It is odd and I have tried to break it down.
First case discussed today - the experts said the evidence was very hard to spot. (They are genuinely world leading). New supporting evidence in recent publication.
Second case - this matches what the pathologist who originally examined the body said, but there is also new supporting evidence because we now have a relevant academic study of air embolism through injection. That publication didn't exist then
Third case discussed today. The prosecution gave the wrong data on timing of tests. New engineering studies have been conducted since the trial to see if it would have been possible for Letby or anyone to poison children in the way described
Fourth case described today. I think here the panel has made a connection the defence lawyer wasn't aware of.
Fifth case described today. The child had a rare bacterial infection - not sure how attuned to this the defence would have been.
Sixth case described today. The judge told the jury it was a matter of choosing who to believe, the convicted murderer or the doctor. The defence had tried to argue some of these points.
Seventh case described today - this is based partly on information that wasn't given to the pathologist who examined the child's body, so his report had a gap. It needed someone to go back over later notes to the coroner and piece things back together. We don't know if the defence had access to this information - they have got some of their information from the Thirlwall Inquiry.
So it's a mixture of answers: new studies, information not disclosed or believed; probably some details missed. 14 world leading experts with more info vs 2 (as far as we know) perfectly competent professionals with far fewer resources and less time.