The bipolar power spectrum (BiPS) is a statistical estimator that measures correlations between pairs of CMB multipoles otherwise assumed independent in a rotationally invariant universe. Single-field inflation predicts vanishing BiPS at all multipoles to high precision. Observations show small but systematic non-zero values at low l, indicating a preferred axis along which CMB power preferentially aligns (Naselsky 2012; Moss 2011).
The standard model assumes inflation produces statistically isotropic primordial perturbations with no preferred direction. The cosmological principle then propagates that isotropy forward to the CMB power spectrum at every angular scale. A persistent non-zero BiPS signal has nowhere to come from in the model except as an unaccounted systematic.
SCT replaces the hot-dense-center with a superluminal collision and the thermalized debris field. From this single change, the cascade is intrinsically directional. The two parent pockets intersected at superluminal v_rel with non-zero impact parameter b (P22), and the multi-stage thermalization (P36, P37, P38, terminating before t ≈ 1 second per P40) preserved the impact-parameter direction as a privileged axis through every cascade stage. Angular momentum is a conserved vector quantity protected by Noether's theorem (P30); thermalization can redistribute it but cannot erase it.
The expected BiPS signal is concentrated at low multipoles (l ≈ 2 to 20) because the largest cascade stages imprint coherence on the largest angular scales, while smaller-scale fluctuations come from later, smaller-amplitude stages with progressively decreasing impact-parameter coherence. The BiPS preferred axis is predicted to coincide with the angular-momentum axis from the impact-parameter geometry (P31, P32, P41), and therefore with the Axis of Evil (recid 24), the hemispherical asymmetry (recid 28), the connected quadrupoles (recid 18), and the CMB dipole direction (P64).
The same M10 mechanism that produces the four other large-angle CMB anomalies produces the bipolar power signal. Five anomalies share one axis because they all trace back to one collision geometry, detected by five different statistical estimators sensitive to the same underlying J-axis imprint. There is no need for special curvaton fields, modulated inflation, or systematic-error explanations.
If the BiPS preferred axis is statistically incompatible with the quadrupole-octupole alignment axis, the hemispherical-asymmetry axis, or the CMB dipole at greater than 3σ (i.e., the five anomaly axes do not share a common direction), the M10 common-collision-axis explanation fails. CMB-S4 polarization measurements will sharpen the cross-axis test in the next few years.