[OUTDATED] FG Efficiency Study - Applying statistics to all of the myths

[kirchj33]

Still need someone to volunteer to run preliminary data on Flay 'EM. 

[Jenkar]

deck	Liquid Antimatter
players	Jenkar

version	1.27		win-rate	18%
$ track ?	enabled		win-rate (n)	n.a.
games	100		Score/h	-166
win-loss-(EM)	18-82-(12)		Score/h (n)	n.a.
time (h:m:s)	03:03:41		FGei[c]* (http://elementscommunity.org/forum/index.php/topic,14626.0.html)	2162
min/game	01:50		FGei[cn]* (http://elementscommunity.org/forum/index.php/topic,14626.0.html)	n.a.

      Statmasta™4000                                                                *assumed card-spin/win: 35%

score/h	FGei(c)	Statmasta™4000	wins	losses	skips	EM/Wins
n.a.    -1016    -1271    -3600    273    -21600    -25714    0    -1328    -1516    -3100    -1469    -472    -973       -1301    -931    -639    -759    1723    725    758    1283    -1113    -19286    452    119    -1600    -1434    -3600	n.a.    -1016    -1271    -3600    2209    -21600    -25714    1722    -1328    -1516    -3100    -1469    959    -973       -1301    -931    -639    -759    8561    4800    4175    7072    -1113    -19286    3722    1732    -1600    -1434    -3600	Akebono Chaos Lord Dark Matter Decay Destiny Divine Glory Dream Catcher Elidnis Eternal Phoenix Ferox Fire Queen Gemini Graviton Hecate Hermes Incarnate Jezebel Lionheart Miracle Morte Neptune Obliterator Octane Osiris Paradox Rainbow Scorpio Seism Serket	1             1                    1                        5    3    3    2            1    1	3     1     2     1      5     5     4     3     4     6     2     7     1     4     1     1     4     2     3     3     2     1     5     1     3     4     3     1		1                                                                  4      1      3      2                        1

Autoskips : Divine glory : "So, what's the next... goodbye."
Dream Catcher : Too little damage, unless his (two) fate eggs are drawn VERY early, BOTH evolve in a Light dragon, and he doesn't draw a purple nymph. The chance is infime, especially since his hand clogs up very quickly.
Decay : NO way to make enough damage.
Osiris : Only chance is him playing Momentumed Pharaohs when you have enough quanta to LS & AM them at the same turn. Impossible in practice, due to A : Enormous Quanta, B : trebuchets.

[Onizuka]

I might as well mention this.

I'll test Flay 'EM.

[kirchj33]

I might as well mention this.

I'll test Flay 'EM.

Wow.  When I told you this earlier in chat, I meant it.... "MVP"

[QuantumT]

I think another useful thing we could do with this that wouldn't require much more work would be to find what the efficiency index would be if we included autoskips.

[kirchj33]

I think another useful thing we could do with this that wouldn't require much more work would be to find what the efficiency index would be if we included autoskips.

Yeah, that's definitely on the agenda after we get more data.  Hopefully, 500 games is a realistic goal for removing those skips.

[QuantumT]

I think another useful thing we could do with this that wouldn't require much more work would be to find what the efficiency index would be if we included autoskips.

Yeah, that's definitely on the agenda after we get more data.  Hopefully, 500 games is a realistic goal for removing those skips.

I mean we create a list of autoskips for each deck, and see how much money you would make if you skipped them everytime they came up instead of playing them.

[kirchj33]

I think another useful thing we could do with this that wouldn't require much more work would be to find what the efficiency index would be if we included autoskips.

Yeah, that's definitely on the agenda after we get more data.  Hopefully, 500 games is a realistic goal for removing those skips.

I mean we create a list of autoskips for each deck, and see how much money you would make if you skipped them everytime they came up instead of playing them.

Yeah, so I think we're on the same page, but I'll try and catch you in chat later tonight just in case we're not.

Basically, what Jangoo and I were discussing is:

1.  Find some arbitrary god-by-god FGei number to eliminate data for each deck, say... for the sake of discussion 1000 FGei or less.
2.  Remove this data and treat it as an autoloss instead, with some number representing the time it takes to identify the god and quit while playing (5 or 6 seconds?)
3.  Recalculate the deck's overall FGei with "skips", possibly renaming this value as FGei(s)
4.  For each deck, post suggested skips based on data (500 games would be ideal for the reason that perhaps someone has had poor luck with too few data)

[kirchj33]

deck	Shak'ars Revenge
players	kirchj33

version	1.27		win-rate	29.09%
$ track ?	enabled		win-rate (n)	28.85%
games	220		Score/h	7
win-loss-(EM)	64-156-(0)		Score/h (n)	0
time (h:m:s)	07:45:09		FGei[c]* (http://elementscommunity.org/forum/index.php/topic,14626.0.html)	3111
min/game	02:07		FGei[cn]* (http://elementscommunity.org/forum/index.php/topic,14626.0.html)	3058

      Statmasta™4000                                                                *assumed card-spin/win: 35%      
      

"

score/h	FGei(c)	Statmasta™4000	wins	losses	skips	EM/Wins
727    -785    -808    -20    -1032    609    297    556    -673    -1263    -253    -876    -653    141       634    -432    911    138    -125    -267    1082    300    151    -747    426    895    -630    -1288    1147	6673    -785    -808    3025    -1032    5736    3756    5963    592    -1263    1859    -876    401    2447       6488    1059    6266    3108    2811    2641    7216    4322    4216    -747    5074    6971    1228    -1288    8585	Akebono Chaos Lord Dark Matter Decay Destiny Divine Glory Dream Catcher Elidnis Eternal Phoenix Ferox Fire Queen Gemini Graviton Hecate Hermes Incarnate Jezebel Lionheart Miracle Morte Neptune Obliterator Octane Osiris Paradox Rainbow Scorpio Seism Serket	6            2          3    3    3    1        1        1    1    4    1    9    4    1    2    3    4    1        4    5    2        3	6     8     7     5     8      3     4     3     9     4     4     6     10     2     4     6     4     8     3     7     1     6     2     6     5     3     13     7     2

"      
      

Excel file: https://rapidshare.com/files/1281279664/Statmasta4000_-_Shakar_s.xls

[Onizuka]

It takes me about 3 seconds to identify the god+quit.

[Jenkar]

It takes me about 3 seconds to identify the god+quit.

About 4-5 for me, that's because my computer lags a weeny bit at the beggining of fights. Onizuka's time is probably right.

[Jangoo]

Just to be sure:

Please try to skip as little as possible for now!


Like kirch already explained:

Most of those unsustainable matches will be "artificially" purged from the data later.