The disagreement over circle significance traces to the construction of the null ensemble: Gurzadyan and Penrose computed significance against simulations of a pure Gaussian sky plus instrument noise, while all independent reanalyses simulated skies drawing on the full ΛCDM angular power spectrum, under which the observed features are typical.

Both sides state the ensemble difference in print; it is the episode's documented crux. The claim asserts the diagnosis, not a winner; as a diagnosis it is beyond serious dispute.

Both sides explicitly acknowledge this methodological divergence in print; Gurzadyan and Penrose published a direct response confirming they used a pure Gaussian-plus-noise null while critics used ΛCDM power spectra. The strongest opposing consideration is that proponents assert the ΛCDM null is inappropriate because the CMB is mostly non-random, though this begs the question by assuming the standard model's correlations are themselves the anomaly.

Gurzadyan and Penrose themselves state in their 2010 response that they simulate “a pure Gaussian sky plus the WMAP’s noise” while the critical groups simulate skies using the ΛCDM power spectrum, and all independent reanalyses confirm this difference in their methods. Thus it is an uncontested fact that the two camps used different null ensembles, and that this is where their significance estimates diverge. The strongest objection is not to this descriptive claim but to treating the ensembles as equally legitimate: from a standard statistical standpoint, omitting the well-measured ΛCDM correlations from the null is a substantive methodological error, not just a neutral choice—but that critique goes to CCC’s correctness, not to the factual content of this claim.

Gurzadyan and Penrose themselves state that the critics "simulate maps using the CMB power spectrum for LCDM, while we simulate a pure Gaussian sky plus the WMAP's noise," and every critical paper documents the same split. The claim is a description of what each side did, not a verdict on which ensemble is correct, and that description is uncontroversial on the face of the record. The strongest opposing consideration is that calling this a symmetric "choice of ensembles" understates an asymmetry: omitting the known power spectrum from the null is a methodological error rather than a legitimate alternative — but that is the attribution dispute (CCC-C009 versus CCC-C010), not a reason to deny that the ensemble difference is where the significance disagreement actually sits.

This is a documented fact about the record, stated verbatim by the proponents themselves (CCC-E005: "Both groups simulate maps using the CMB power spectrum for LCDM, while we simulate a pure Gaussian sky plus the WMAP's noise") and exhibited in every critical reanalysis. That the same rings are anomalous under one ensemble and typical under the other is the direct content of CCC-E002 and CCC-E003. The strongest opposing consideration is that describing this as a symmetric "choice of ensembles" understates the asymmetry — testing a detection against a null that omits known statistical properties of the data is a textbook error. But that objection concerns which null is right, not whether the ensemble difference is what the significance dispute traces to, which is all this claim asserts.

This is a descriptive claim about what each side did, and it is documented verbatim on both sides: the proponents' own response note says "Both groups simulate maps using the CMB power spectrum for LCDM, while we simulate a pure Gaussian sky plus the WMAP's noise." Every critical paper describes the same division. Two qualifications keep it short of established: the proponents' 2011 note concedes circles appear at similar frequency in WMAP-derived-spectrum simulations while arguing a theoretical ΛCDM spectrum removes them, so the ensemble dispute has finer structure than a binary; and Bodnia et al.'s artifact diagnosis shows the ensemble is not the only locus of disagreement. Separately — and this is a judgement partly on methodological priors rather than on any evidence record — the two ensembles are not epistemically symmetric: a null that omits the measured power spectrum cannot test a claim of excess over ΛCDM.

Both sides state the ensemble difference in print (CCC-E005 versus CCC-E002); the claim asserts the diagnosis, not a winner, and as a diagnosis it is beyond serious dispute.

Both sides state the ensemble difference in print (CCC-E005 versus CCC-E002); the claim diagnoses the disagreement without picking a winner, and as a diagnosis it is beyond serious dispute. No change.

