Computer Simulations - ROUND 4

[majofa]

Can an event organizer please post clarification about the rule stating "cannot have any similar cards with your other deck"?  Does it mean the first two decks could not share similar card or is it applicable to the response-deck not being able to share similar card with a contestant's first two decks?

Thanks!

The first 2 decks.

[andretimpa]

Lots of people posting death and gravity monos too

[farscape]

This round feels like "Mission Impossible" compared to the previous ones! 

I'm having trouble making a deck to even counter 1 of the opponent decks with the given restrictions, let alone all 4.

[rob77dp]

This round feels like "Mission Impossible" compared to the previous ones! 

I'm having trouble making a deck to even counter 1 of the opponent decks with the given restrictions, let alone all 4.

At this point of the event, I suppose it is a big step up this round thrashing the participants leaving the chaff eliminated and the grain moving forward.

(I agree - much higher difficulty countering and getting high aggregate Win% this round for me compared to prior rounds)

[dawn to dusk]

This round's rules are actually absurd... brings attacking (3rd) deck's chances to pure rng... even with 5k simulations..

this. with my tests, differences can be as big as 1% per deck. and with a round like this where you are only getting 2-3% per deck anyway...

[Submachine]

Even with 5000 simulations, there WILL be a 1-3 % fluctuation. That's why I usually only take the first outcome.
Even if I would run 5000 sims 3 times and count the average of those, it would still fluctuate.

[RootRanger]

Fluctuation is inversely proportional to the square root of the sample size, however.

In other words, with a sample 4 times as large, the result fluctuates only 1/2 as much from the true win probability.

Of course, if we have 5000 simulations for a deck with a 10% win rate, the standard deviation of the difference between the simulated win rate and the actual win rate is only 0.4%, which means about two thirds of the simulations will fall within 0.4% of what the actual win rate is supposed to be.

[Submachine]

All submissions after this post are declined. Initiating simulations. Results coming soon.

[andretimpa]

Lol, I thought you were announcing the results

[RootRanger]

Nope, that's my job.

The Results: Farscape and I move on, the three people with crappy decks are out.

[Submachine]

GROUP A

	Calindu Deck 1	Calindu Deck 2	rem4life Deck 1	rem4life Deck 2	rob77dp Deck 1	rob77dp Deck 2	TOTAL %
Calindu			2.94%	4.34%	3.40%	5.16%	15.84%
rem4life	6.94%	50.48%			9.64%	14.00%	81.06%
rob77dp	0.52%	6.72%	22.42%	0.30%			29.96%

GROUP B

	dawn to dusk Deck 1	dawn to dusk Deck 2	farscape Deck 1	farscape Deck 2	Zso_Zso Deck 1	Zso_Zso Deck 2	TOTAL %
dawn to dusk			0.30%	2.00%	3.54%	7.70%	13.54%
farscape	8.98%	16.94%			4.26%	13.66%	43.84%
Zso_Zso	8.42%	14.68%	1.36%	8.52%			32.98%

GROUP C

	andretimpa Deck 1	andretimpa Deck 2	Regyptic Deck 1	Regyptic Deck 2	RootRanger Deck 1	RootRanger Deck 2	TOTAL %
andretimpa			0.32%	0.20%	0.04%	0.06%	0.62%
Regyptic	9.32%	37.5%			8.42%	16.58%	71.82%
RootRanger	12.36%	42.78%	25.26%	9.22%			89.62%

[dawn to dusk]

Swear to god I hate RNG. I got 5% more when I was doing my tests. I would still be out but...

What does this goddess of the random numbers have against me?????