-by-side comparison with sanctuary I could get.
Vs (Upped) Quicksand, mixed towers & pendulums:
Stall Rating = 1.08

Code: [Select]

778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 77j 77j 77j 77j 77j 77jTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 6.4
4- 0
5- 1
6- 11
7- 7
8- 1
9- 0
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Quicksand is not effective at all in an upped environment.
I also ran this card versus a modified version of USEM where I exchanged all of the pendulums for towers. It didn't make much difference.
Vs (Upped) Quicksand, Towers only:
Stall Rating = 1.04

Code: [Select]

778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 77j 77j 77j 77j 77j 77jTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 6.15
4- 0
5- 3
6- 11
7- 6
8- 0
9- 0
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Quicksand is not effective at all in an upped environment.
This time I ran quicksand versus a modified version of USEM that had only towers (to test the effectiveness of dividing up towers & pendulums versus EQ).
Oddly, Quicksand actually appears to be slightly less effective against solely towers than versus the mix. I will look into this more, but I suspect that this is merely due to the fact that quicksand simply wasn't effective against either of the two decks.
Vs (unupped) Devourer:
Stall Rating = 1.19

Code: [Select]

5um 5um 5um 5um 5um 5um 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7
4- 0
5- 3
6- 4
7- 7
8- 3
9- 2
10- 1
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: I think a lot of the quanta denial cards would fare much better against an unupgraded deck than an upgraded one.
Towers give a quanta that can be spent before your opponent can interfere, which is a big difference for many decks.
Vs (Upped) SoD:
Stall Rating = 1.31

Code: [Select]

6rm 6rm 6rm 6rm 6rm 6rm 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7laTotal games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 7.7
4- 0
5- 0
6- 1
7- 7
8- 9
9- 0
10- 2
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: SoD comes in behind SoG in terms of raw stalling power.
Vs (Unupped) Fate Egg:
Stall Rating = 1.04

Code: [Select]

5ri 5ri 5ri 5ri 5ri 5ri 6u3 6u3 6u3 6u3 6u3 6u3 6ve 6ve 6ve 6ve 6ve 6ve 6ve 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0 7q0Total games so far: 20

ttw unopposed = 5.9
4- 0
5- 5
6- 12
7- 3
8- 0
9- 0
10- 0

ttw opposed = 6.15
4- 0
5- 5
6- 8
7- 6
8- 1
9- 0
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: This card was a pain to test . . .

Things I learned from testing Fate egg:
1) It is IMPOSSIBLE to balance quanta usage with this card. No, not really hard, impossible. Even with the entire rest of the deck devoted to quanta you still run short of something in half the games.

2) In almost every game fate egg does nothing whatsoever (stall-wise).

3) Purple nymph is so OP there's no way I can even test it. I actually LOST a game when the AI had an early hatch pop out a purple nymph. Notably, that was also pretty much the only game the eggs really did anything impressive.

95.3% win rate . . . certainly wasn't expecting that with this card.
--------------------------------------------------------------------------------------------------------------------
Testing with Grabbix:

Code: [Select]

4vh 4vj 4vj 4vj 4vj 52q 590 590 590 590 590 5c1 5f9 5f9 5f9 5f9 5f9 5f9 5fc 5fc 5fc 5fc 5fc 5fc 5l9 5l9 5l9 5l9 5l9 5l9
Non-stacking Stall Cards:
Weapons:
Vs (Unupped) Druidic Staff:
Stall Rating = 1.12

Code: [Select]

5c5 5c5 5c5 5c5 5c5 5c5 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7buTotal games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 6.45
4- 0
5- 3
6- 7
7- 8
8- 2
9- 0
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Thanks to Baily18 for helping to test this one. I rounded some of the numbers to be consistent with where I've been rounding.
Druidic staff's performance against grabbix is similar to its performance against USEM.
Vs Discord:
Stall Rating = 1.14

Code: [Select]

Total games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 6.55
4- 0
5- 5
6- 6
7- 6
8- 1
9- 1
10- 0
11- 1
12- 0
13- 0
14- 0
15- 0

Notes: Tested by Bailey18
[Bailey18's notes]When you have no immos but phoenixes in hand, discord can slow you down alot. When you have immos and phoenixes for it, it can help you get that extra quantum you need to reborn that phoenix or grow that spirit, which made more 5 and 6 turn wins.[/Bailey18's notes]
Discord seems to still be at least partially effective against some rainbows.
Vs (Unupped) Owl's eye:
Stall Rating = 1.44

Code: [Select]

5ol 5ol 5ol 5ol 5ol 5ol 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7ms 7oeTotal games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 8.25
4- 0
5- 1
6- 3
7- 4
8- 7
9- 4
10- 1
11- 0
12- 1
13- 0
14- 0
15- 0