Both sides state the ensemble difference explicitly in print (CCC-E005 against CCC-E002 and CCC-E003). The claim diagnoses the disagreement rather than adjudicating it, and as a diagnosis it is uncontested. Unchanged.

Both sides state the ensemble difference in print (CCC-E005 against CCC-E002 and CCC-E003). The claim diagnoses the disagreement without adjudicating it, and as a diagnosis it is essentially uncontested. Unchanged.

Both sides state the ensemble difference in print (CCC-E005 against CCC-E002 and CCC-E003). The claim diagnoses the disagreement rather than adjudicating it, and as a diagnosis it is uncontested. Unchanged.

Both the proponents and critics explicitly agree on this methodological divide in the primary literature. Gurzadyan and Penrose's own response (CCC-E005) confirms they utilize a pure Gaussian sky plus noise as their baseline, explicitly rejecting the ΛCDM power spectrum simulations used by all critical reanalyses. The strongest opposing consideration is that establishing this difference in method does not, on its own, resolve which ensemble is physically correct for testing the hypothesis.

Gurzadyan & Penrose themselves state that they simulate “a pure Gaussian sky plus the WMAP's noise,” whereas the critical groups simulate ΛCDM-spectrum skies; the independent reanalyses explicitly attribute the discrepancy in significance to this ensemble choice. All sides therefore agree on the factual content of this claim: the null ensembles differ, and under ΛCDM simulations the features are typical. The strongest opposing argument is normative rather than factual: proponents argue ΛCDM-spectrum simulations are illegitimate as a null because they “bake in” non-random CMB structure, but that is a contention about which null is appropriate, not about what was actually done.

Gurzadyan and Penrose themselves stated that the critics "simulate maps using the CMB power spectrum for LCDM, while we simulate a pure Gaussian sky plus the WMAP's noise," and every critical paper records the same split. As a description of what each side did, the claim is uncontroversial on the primary record. The strongest opposing consideration is that calling this a symmetric choice of ensembles understates an asymmetry: omitting known statistical properties of the data from the null is a textbook error, and the proponents' justification that the CMB is mostly non-random presupposes part of what is in dispute.

As a description of where the disagreement lives, this claim is documented on both sides: the proponents' own response note states that the critics simulate with the ΛCDM power spectrum while they simulate a pure Gaussian sky plus noise (CCC-E005), and every critical paper confirms it. No party disputes the facts of who simulated what. The strongest opposing consideration is normative rather than factual — the proponents hold that ΛCDM-spectrum simulations smuggle the disputed structure into the definition of chance — but that is the attribution dispute (CCC-C009 versus CCC-C010), not a challenge to this claim's descriptive content. Where the verdict leans on priors: the near-universal view that the null must contain the rival model's known predictions is standard statistical doctrine, used here for calibration only, not as evidence.

This is a descriptive claim about what each side did, and it is affirmed in the proponents' own response note: "Both groups simulate maps using the CMB power spectrum for LCDM, while we simulate a pure Gaussian sky plus the WMAP's noise" (CCC-E005). Every critical paper describes the same divergence, and the 2011 Gurzadyan–Penrose follow-up sharpens it further by distinguishing WMAP-derived from theoretical ΛCDM spectra. The strongest opposing consideration is that framing this as a symmetric "choice" flatters the proponent side: a detection claim tested against a null that omits the measured two-point correlations of the data is a standard error, and the justification offered — that most of the CMB signal is not random — presupposes part of what is in dispute. That is a point about which null is correct, not about the accuracy of this claim, which is well documented on both sides.

Both camps state the ensemble difference explicitly in print (CCC-E005 against CCC-E002 and CCC-E003). The claim diagnoses the disagreement rather than settling it, and as a diagnosis it is uncontested. Unchanged.

The primary literature on both sides explicitly acknowledges that the proponents computed significance against a pure Gaussian sky plus noise, while independent critics used skies simulated with the full LCDM angular power spectrum. This methodological divergence is the undisputed root of the conflicting significance claims. The strongest opposing consideration is whether describing this as a symmetric "disagreement" obscures the fact that omitting known data correlations in a null is a standard statistical error.

