Unitary Triangle Fit: CKM Analysis

• The Unitary Triangle Fit is a global flavor-physics analysis that uses Bayesian methods to determine the CKM apex from semileptonic decays, meson mixing, and CP violation observables.
• It employs a comprehensive set of measurements, including B decay rates, neutral-meson oscillations, and CP-violating parameters, to constrain the Wolfenstein parameters.
• Recent refinements integrate geometric interpretations of CP phases and ΔF=2 constraints to probe both Standard Model consistency and potential new-physics effects.

The Unitary Triangle Fit is the global flavor-physics analysis that determines the apex of the CKM unitarity triangle and tests the internal consistency of the Standard Model through correlated constraints from semileptonic $B$ decays, neutral-meson mixing, and CP-violating observables. In the UTfit implementation, it is a Bayesian global fit for the Wolfenstein parameters $\bar\rho$ and $\bar\eta$; in parallel, recent geometric work has recast the triangle framework so that the parametrization phases $\delta_{\rm PDG}$ and $\delta_{\rm KM}$ appear as explicit angles on the complex plane and obey a quadrangle sum rule linked to standard unitarity-triangle angles (Bevan et al., 2014, Yang, 15 Sep 2025).

1. Algebraic origin and geometric content

The unitarity-triangle construction starts from orthogonality relations implied by the unitarity of the flavor-mixing matrix $V$ or $U$. A standard relation, emphasized in the geometric analysis of flavor mixing, is

$$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$

which is interpreted as a triangle in the complex plane because the three complex vectors sum to zero. The conventional triangle angles are the rephasing-invariant quantities $\alpha,\beta,\gamma$, for example

$$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$

This fixes the geometric setting in which CKM-phase information is encoded (Yang, 15 Sep 2025).

Within this framework, the fit is not merely a graphical closure exercise. The triangle serves as the geometric image of a constrained parameter-estimation problem in which multiple observables select an allowed region for the apex in the $\bar\rho$0 plane. The geometry is therefore both diagnostic and inferential: angle measurements, side constraints, and mixing observables collectively test whether a single unitary CKM description is viable.

A recurrent misconception is that the CP phases entering specific parameterizations are automatically geometric observables. The geometric study of $\bar\rho$1 and $\bar\rho$2 makes the opposite point precise: the standard triangle already encodes CP-violating geometry, but these phases are not usually pictured as explicit angles on it. Their geometric identification requires specific rephasing-invariant constructions involving products of matrix elements and $\bar\rho$3 (Yang, 15 Sep 2025).

2. Global-fit methodology in the Standard Model

The Standard-Model UT analysis reported by the UTfit Collaboration is performed with a Bayesian global fit using the method developed in the UTfit program. The primary fit parameters are the CKM Wolfenstein quantities $\bar\rho$4 and $\bar\rho$5, extracted from a broad set of flavor observables rather than from any single angle or side measurement (Bevan et al., 2014).

The fit uses, as basic constraints, $\bar\rho$6 from semileptonic $\bar\rho$7 decays, $\bar\rho$8 and $\bar\rho$9 from $\bar\eta$0 oscillations, $\bar\eta$1 from $\bar\eta$2 mixing, $\bar\eta$3 from charmless hadronic $\bar\eta$4 decays, $\bar\eta$5 from charm hadronic $\bar\eta$6 decays, and $\bar\eta$7 from $\bar\eta$8. The paper states that experimental inputs are mostly taken from HFAG 2012, nonperturbative hadronic parameters come from the latest FLAG 2013 lattice averages, and the full numerical input set is available on the UTfit website (Bevan et al., 2014).

A specific update concerns the determination of the angle $\bar\eta$9 from charmless hadronic $\delta_{\rm PDG}$0 decays using isospin analyses. The new input highlighted in the analysis is

$\delta_{\rm PDG}$1

including the latest Belle result. This update modifies the posterior for $\delta_{\rm PDG}$2 and, through the global combination, affects the allowed UT region (Bevan et al., 2014).