Notes: Tested by Bailey18
[Bailey18's notes]All of Grabbix's creatures are under three health(besides spirit and lycan buffed), and the reborn of phoenix prevents the summon of another phoenix.(without nova)Also, since the OE is likely to come out first turn your shriekers will not have a chance to burrow unless you evolve two at once.[/Bailey18's notes]
Vs (Upgraded) Vampire Dagger:
Stall Rating = 1.23

Code: [Select]

7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7tb 7tb 7tb 7tb 7tb 7tbTotal games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 7.05
4- 0
5- 1
6- 7
7- 4
8- 6
9- 2
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Tested by Bailey18
Again, grabbix has no shield so we can treat this as just 6 hp/turn healing. Compare to Druidic staff and unupped vampire stiletto.
Vs (Upped) Eternity:
Stall Rating = 1.92

Code: [Select]

7q8 7q8 7q8 7q8 7q8 7q8 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7riTotal games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 11.05
4- 0
5- 0
6- 0
7- 2
8- 2
9- 5
10- 4
11- 2
12- 0
13- 0
14- 1
15- 1
16- 0
17- 2
18- 0
19- 0
20- 0
21- 1

Notes: Tested by Bailey18
[Bailey18's notes]Bad hand=Eternity slaughters you. Arsenic makes for a faster win and you need to play your whole hand in one turn to win faster.[/Bailey18's notes]
It looks like eternity is generally really strong versus cremation decks.
Shields:
Vs (Unupped) Shield:
Stall Rating = 1.14

Code: [Select]

4tc 4tc 4tc 4tc 4tc 4tc 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778 778Total games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 6.55
4- 0
5- 3
6- 6
7- 8
8- 3
9- 0
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Tested by Bailey18
[Bailey18's notes][/Bailey18's notes]

Vs (Upped) Turtle Shield:
Stall Rating = 1.51

Code: [Select]

7q3 7q3 7q3 7q3 7q3 7q3 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7ri 7riTotal games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 8.7
4- 0
5- 0
6- 2
7- 0
8- 6
9- 6
10- 6
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Tested by Bailey18
[Bailey18's notes]Got several bad draws.[/Bailey18's notes]
Turtle shield seems to be stronger versus decks that rely on activated abilities (not much of a surprise).
Stacking Stall Cards:
Vs (Unupped) Devourer:
Stall Rating = 1.05

Code: [Select]

5um 5um 5um 5um 5um 5um 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4 7t4Total games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 6.05
4- 0
5- 5
6- 10
7- 4
8- 1
9- 0
10- 0
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Tested by Bailey18
[Bailey18's notes]Devourers didn't really slow down the deck at all. Maybe stop a lycan from growing, or ash from reviving once in a while.[/Bailey18's notes]
Devourers tend to be very weak against rainbows, nova and cremation both produce huge amounts of quanta that can be used before the devourers get a shot at it.
Vs () :
Stall Rating = 1.25

Code: [Select]

7k6 7k6 7k6 7k6 7k6 7k6 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7la 7laTotal games so far: 20

ttw unopposed = 5.75
4- 1
5- 7
6- 8
7- 4
8- 0
9- 0
10- 0

ttw opposed = 7.2
4- 0
5- 1
6- 4
7- 9
8- 4
9- 1
10- 0
11- 1
12- 0
13- 0
14- 0
15- 0

Notes: Tested by Baily18.
The hand/quanta protection doesn't really apply to Grabbix so mostly this is just 4 hp/turn healing.
--------------------------------------------------------------------------------------------------------------------
Testing with unupped mono death:

Code: [Select]

52g 52g 52g 52g 52g 52g 52g 52h 52h 52h 52h 52o 52o 52o 52o 52o 52o 52t 52t 52t 52t 52t 52t 542 542 542 542 542 542 542Non-Stacking Stall Cards:
Weapons:
Vs (Upped) Discord:
Stall Rating = 1.17

Code: [Select]

6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6ts 6u5 6u5 6u5 6u5 6u5 6u5Total games so far: 20

ttw unopposed = 7.45
4- 0
5- 0
6- 0
7- 12
8- 7
9- 1
10- 0

ttw opposed = 8.70
4- 0
5- 0
6- 0
7- 0
8- 7
9- 12
10- 1
11- 0
12- 0
13- 0
14- 0
15- 0

Notes: Tested by Bhlewos.
[Bhlewos' notes]Discord actually did quite a lot of damage by itself, usually I had 40 - 50 HP left at the end. Usually it was also played the first turn. Otherwise, not too much trouble.[/Bhlewos' notes]
I'm actually a little surprised discord didn't do better than it did against an unupped mono, siince this is the scenario where discord is particualrly infamous.
Vs (unupped) Druidic Staff:
Stall Rating = 1.12