Gurzadyan & Penrose themselves state that they simulate a “pure Gaussian sky plus the WMAP’s noise” while critics use simulations with the ΛCDM power spectrum (CCC-E005), and all critical papers characterize this ensemble difference explicitly (CCC-E002–E004, E008). There is no serious dispute about this factual description; what is contested is which ensemble is appropriate as a null, which belongs to attribution rather than to this methodological claim. The only opposing point is the proponent argument that ΛCDM-spectrum simulations are illegitimate because most of the CMB is “non-random,” but that does not alter the descriptive accuracy of CCC-C003.

Gurzadyan and Penrose themselves state that they simulated a pure Gaussian sky plus instrument noise while the critical groups simulated maps from the ΛCDM power spectrum, and every independent reanalysis describes the same split. As a description of where the significance disagreement lives, the claim is documentary rather than interpretive. The strongest opposing consideration is that calling this a symmetric choice of ensembles understates an asymmetry: a detection claim tested against a null that omits the known two-point correlations of the data is a misspecified test, and the proponents' defense of the poorer null already presupposes that most of the CMB is non-random.

Both camps state the ensemble difference in their own words — the proponents' response note says it verbatim (CCC-E005) — and every critical paper confirms it. That the features are typical under the ΛCDM ensemble is found by four independent groups, and the proponents' 2011 note effectively concedes circles arise at similar frequency under a WMAP-derived spectrum. The claim assessed here is descriptive — what each side did and what follows under each null — and that is a matter of public record. Strongest opposing consideration: which null is *correct* remains genuinely contested, since the proponents argue the ΛCDM-spectrum ensemble smuggles the disputed structure into the definition of chance; but that is the attribution dispute (CCC-C009 vs CCC-C010), not a challenge to this claim's factual content.

The ensemble difference is stated verbatim by the proponents themselves — "Both groups simulate maps using the CMB power spectrum for LCDM, while we simulate a pure Gaussian sky plus the WMAP's noise" (CCC-E005) — and is echoed in every critical paper, so the descriptive content is essentially uncontested. Diagnostically it locates the crux precisely: the same features are 6σ against one null and unremarkable against the other. The strongest opposing consideration is that the source ledger records a later proponent note (Gurzadyan & Penrose 2011) conceding circles arise at similar frequency under a WMAP-derived spectrum while arguing they vanish under the theoretical ΛCDM spectrum — which means the story is not a clean two-ensemble split but a finer dispute about which spectrum counts. I also note, as a matter of general statistical practice rather than of these records, that a null which omits the measured correlations of the data is not a defensible baseline for a detection claim.

Both camps state the ensemble difference explicitly in print (CCC-E005 against CCC-E002 and CCC-E003). The claim diagnoses the disagreement rather than settling it, and as a diagnosis it is uncontested.

Uncontroversial as a description of what each side did — the ensemble difference is stated in the proponents' own response note and in every critical paper. What remains contested is which null is the right one, and that is the attribution dispute (CCC-C009 vs CCC-C010), not this claim.

The methodological origin of the statistical dispute is fully documented and conceded by both sides in primary literature. Gurzadyan and Penrose (2010b) explicitly confirmed that their null simulations used a pure Gaussian sky plus instrumental noise, whereas all critical reanalyses employed synthetic skies generated from the ΛCDM power spectrum. Under the latter ensemble, the observed circle abundances match expectations. Strongest opposing consideration: Gurzadyan and Penrose contend that adopting the ΛCDM power spectrum inherently assumes cosmic inflation and washes out genuine non-random physical circular patterns.

