Let's get my definition of "formal validity" out into the sunlight, for understanding and discussion.
Then I'll invite you to do the same.
Let x = "This statement is false" be an element of the set S of all self-referential statements of the form "This statement is B".
Let y = "This statement is true" be another element of S.
Let T be the property of Truth.
Let F be the property of Falseness.
Then x becomes x = F(x).
and y becomes y = T(y).
Then, because F and T are formally defined over the set S, and x and y are elements of S, both statement are formally valid.
y is formally valid and decidable: T(y).
x is formally valid, but not decidable: F(x) leads to ^x or ^F(x).
So we cannot conclude F(x), and we also cannot conclude T(x).
x is formally valid and undecidable.
---------------------
In the realm of arithmetic,
Let p = "1 + 1 = 1 + i" be an element of the set E of equations of the type A = B where A and B are expressions.
Let q = "1 + 1 = 1 + 1" be another instance
And r = "1 + = x 4 - x 3 - / i 13 / = 45 / -" be considered as be a third.
Let F and T be the properties defined above over the set E.
p is formally valid and decidable: F(p).
q is formally valid, and decidable: T(q)
r is not of the form A=B [formally invalid]; it's not an element of E.
Therefore F and T do not formally apply to r.
r cannot be discussed with respect to F and T.
Now you have my definition of formal validity.
In this case it means + and x and / and - are binary operators, taking prefix and postfix arguments.
It means elementary terms, 1, i, 4, 3, 13, 45 are proper arguments of these binary operators.
It means = is a binary relationship taking two numbers or expressions and asserting their equivalent value.
p and q follow these formal rules. One happens to be false, the other true.
r does not; it is therefore formally invalid.
It's not the case that r is undecidable with respect to truth: T(r) and F(r) don't apply.
From this, see that r and x [above] are not similar.
In a previous post, you said in effect that because r is invalid, so is x.
Since my definition of valid does not permit this, let me ask for a definition that does:
What's your definition of validity?
You seem to define valid as: if F(a), then a is invalid.
Is that how you define valid: if a is false then it's invalid?
Why use two terms [false and invalid] confusingly, in different ways and for different purposes
[false means not true and invalid means we can dismiss it] that in the end have the same definition?