You are doing it wrong:
Dusk Mantle: 50% of blocking the incoming attack.
Turtle Shield: Set the creature in a time bubble for 1 turn if they attack.
The 2nd shield will "freeze" the creature every other turn. So if he has 10 turns, he will only attack during 5. The other 5 the creature is "Frozen". It doesn't attack and it can't use its ability
The 1st shield says that it has a probability of 50% of blocking the incoming attack.
That being said, it means that it MAY or MAY NOT block it.
If a creature attack for 10 turns:
The probability of blocking r attacks is:
B(r/n=10, p=0,5) = (n/r)*[(p)^r]*[(1-p)^n-r]
As p = 1-p; for p being 0,5 then:
B(r/ n=10, p=0,5 = (10/r)* 0,5^10 = (10/r)* 0.0009765625
r= 0 ; P= 0.001
r= 1 ; P= 0.010
r= 2 ; P= 0.044
r= 3 ; P= 0.117
r= 4 ; P= 0.205
r= 5 ; P= 0.246
r= 6 ; P= 0.205
r= 7 ; P= 0.117
r= 8 ; P= 0.044
r= 9 ; P= 0.010
r=10; P= 0.001
This means that chances of blocking EXACTLY 5 attacks from 1 creature is = 24,6%
Chances of blocking AT LEAST 4 attacks from 1 creature (sum from r=4 to r=10) = 82.8%
Chances of blocking AT LEAST 5 attacks from 1 creature (sum from r=5 to r=10) = 62.3%
Chances of blocking AT LEAST 6 attacks from 1 creature (sum from r=6 to r=10) = 37.7%
( 'at least 4' stands for "blocking either 4, 5, 6, 7, 8, 9 or 10 attacks", ergo, blocking by minimun 4 of the attacks)
B(r/n,p)
B=Bernoulli distribution
r= succesful event
n= total events
p= probability of succesful event
(n/r) = combinatory N , R = (n!) / [(n-r)! * r!]
This number calculates the different arrangements of having r in n events for example if n=10 and r=4:
T=true =r
F=false= ~r (not r)
T T T T F F F F F F
T T T F T F F F F F
T T T F F T F F F F
etc, for this there are 210 possible combinations or arrangements.
Now you may understand how Fog shields, Dask Mantle and Thorn Capparace, Permafrost are WAY different from Turtle Shield.
(By the way, your reasoning isnt entirely false, the thing is that during long time probability was conceived as "if i repeat this experiment n times for n a great number, then the probability = p". But this is a misconception)
[edit: faulty math and minor spelling
]