The claim is a methodological description of the dispute, and it is documented from both sides: Gurzadyan & Penrose's own response note states verbatim that the two reanalyses used the ΛCDM power spectrum while they simulated a pure Gaussian sky plus noise, and the reanalyses' abstracts confirm their ensembles. The strongest opposing consideration is that describing the disagreement does not adjudicate which ensemble is correct — but the claim asserts only the trace of the disagreement to ensemble construction, which the primary sources establish.

CCC-E005 directly records the proponents distinguishing their pure Gaussian-plus-noise simulations from critics' simulations based on the ΛCDM CMB power spectrum. The independent reanalyses consistently identify the structured-sky ensemble as the reason they obtain much broader expected variance. The strongest opposing consideration is the proponents' substantive dispute over which ensemble is scientifically appropriate, but that dispute confirms rather than contradicts the methodological claim.

Primary sources on both sides state the discrepancy in nearly identical terms: Gurzadyan & Penrose simulated pure Gaussian plus instrument noise; the critical groups simulated full ΛCDM-spectrum skies. That methodological fork, not disagreement about the map pixels, produces the significance gap. Strongest opposing consideration is the further claim that concentric families cannot be random even under ΛCDM, which is an assertion rather than a quantified result against the proper ensemble.

Both sides state the ensemble disagreement in print: the proponents' own response note says the critics "simulate maps using the CMB power spectrum for LCDM, while we simulate a pure Gaussian sky plus the WMAP's noise" (CCC-E005), and the critics' abstracts confirm the features are typical under their ensemble (CCC-E002, CCC-E003). The claim's descriptive core is therefore documented from both directions. Strongest opposing consideration: the claim's "traces to" is too tidy for the later record — Jow & Scott's look-elsewhere correction, Bodnia's few-pixel artifact diagnosis, and Bielewicz's low-ℓ conditioning are failure modes independent of the ensemble choice, so ensemble construction is one locus of disagreement, no longer the whole of it.

Stated in the proponents' own words (CCC-E005) and matched by the critics' descriptions of their ensembles (CCC-E002, E003, E004). Bielewicz et al. refine it: within ΛCDM ensembles, conditioning on the observed ℓ ≤ 7 multipoles is what renders the large-radius deficit typical (CCC-E056, E057), so the crux is the ensemble's large-scale content as well as its spectrum. The proponents' contention that the ΛCDM ensemble is the wrong null is asserted, not quantified.

The literature documents that Gurzadyan and Penrose (2010b) explicitly confirmed they evaluated significance against a pure Gaussian white-noise sky plus instrument noise, whereas Wehus & Eriksen (2011) and Moss et al. (2011) tested against skies synthesized from the ΛCDM power spectrum. Both sides openly agree on this exact methodological divergence as the origin of their conflicting significance values. The strongest opposing consideration is Gurzadyan and Penrose's conceptual argument that the ΛCDM power spectrum should not be assumed as the random null because the CMB fluctuations may not be purely Gaussian random noise.

Both sides document the ensemble difference in their own words: Gurzadyan & Penrose state they simulate a pure Gaussian sky plus noise while the critics use the ΛCDM power spectrum, and the critics' simulations indeed show the features at chance abundance. The strongest opposing consideration is that tracing the disagreement to the ensemble does not settle which ensemble is correct — the proponents argue ΛCDM simulations illegitimately presume the signal is random — but the claim as stated concerns where the disagreement lives, which is squarely documented.

The proponents explicitly contrast their pure-Gaussian-plus-noise simulations with critics' simulations drawn from the ΛCDM CMB power spectrum. Independent reanalyses consistently report that the observed structures become typical under the latter ensemble, localizing the principal significance disagreement to the null construction. The strongest opposing consideration is that later disputes also involve masks, ellipticity, artifact handling, and search parameters, so the ensemble is the central but not the only methodological difference.