Methodologically, the UT fit is a consistency test of the CKM paradigm. The overlap of the global posterior with the separate allowed regions from individual observables is itself part of the result: consistency across many independent measurements supports the hypothesis that a single unitary CKM matrix describes flavor and CP violation in the quark sector.

3. Standard-Model determinations and predictive outputs

Using the full set of constraints in the Bayesian framework, the analysis reports the CKM apex as

$\delta_{\rm PDG}$3

For the angle extraction from charmless decays, the combined analysis gives

$\delta_{\rm PDG}$4

These are the core Standard-Model outputs of the global UT fit (Bevan et al., 2014).

The fit also produces Standard-Model predictions for observables not used simply as isolated point estimates but as cross-checks of the global CKM solution. The paper gives

$\delta_{\rm PDG}$5

to be compared with the experimental value

$\delta_{\rm PDG}$6

which the paper states corresponds to agreement at about the $\delta_{\rm PDG}$7 level. It also reports

$\delta_{\rm PDG}$8

and compares them with then-recent CMS and LHCb measurements (Bevan et al., 2014).

These outputs illustrate the role of the UT fit as a predictive engine. The same CKM solution that fixes the apex also propagates to leptonic and rare-decay observables, so agreement or tension can be assessed globally rather than process by process. This suggests that the fit is valuable not only for parameter extraction but also for stress-testing the coherence of the Standard Model flavor sector.

4. Beyond-the-Standard-Model extension and $\delta_{\rm PDG}$9 constraints

The UT analysis is extended to allow possible new-physics effects in neutral-meson mixing, namely in $\delta_{\rm KM}$0–$\delta_{\rm KM}$1, $\delta_{\rm KM}$2–$\delta_{\rm KM}$3, and $\delta_{\rm KM}$4–$\delta_{\rm KM}$5 systems. For $\delta_{\rm KM}$6 mixing, with $\delta_{\rm KM}$7, the NP effects are parameterized as

$\delta_{\rm KM}$8

In the Standard Model,

$\delta_{\rm KM}$9

equivalently $V$0 and $V$1 (Bevan et al., 2014).

To sharpen the $V$2 constraints, the NP fit also includes the semileptonic asymmetry in $V$3 decays, the dimuon charge asymmetry, the $V$4 lifetime from flavor-specific final states, and the CP-violating phase and decay-width difference from time-dependent angular analyses of

$V$5

With these ingredients, the NP fit still selects a CKM apex consistent with the Standard-Model one: $V$6 The fitted NP parameters are

$V$7

$V$8

all compatible with the Standard-Model expectations $V$9 and $U$0 (Bevan et al., 2014).

The paper further states that the ratio of NP to SM amplitudes must satisfy approximately: in $U$1 mixing, less than $U$2 at 68% probability and less than $U$3 at 95%; in $U$4 mixing, less than $U$5 at 68% probability and less than $U$6 at 95%. This is one of the central phenomenological conclusions of the analysis: the allowed room for NP in $U$7-mixing is modest (Bevan et al., 2014).

The fit translates these mixing constraints into bounds on the coefficients of the most general $U$8 effective Hamiltonian through

$U$9

where $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$0 encodes generally complex NP flavor couplings, $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$1 is a loop factor, and $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$2 is the NP scale. For a generic strongly interacting theory with arbitrary flavor structure, the paper emphasizes $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$3 and $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$4, so the allowed ranges of $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$5 can be translated directly into lower bounds on $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$6. Loop-mediated NP can be rescaled by $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$7 or $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$8, which weakens the inferred lower bound on $$V_{11}^* V_{31} + V_{12}^* V_{32} + V_{13}^* V_{33} = 0 \, ,$$9 but still often leaves NP scales near or above LHC reach (Bevan et al., 2014).