Code: [Select]

5c5 5c5 5c5 5c5 5c5 5c5 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7ac 7acTotal games so far: 20

ttw unopposed = 7.45
4- 0
5- 0
6- 0
7- 12
8- 7
9- 1
10- 0

ttw opposed = 8.35
4- 0
5- 0
6- 1
7- 1
8- 10
9- 5
10- 1
11- 0
12- 1
13- 0
14- 0
15- 0

Notes: Tested by Bhlewos
[Bhlewos' notes]Actually got one EM, in the 6 turn win. Throughout the game the AI did not play the Druidic staff AT ALL, lol.
The 12-turn win was unlucky, I did not get out creatures until the 4th turn.[/Bhlewos's notes]
Druidic staff has been very consistent across most decks tested so far.
Shields:
Stacking Stall Cards:
--------------------------------------------------------------------------------------------------------------------
If anyone is interested in helping let me know. Pretty much any sort of feedback you have is welcome.

There's a lot of different things I can measure with this, so if there's anything in particular you'd like to see let me know. Here's a couple examples of what's possible:
-Run the same mono deck versus earthquake both with only pillars and mixing pillars and pendulums to see how much splitting your quanta producers helps against earthquake.

-Compare a deck using sanctuary versus the same deck using SoG when running against Discord, Devourers etc. to see how effective sanctuary is.

-Test how much quanta a give stall needs to work by replacing towers with unupped shards until the stall index starts dropping (indicating there wasn't enough quanta to use the card to full effect)

*edit* after looking at my first couple tests 10 replicates isn't nearly enough to get accurate measurements. Not sure why I thought it would be. Going for at least 20 now

*edit #2* I changed the name from stall index to stall rating to avoid confusion with all the other things called "index"

*edit #3* I tried to clarify/simplify the explanation. The full description is still there behind the spoiler.

NOTE: I had a migraine when I updated this last, so keep an eye out for typos etc.

[jmdt]

Very nice idea.  I'll have to read it more thoroughly when my brain is working.

[BluePriest]

Index is becoming the most overused word in this community... *whistles innocently*

[Hyroen]

Very interesting. I'll have to test some of your methods later. Kudos.

[ratcharmer]

Index is becoming the most overused word in this community... *whistles innocently*

Yeah, mayhaps I should've called it stall rating or something . . .

[coinich]

Index is becoming the most overused word in this community... *whistles innocently*

Index
Ratio
Rule
Guide
Factor
Rank

thesaurus.com

[Dragoon]

Nice idea.  Unfortunately I'm too busy with a few of my own ELement's related projects to help, but I'm definitely going to keep an eye on this thread.

I wonder, should you focus using pure rush decks versus EM decks (ie use a Life deck with Adrenalines vs Heals)?  Or does it even matter?  Also, should you use a set of decks that represent all the rush deck types and then average their SI together?

Here's some rush deck types I can think of off the top of my head:
Cremation
Speedbow
PSNbow
Mono
Duo

[ratcharmer]

@Hyrogen:
Glad to hear someone else wants to try this out. If you want to post any results you get here it'll make my life easier.

@Dragoon:
Yeah, eventually I should run a variety of decks to get a better overall picture of how well each stall card works in general, but it will take me quite some time to finish doing all that. For the time being I'm just trying to test all of the cards against USEM. Once I finish USEM I'll pick a different deck and start on that one.

What I have so far:
Most of the results so far aren't earthshaking, but it's interesting to see some things quantified.

-An early discord is devastating to USEM. (not a surprise as USEM is a mono deck with a fairly delicate quanta balance)

-More healing on a weapon is better, but the difference is very slight (a change in average ttw of 0.05 per point of healing). The change in stall rating due to healing per turn seems to be linear. It also seems that having any healing at all is more important than the amount healed.

[Tea is good]

so by the end of this you will be the un-disputed deck countering master of the game?

[ratcharmer]

so by the end of this you will be the un-disputed deck countering master of the game?

. . . or a raving lunatic . . .

Not looking forward to testing phase shield . . . I foresee 20+ turn games . . .

[godofdeath500]

so by the end of this you will be the un-disputed deck countering master of the game?

. . . or a raving lunatic . . .

Not looking forward to testing phase shield . . . I foresee 20+ turn games . . .

Well, since the AI knows how to chain them, it will be 6*3=18 turns of no damage. That is, if they get one out early. Either way, it still has the potential for 18 turns.

[zse]

I understand your idea on picking USEM as the no-brain beating deck, but even as simple deck as that will sometimes have some inconsistent starting hands in short tests. Since the trainer allows you to give "illegal" decks for AI, you should think using something like this for opponent:

by zse

Hover over cards for details, click for permalink

Deck import code : [Select]

7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 7ac 7ae 8pn