Old science: Make the theory fit the data.
New science: Make the data fit the theory.
New science: Make the data fit the theory.
Not that new. In 1919 the astronomer Eddington claimed he had verified Einstein's prediction of the bending of light by gravity, by observing light bent by the sun.
A later investigation revealed that he had published only the 15% of his observations that matched the prediction, and had left out the 85% that did not.
By a strange coincidence, the observations he published had been made with a 4 inch telescope, which simple calculation shows could not deliver the accuracy he claimed. The 10 inch telescope results he omitted had confirmed Newton's calculations.