Need input/opinions on best 1.17 FG grinder

[zhen_rogue]

Have tried using ivalmain's deck for 40 matches now, results are:
16 = game quit/loss due to no quantum towers in initial draw, and no towers/supernovas by draw 2
18 = loss
6 = wins
15% Win Rate

The "bad draw" component seems really high on this deck for some reason, i'm wondering if simply adding a quantum tower, or maybe changing the supernovas to novas could make a noticeable difference.
Granted, 40 isn't a very big number to base final conclusions on, i'll keep playing with it a bit and see if my luck was simply bad.

[elfreth]

I'm getting the same bad initial draw.  Just this morning: 1 win vs 6 losses.  Most were just no chance to play either no quanta, no entropy to play the several supernovas in my hand, no shields, or no hourglasses.

I do get the deck idea, it just doesn't draw fast unless you draw an hourglass, and that's certainly not guaranteed.

[ivalmian]

Have tried using ivalmain's deck for 40 matches now, results are:
16 = game quit/loss due to no quantum towers in initial draw, and no towers/supernovas by draw 2
18 = loss
6 = wins
15% Win Rate

The "bad draw" component seems really high on this deck for some reason, i'm wondering if simply adding a quantum tower, or maybe changing the supernovas to novas could make a noticeable difference.
Granted, 40 isn't a very big number to base final conclusions on, i'll keep playing with it a bit and see if my luck was simply bad.

Don't quit if you don't get quanta the first or second turn..
seriously... don't, you have at least a 50% chance to win still..

Also, with three hourglasses, the chance of getting an hourglass in the first 8 cards (initial 7 + first draw) is statistically 60%. You can calculate it as (1 - (32!*27!)/(24!*35!)) * 100%....

Chance of getting a shield is even higher (since there are 4 of them instead of 5)... chance of getting a qt are higher then that of scaregirl's or dragoon's decks and those were kings of fg before sundial nerf...

[elfreth]

I appreciate the formulas and %'s to get the critical cards.  I even agree w/ the math.  The problem I'm finding is that it's not happening in practice.  Such is the nature of statistics, you can just have a "bad run", well...my "bad run" has been going for 50 games or so....still haven't won any more this morning. 

In fact, just lost to Elidnis, which last night I had a 100% win rate against.  Nothing to do when you don't see an hourglass for the first 15 cards and no creature in the first 18 cards, and only one shield...crazy draws huh!

[ivalmian]

right... it's just I don't understand how you can have runs whose probabilities are on the order one in a billion..

For example, I don't think zhen_rogue was using my deck card to card or that he waited for  2 turns... since the probability of not getting a qt in the first 2 turns in 16 of 40 games is
ONE IN A QUADRILLION (this is not a joke, calcs are below)!!! No one, ever, is this unlucky.... I mean.. you're more likely to be killed by a meteor... Also, loosing 18 of 24 games where you have pillars in starting hand is also obscenely high.. are you not running my deck upped? I mean, it's either that or you're seriously not using the deck correctly... Also, probability of getting a shield in the first two turns is 71%... which is higher than most other fg grinding decks...

So, I don't know what you guys are doing...but I get fairly high win percentage and it's not because I am exceptionally lucky..

Anyways, here are some calculations on how this deck behaves... this not a matter of opinion, this is math.. I am attaching a screen shot from Wolfram Mathematica

*note in the image it says prob of not getting towers in 18 of 40 instead of 16 of 40... but the calcs are don't for 16 of 40... also, prob of not drawing in 18 of 40 is one in like 10^19




Also.. I guess I'll start recording my games against which FG I played and win/lose/elemental so you guys can have something to compare to...

[Delreich]

Also, with three hourglasses, the chance of getting an hourglass in the first 8 cards (initial 7 + first draw) is statistically 60%. You can calculate it as (1 - (32!*27!)/(24!*35!)) * 100%....

55% for first eight cards, 60% for first nine. For more advanced calculations, or for those less maths interested: http://stattrek.com/Tables/Hypergeometric.aspx (http://stattrek.com/Tables/Hypergeometric.aspx)
Nitpicking aside, that's still a 17% chance of not getting a single hourglass in the first 15 cards, meaning one in six games you'll pretty much be screwed.

[elfreth]

Well, here's today's testing of this deck so far (yes I built it all upgraded to begin with):

Code: [Select]

FG | Win | Loss |
Scorpio     |    0  |     3     |
Chaos Lord  |    1  |     3     |
Morte       |    0  |     1     |
Gemini      |    0  |     4     |
Hermes      |    1  |     1     |
Fire Queen  |    1  |     1     |
Elidnis     |    0  |     1     |
Rainbow     |    1  |     3     |
Obliterator |    0  |     4     |
Graviton    |    0  |     1     |
Miracle     |    2  |     4     |
Incarnate   |    0  |     1     |
Destiny   |    2  |     0     |
---------------------------------
Totals: 8       2729.6%

How much of that is very poor opening draws or very good for the FG...hard to tell.  Stopped trying to keep track of the horrible starts early on.  I love the idea of this deck, and one good thing about it, it's finally gotten me over worrying about my win/loss %.  Testing this  has blown it all to hell heh heh

[zhen_rogue]

Believe me, I agree with the mathematics behind your probability and this deck - however, the end result is the same... I have a nasty string of bad draws/luck.
And yes, the cards are upgraded and exactly the same as what you posted, card for card.

My only speculation has to do with the random seed in shuffling being not-so-random.
There was a thread before that dealt with top-50 opponents becoming more like "top 5", and the fix was exiting/clearing cache/reloading Elements to garner a new random seed and fix the stagnant results.
Perhaps this same random seed issue also applies to deck shuffling, and for true results one should clear their cache and reload Elements every 5 games, every other game, or maybe every game.
Beyond this, i'm not sure what could cause the string of 'bad draws', when other people have nothing but 'good draws'.

Base point: I don't think pure probability equations apply to this game in its current state, there's another factor (or factors) that skew/narrow the random component.

[yendorii]

Base point: I don't think pure probability equasions[sic] apply to this game in its current state, there's another factor (or factors) that skew/narrow the random component.

Glad I'm not the only one that thinks this then.

[elfreth]

Base point: I don't think pure probability equasions[sic] apply to this game in its current state, there's another factor (or factors) that skew/narrow the random component.

Glad I'm not the only one that thinks this then.

Put me in this camp too...thanks to zhen_rogue for stating it so more eloquently.

[pervepic]

I think that Ivalmian's deck is close to perfect. If lucky, it beats also Divine Glory and Rainbow. But I would make few additions there: add 2 supernovas, 1 quantum tower and 6(!) sundials. That does not make this deck slower, but even quicker and more defencive.

[elfreth]

Upped sundials or non-upped?