Direct <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mi>C</mml:mi><mml:mi>P</mml:mi></mml:math> violation and the <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mi mathvariant="normal">Δ</mml:mi><mml:mi>I</mml:mi><mml:mo>=</mml:mo><mml:mn>1</mml:mn><mml:mo stretchy="false">/</mml:mo><mml:mn>2</mml:mn></mml:math> rule in <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mi>K</mml:mi><mml:mo stretchy="false">→</mml:mo><mml:mi>π</mml:mi><mml:mi>π</mml:mi></mml:math> decay from the standard model

We present a lattice QCD calculation of the I 1=2, K decay amplitude A 0 and 0 , the measure of direct CP violation in K decay, improving our 2015 calculation [1] of these quantities. Both calculations were performed with physical kinematics on a 32 3 64 lattice with an inverse lattice spacing of a -1 1.378468 GeV. However, the current calculation includes nearly 4 times the statistics and numerous technical improvements allowing us to more reliably isolate the ground state and more accurately relate the lattice operators to those defined in the standard model. We find ReA 0 2.990.320.59 10 -7 GeV and ImA 0 -6.980.621.44 10 -11 GeV, where the errors are statistical and systematic, respectively. The former agrees well with the experimental result ReA 0 3.320118 10 -7 GeV. These results for A 0 can be combined with our earlier lattice calculation of A 2 [2] to obtain Re 0 = 21.72.66.25.0 10 -4 , where the third error represents omitted isospin breaking effects, and ReA 0 =ReA 2 19.92.34.4. The first agrees well with the experimental result of Re 0 = 16.62.3 10 -4 . A comparison of the second with the observed ratio ReA 0 =ReA 2 22.456, demonstrates the standard model origin of this "I 1=2 rule" enhancement.