The G15 with its teeny, tiny pixels out-performs the 1Dx in DR even with its enormous pixels.
Yes, that G15 is really beating the pants off the 1D X in terms of DR. Wait, I'm wrong. For a minute there, I thought DR was like golf where lower numbers were better. Oops.
1Dx: 11.2 stops at base ISO: http://sensorgen.info/CanonEOS-1D_X.html
G15: 11.5 stops at base ISO: http://sensorgen.info/CanonPowershot_G15.html
And according to DxO's measurements,
1D X: 11.77 stops at base ISO
G15: 11.52 stops at base ISO
But the bottom line is that it's wash - 1/3 stop either way, and the direction of the difference varies by the source of the data.
Also, you stated that the G15 'out-performs' the 1D X…but you neglected to qualify that statement with 'at base ISO' - and those base ISOs are different (ISO 80 for the G15, ISO 100 for the 1D X).
From the data in your the tables you linked, at the same ISO the 1D X has the advantage…and it's an advantage that gets progressively larger as ISO is increased. Maybe you have the same attitude that the DxOMark Score is based on - only base ISO matters. Have you ever raised your ISO setting above the base ISO? If so, stating that, "The G15 ... out-performs the 1Dx in DR," is pretty hypocritical, when even at ISO 100 that's not true, and starting at ISO 800 the 1D X has a 2-stop advantage.