What defenses have you lost to recently and why?


#1

We have the topic asking people about what attack and defense they use, but for me I’d like to hear about the few losses people have. Most people will say “this team kills almost everything”

That’s what interests me the most, the “almost”. Most people can beat 90% of the teams on a region easily (defense is way harder now) but there are a few times those of us who are still mortals lose.

I will start. I lost to a Miraboobs team with windowless weapons and Shiva Force Rick.
Why did I lose?

  • I was playing around with Rick to get him to pop, stunned myself and I let Tyreese rush instead of confusing him. Ty wiped a line, Zeke poppped and I was down 3-5 and stunned

Next raid I didn’t stun myself round 1 and won 5-0 :smiley:

That Shiva Force Rick defense would have been really nice before 6*


#2

I have a bad habit running into fully upgraded lagendary teams on Russian regions


#3

Only lost to one ranged team with 6*s since I had a full team of them. It was mirabelle with ty siddiq Monica yumiko and one other I can’t remember. All trait weapons and I have no trait weapons.

Carl teams only lost to a 4x Priya team, and timed out once against a Carl Yvette 2x shiva zeke team. Also with trait weapons and I have none.

My team is Carl (3t80) barker (2t70) Yvette (2t70) shiva (3t80) and zeke (3t80). Against melee tend to bring in Samurai Jesus lead.


#4

Dang it I forgot about all the scumbag cheaters


#5

Went for the shield Michonne, ending up hitting Mirabelle. Michonne shielded, I couldn’t kill the Michonne and Mirabelle stunned four of my teammates. Zeke popped off, and killed my Abe and Timothy which pretty much meant loss, so yeah.


#6

I only lose to Carl teams and thats if they have 2 to 3 stun swords that trigger turn 1.


#7

if you have a good team you generally lose to RNG, zeke popping shields a lot, consecutive 5/5 abs defense procs, or opponent stunners knocks your strategic toons on 1st turn and ruins your strategy and stuff like this :grin:


#8

there is only 2 viable teams in the game… Carl or Boobs… thats it… pretty fucking boring.


#9

It’s actually just rng that rarely fucks me. Full stun-impair team with Hershel popping guardian every turn. It’s still doable most of the time but sometimes I’m just unlucky with it. With Carl teams it’s not nearly as bad cause they don’t stun/impair you and do way less damage.

Rng is such amazing game design btw <3


#10

Since they’ve posted that a standard AD or stun defense has a 30% chance of working, the odds of all five missing is 0.2% (0.3^5) chance of happening …
which is why its hard to believe those odds.


#11

Recalculate, that’s one way to say “what are the odds” but there are many ways to look at that

The more holistic approach says what are the odds of 5 AD in a row, given hundreds of thousands of instances (CRW) and the odds are actually high that it will happen, often.

http://www.espn.com/chalk/story/_/id/11863567/betting-five-biggest-cognitive-biases-impair-sports-bettors-most


#12

I lose to a guy who has four 6* Shivas under a Carl lead


#13

as it’s a programming code, math is likely the only way to look at it.

Hit: Success = 70%, Fail = 30%

5/5 Failed hits = Fail x Fail x Fail x Fail x Fail
= 0.3 x 0.3 x 0.3 x 0.3 x 0.3
= 0.00243 or 0.24% or once out of every 400 tries

if it was 50/50 like on SR Armoured Walkers it would be 3% chances for 5/5 misses… which for me seems like it happens at least once every other stage attempt.

As for a stun shield… the simplified odds of getting 3 of 5 stunned while attacking is:
=.7 x .7 x .3 x .3 x .3
= 0.031 or 3%
Again, it seems like that happens a lot more than that. So it is fair to question if it is a bug or not.

@CombatDevIl


#14

This affirmation is not strong enough to caracterize a bug. There is no secret in calculating simple probability, so it is not likely a bug, just you noticing more when it happens than when it doesn’t.


#15

Confirmation bias. And the odds of missing are 70% not 30%

So 16.8% chance of missing all 5


#16

no doubt… but it’s the start.

So I guess the question is, where do we take it from here for anything that might look bug-ish? What sample size is sufficient and what is helpful structure to present data?


#17

Haven’t lost a raid since the 6s came out


#18

I can assure there is no bug on chance calculation. It is programming 101 math.


#19

A fail is not hitting or being blocked. (30% chance of happening)
17% is for successfully hitting your target 5 times out of 5 tries.


#20

That’s only 1 way to look at it.

The other way to look at it (the more statistically significant way) is to look not at your account over the course of a few raids.

But to look at the % chances for anyone playing TWD to encounter a 5x AD/STun proc rate.

Cosider the roulette wheel.

What are the chances that a roulette wheel will hit the same number, 9x in a row? Seems like a low chance?

But what about the thousands of roulette wheels in vegas, all spnning hundreds of times per hour. Now consider it over a period of time

Now ask it a different way
What are the chances that any roulette wheel will hit the same number 9x in a row given such a huge sample size?

The answer is, it is expected to happen every 15 months!