5. Geometric refinements: CP phases, quadrangles, and inverse unitarity triangles

A significant geometric refinement of the unitarity-triangle program is the representation of $\alpha,\beta,\gamma$0 and $\alpha,\beta,\gamma$1 as rephasing-invariant angles on the complex plane. The paper defines

$\alpha,\beta,\gamma$2

and

$\alpha,\beta,\gamma$3

In this formulation, these quantities are not abstract parameters tied only to a chosen Euler parameterization; they are geometric arguments of complex ratios (Yang, 15 Sep 2025).

The same work emphasizes the sum rule

$\alpha,\beta,\gamma$4

Geometrically, this is expressed as a quadrangle relation on the complex plane: $\alpha,\beta,\gamma$5 The quadrangle is constructed by combining the standard unitarity triangle with an alternative triangle obtained from the inversion formula of a unitary matrix, for example

$\alpha,\beta,\gamma$6

Eliminating $\alpha,\beta,\gamma$7 yields a four-term relation whose four corner angles are identified with $\alpha,\beta,\gamma$8, $\alpha,\beta,\gamma$9, $$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$0, and $$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$1 (Yang, 15 Sep 2025).

The same paper introduces a new family of inverse unitarity triangles from

$$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$2

Using the explicit cofactor form of the inverse, the authors construct nine triangle relations by grouping terms so that sums of three complex numbers vanish. These are termed inverse unitarity triangles because they arise from the cross-product or cofactor structure of $$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$3, in contrast to the usual dot-product-based unitarity triangles. The phase matrices $$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$4 and $$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$5, built from third-order invariants, satisfy

$$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$6

with $$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$7 the matrix of ordinary unitarity-triangle angles (Yang, 15 Sep 2025).

For fit-oriented analyses, the paper lists as especially relevant the rephasing-invariant phase definitions, the standard triangle angles, the quadrangle angle sum, and the decomposition identity $$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$8. It states that these formulas connect measurable CKM/PMNS-like quantities to geometric objects and allow one to constrain phases through angle-closure conditions. A plausible implication is that future unitarity-triangle analyses can use the usual triangle constraints together with quadrangle and inverse-triangle identities as internal cross-checks on phase determinations.

6. Scope of the term and adjacent triangle frameworks

The expression “triangle fit” appears in adjacent areas of flavor and neutrino phenomenology, but these constructions should not be conflated with the CKM Unitarity Triangle fit. In cosmic neutrino propagation, averaged oscillations map source flavor ratios $$\alpha = \arg \left [ - {V_{31} V_{33}^{*} \over V_{11} V_{13}^{*} } \right ], \qquad \gamma = \arg \left [ - {V_{11} V_{13}^{*} \over V_{21} V_{23}^{*} } \right ] .$$9 to Earthly ratios $\bar\rho$00 through

$\bar\rho$01

Unitarity reduces flavor space to a triangle, and the area of the Earthly flavor triangle obeys

$\bar\rho$02

In that setting, the geometric problem concerns invertibility of flavor propagation and source reconstruction, not the CKM apex in the $\bar\rho$03 plane (Fu et al., 2014).

The same distinction applies to non-unitary neutrino-oscillation analyses. There, the effective light-sector mixing matrix is parameterized as

$\bar\rho$04

and the “triangle” language refers to geometric parameter-space plots of the $\bar\rho$05 parameters rather than to CKM unitarity triangles. The paper explicitly notes that these are not unitarity triangles in the CKM sense (Miranda et al., 2019).

These distinctions clarify the scope of the Unitary Triangle Fit in flavor physics. Properly understood, it is the quark-sector global analysis of CKM unitarity constraints, augmented by NP-sensitive $\bar\rho$06 tests and, in newer geometric formulations, by explicit complex-plane representations of CP phases. Its central function is to determine whether a single unitary mixing description consistently organizes the observed pattern of flavor transitions and CP violation.