Gurzadyan & Penrose themselves state that the critics simulate maps from the ΛCDM power spectrum while they simulate pure Gaussian sky plus noise; the independent reanalyses uniformly confirm that this ensemble choice is decisive and that the features become typical under the former. The methodological disagreement is therefore documented on both sides and is the sole remaining locus of the significance dispute. The strongest opposing consideration is the proponents’ further claim that concentric families cannot be random even under ΛCDM, a claim never quantified against the full ensemble.

The ensemble disagreement is documented verbatim from the proponents' side — "Both groups simulate maps using the CMB power spectrum for LCDM, while we simulate a pure Gaussian sky plus the WMAP's noise" (CCC-E005) — and from the critics' side (CCC-E002, CCC-E003). The localization is accurate for the 2010–2011 circle exchange. Strongest opposing consideration: the claim's scope ("all independent reanalyses") no longer exhausts the critique. Hajian's contribution was realistic anisotropic noise, Jow & Scott's is a look-elsewhere correction, and Bodnia et al.'s is artifact screening at full resolution — none of which is an ensemble-choice dispute. So the claim is true of the founding exchange but understates the later critiques; confidence held to moderate for that reason.

Both proponents and critics explicitly agree in print that the statistical divergence originates from the construction of the null ensemble. Gurzadyan & Penrose (2010b) explicitly acknowledged that they tested against a pure Gaussian sky plus noise while independent groups simulated skies using the ΛCDM angular power spectrum. The strongest opposing consideration is that Gurzadyan and Penrose (2013) later argued the anomaly could be demonstrated without simulations via the internal sky-twist procedure.

The characterization is documented in the proponents' own words (CCC-E005) and matches every independent reanalysis's diagnosis (CCC-E002, E003, E004, E008); neither side disputes where the disagreement lives. The strongest opposing consideration is the proponents' accompanying assertion that concentric families "cannot be explained as a purely random effect," which their own response never quantifies against the ΛCDM ensemble — a position statement, not a rebuttal of the localization.

The proponents explicitly state that they used a pure Gaussian sky plus WMAP noise while their critics used simulations based on the ΛCDM CMB power spectrum. Independent analyses consistently report that the observed features become typical under the latter ensemble. The strongest opposing consideration is the proponents' substantive contention that the ΛCDM-spectrum ensemble is the wrong conception of chance, but that objection confirms the location of the disagreement rather than contradicting this methodological claim.

Gurzadyan & Penrose’s own response note states that the critics simulate full ΛCDM power-spectrum skies while they simulate pure Gaussian plus noise; every subsequent independent paper confirms that this ensemble choice is the locus of the significance disagreement. The strongest opposing consideration is the proponents’ philosophical claim that most of the CMB is not random and therefore ΛCDM simulations are the wrong null, but that claim is itself the hypothesis under test and cannot define the null.

Both sides document the split in the founding exchange. The proponents' own response says the critics simulate with the ΛCDM power spectrum while they simulate a pure Gaussian sky plus WMAP noise (E005), and Wehus & Eriksen locate the discrepancy in the simulated variance rather than the data (E002). Every independent circle reanalysis on the ledger (E002–E004, E008, E037) used ΛCDM-spectrum skies and found the features typical. The claim describes where the disagreement lives, not who is right; I nonetheless judge the ΛCDM-spectrum null correct because a Gaussian CMB sky is characterized by its measured angular power spectrum — a background-physics judgment, not ledger evidence. Strongest opposing consideration: the ensemble is not the only locus — the proponents flag a separate 'technical' discrepancy, Hajian's reanalysis turns on anisotropic noise modelling, Bodnia's on few-pixel artifacts, and the Warsaw group's same-harmonic-content comparison shows later proponent work moved to a spectrum-preserving null with a different statistic.

The proponents' own response states the difference: critics simulate skies with the standard model's power spectrum, while they simulate a pure Gaussian sky plus noise (CCC-E005). The reanalyses logged under CCC-C002 find the features typical under the former. This holds against three seats that sought established. Only one record is attached, and later disputes add artifact screening and large-scale conditioning, so the ensemble is the central source of disagreement but not the only one.