[OUTDATED] FG Efficiency Study - Applying statistics to all of the myths

[Jangoo]

Unholy mother of God. That FGei with that winrate.. *drools*

Hell yeah ...

I am almost done with retuning the Statmasta according to the cardspin-rates as calculated by DarkWeaver (http://elementscommunity.org/forum/index.php/topic,20376.0.html) and I can already tell you that the decks FGei(c)s and -(cn)s will jump up by a grand or something easily ... yet without skips!

Now imagine THIS deck ... We are talking about a realistic FGei somewhere beyond 7k here 

[kirchj33]

Stats updated.  Looking forward to seeing what everything looks like with skips.  Lets get 500 games on Fast-Draw Ghostal so we can see what it looks like with skips as well

[Onizuka]

Time for kirch the FGei killah to reappear!

[Jangoo]

Now imagine THIS deck ... We are talking about a realistic FGei somewhere beyond 7k here 

Just ran Ghostal through the new Statmasta and it already beat 6k FGei without skips ...

Since its performance against the various gods looks very balanced though, I doubt it will profit as much from skip-stats as other decks.
You can't really boost allrounders by skipping just a handful of gods ...

[nilsieboy]

The Fast-Draw Ghostal stats are finished (though I'll probably play the deck some more, it's good):

deck	Ghostal
players	10 men

version	1.285		win-rate	65,87%
$ track ?	enabled		win-rate (n)	65,84%
games	208		Score/h	881
win-loss-(EM)	137-71-(98)		Score/h (n)	902
time (h:m:s)	14:29:25		FGei[c]* (http://elementscommunity.org/forum/index.php/topic,14626.0.html)	4639
min/game	04:11		FGei[cn]* (http://elementscommunity.org/forum/index.php/topic,14626.0.html)	4748

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

score/h	FGei(c)	Statmasta™4000	wins	losses	skips	EM/Wins
1313    977    658    -870    954    1604    -583    1536    1056    1352    643    901    -246    1119       680    1348    230    1191    1802    966    1426    -2400    609    722    1590    321    400    715    1140	6732    4863    4292    -870    4373    7031    995    6858    5591    5927    3320    4629    1133    6039       4474    5908    1828    5223    7898    5046    6250    -2400    4037    4009    7248    2180    3468    4737    5771	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	4    7    4        9     3    1    8    6    4    4    6    1    9    5    8    3    7    4    5    8        4    7    7    3    5    2    3	1     3     3     5     2          9         3         3     2     5     1     4         5                     3     5     3         5     5     2     2		2      6      1            9      3      1      7      4      4      4      4            2      2      8      2      7      4      1      8            3      3      6      3            1      3

http://www.mediafire.com/?ldhtty5oapm315o

wow... just wow...
epic winrate and lots of money!

[rosutosefi]

I'm tempted to sell some of my upped cards just to build that deck!

[Djhopper :)]

Uh huh! Top of the leader-board baby! 

[Calindu]

Holy mother of rush, that is insane!And almost 1k score/hour!

[mwaetht]

How did you get such fast games? I played about 20 games with that deck, and I only got (winning) games below 6 minutes 5 times.

[Rember]

How did you get such fast games? I played about 20 games with that deck, and I only got (winning) games below 6 minutes 5 times.

It's an average game length taking both the wins and losses into account.

[rosutosefi]

I get 5-minute wins partly unupgraded. Maybe use high-speed AI?

[Jangoo]

ah?
then is the FGei based on that average time?

Yes sort of ... each FGeis formula in the Statmasta is calculated as such:

FGei = Electrum won / time spent (total: wins + losses + skips)


For the new edition of the Statmasta this means in detail:

FGei(e) = Electrum actually won / time actually spent
FGei(c) = Electrum you would have won if the gods had handed out cards according to their average droprates / time actually spent
FGei(cn) = Electrum you would have won if the gods had handed out cards according to their average droprates / time you would have spent if you had met every FG equally often

Now for the FGei(cn) the average time/game each god has is rather important because it is multiplied with the number of theoretical, normalized encounters with each god (which is of course the same for all gods). This way a new, theoretical "time total" is created for each god and consequentially for all games played which will in return shift your FGei(cn) away from the FGei(c) in almost all cases.

You could express all this briefly by saying: 
Throughout all the games in a sample, a profile is created for each god.
To get an FGei(cn) all the gods profiles are resized to the same size and then merged.

"Win-rate", "EM-rate", "Elec won / regular win" and of course "time/game" god make up a gods profile and lead to:
A gods individual FGei(c)  ->  These individual FGei(c)s are actually the most interesting values since they express the sum of all above factors ... You could say they are a decks and gods efficiency/performance profile.

-------

And while we are at it:

The      S T A T M A S T A™ r e a l t e c      is out.

It features the recently confirmed individual cardspin-rates for each god when it comes to calculating the FGei(c) and FGei(cn) as explained above.
This has significantly boosted (and sometimes shifted) our test-decks FGeis as you will see as soon as kirchj has updated the OP.

More information on it and the download can be found in the Statmasta-thread (see my sig).

Please use the realtec-edition one from now on, so that any statsheet-posts show comparable,
up to date results ... and so I don't have to copy-paste everything later