A cot death expert has admitted his use of statistics at the 1999 trial of Sally Clark for the murder of two of her children was "insensitive".
Sir Roy was speaking during his defence case to the GMC
However, Professor Sir Roy Meadow defended his evidence and denied serious professional misconduct.
He told the General Medical Council that he stood by his controversial estimate of the likelihood of the two Clark children dying from cot death.
Mrs Clark was cleared of killing her two sons Harry and Christopher in 2003.
'Ball park figure'
Sir Roy told the GMC that at a committal hearing for Mrs Clark, he was asked by defence counsel about the one in 1,000 figure for sudden infant death syndrome.
He said he did not remember where the figure had come from, but when it appeared in an article he had written in an eminent paediatric journal, no one had questioned it.
Sir Roy said it was a "reasonable ball park figure" for the incidence of sudden infant death syndrome for the population as a whole.
He said he had squared it to reach a figure of one child per million for two such deaths in the same family
He then calculated that because of Sally Clark's age and the family's relative affluence, the likelihood in their case was one in 73 million - the figure he used during her trial.
It was later disputed by the Royal Statistical Society and other experts have said that once genetic and environmental factors are taken into consideration, the odds are closer to 200 to one.
'Horse racing' odds
Sir Roy said he took the one in 73 million figure to be a "mean figure and there would be a degree of variation on either side".
Sally Clark: Served three years after being wrongly convicted of killing her two sons
Angela Cannings: Served 18 months after being wrongly convicted of killing her two sons
Donna Anthony: Served six years after being wrongly convicted of killing her son and daughter
Trupti Patel: Cleared of killing three of her children
At Monday's hearing, under questioning from his own barrister, Nicola Davies QC, he said he had used analogies during the trial to try and make statistics more comprehensible.
He had said the likelihood of both children dying from cot death was like that of a punter successfully backing an 80-1 shot at the Grand National four years in a row.
But he added: "I am uncomfortable with it. I don't like the racing analogy.
"The situation in which I was giving evidence was one in which there was a family grieving, two children had died and the mother had been accused of murder.
"To raise a subject like the Grand National, that is exciting and vibrant, is insensitive and I should have thought of more appropriate odds."
Last week he told the GMC he did not see himself as an expert statistician.
Sir Roy also gave evidence as an expert witness in the trials of two other women, Angela Cannings and Donna Anthony, who were both freed on appeal after being convicted of murdering their children.
He told the GMC he had concerns over the evidence he saw regarding both Harry and Christopher's deaths.