I took him to be saying that the X-ray emitting gases were also further out because gravitationally attracted by the dark matter, but on re-reading his article it appears he did NOT say that, so I withdraw that objection. There's no reason to think the collision didn't occur (and every reason to think it did), but whether it did is not relevant to his position.

Perhaps so, but the evidence for any one explanation (or group of explanations) still seems inconclusive.

An analogy to other types of known normal matter doesn't really help the cause of exotic dark matter. You're essentially listing other explanations for missing mass that require no exotic dark matter.

Ultimately, yes, and if dark matter and/or dark energy are invalid concepts then those ardently contending they're the only possible explanations and pursuing proof of same today will be much chagrined by tomorrows scientists exposing their errors. If that's how things develop, however, I'd prefer the process take decades rather than centuries, and the more inertia an erroneous theory has within the community the more evidence it will take to overcome it.

That's fair enough, provided we don't inadvertently allow it become an excuse for never finding direct evidence. Otherwise there's no difference between dismissing the lack of direct evidence for dark matter and dismissing the lack of direct evidence for MACHOs; it's just a question of which theory unsupported by direct evidence one prefers.

I don't think it's flimsy, as such, but I do think it inconclusive.

Theoretical estimates of normal dark matter could be off.

Less of it could lie in the immediate vicinity of lensed objects but more along the complete path between us and them.

The image itself could be distorted if the lensing mass is so close it strongly affects parts of the background but others little or none.

Those are just the most obvious solutions occurring to one layman, acknowleding that all may be untenable. Regardless, the basic principle that experiments and observation should seek new theories and test existing ones rather than simply to prove existing ones remains valid. An anomaly alerted us to a problem, but just because a solution that may be tenable was found doesn't mean we should just set out to prove that solution and cease bothering to look for one that fits the evidence better, particularly if one exists that both fits the evidence better AND is simpler.