But it's a backward observation. A trained professional can certainly note that the general disposition of the damage is consistent with an impact over a relatively small area. Application of stress will cause different sections of the hull plating to respond in different ways. An elastic response won't be evident in the final form, except possibly in the form of small surface microfractures visible only by microscope. These would not seriously impair the structure performance of the hull plating; such microfractures are commonly incurred in ordinary operation.
A plastic response will be evident in the form of deformations that nevertheless to do not require the material to yield. Most minor ship collisions result only in plastic deformation. And in fact, such behavior is desired because it allows collided ships to retain watertight integrity.
A yield response occurs when materials or welds part. This can occur in hull plating as a shear failure or as a brittle fracture. Temperature affects the ductile-to-brittle transformation point in all metals except for aluminum. Strain rates also dictate yield behavior. That is evident in the final disposition in obviously parted materials.
To say a "puncture" causes "deformation" isn't really a thing. Application of stress results in a combination of the three responses I summarized. One response mode does not cause another. Deformation and parted yields are two possible outcomes of the same initiating cause.
What remains suspicious for me is the lack of paint loss at the impact site and the lack of any paint or coating transfer to the damage site. Impacts almost always result in the loss of paint, especially where scraping or grinding occurs, such as in the proposed submarine impact scenario. What would cause metal to fracture without loss of paint?
The hole seems to occur at or near the joint between two hull plates. This kind of damage is consistent with a collision, of course. But it's also consistent with differential in-plane loading, or "springing" hull plates, as it's sometimes noted commonly. In-plane loading occurs when there is compression trying to push two edges of a material together. It can also occur when there is a moment vector perpendicular to the material plane. At joint edges, the plates can subduct. But buckling is also an extremely common material response of thin sheets to in-plane loading.
This article features a ship whose hull has failed in a way that caused the hull plating to buckle outward in response to in-plane angular loading. Note that the welds have failed at the greatest extent of buckling.
https://www.marineinsight.com/naval-architecture/ships-hull-fail-at-midship-region/
Absent any reinforcement, there is no guarantee whether a ship's hull plate will buckle inwards or outwards (or both) in response to such in-plane loading. And if the buckle axis is perpendicular to a seam, and the seam fails, there is no guarantee that the seamed plates will buckle in the same direction. The damage I see at the breach in Estonia's hull is not entirely inconsistent in my opinion with buckling damage. And it explains the lack of coating damage.
