{"id":"304f760e-f89e-4ad8-8877-90fef656cdf8","arxiv_id":"2608.13269","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":5,"one_line_summary":"Target-specific stellar apsidal rates for CH Ind and SW CMa account for the reported excesses, so neither system shows a significant circumbinary-planet signal.","lead":"This paper re-checks two stars that were reported to host planets orbiting both members of a double-star system, using measured stellar sizes and internal structure. It finds the planets are not needed: ordinary stellar stretching, tides, and rotation explain the observed orbital drift.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"CH Ind consistency claim is not tested against the published 1.30±0.10 rate; the new 1.82±0.97 measurement is too imprecise, and the unpublished Thornton timings leave the discrepancy unresolved.","rationale":"The reader's weakest assumption correctly identifies the CH Ind k2 inference as a soft spot. I agree that the coeval MESA fit is imperfect and that the adopted 0.05–0.20 dex systematic is a sensitivity assumption, not a measured model error. However, I view the more load-bearing issue as the unreconciled difference between the paper's CH Ind observed rate and Thornton et al.'s published 1.30±0.10 value. The paper cannot reconstruct the latter because the timings are unpublished, and its own rate has roughly ten times larger uncertainty. This means the CH Ind consistency claim is not a strong test of either the stellar model or the original candidate; it is a weak consistency. For SW CMa the argument is much stronger: an independent target-specific prediction (0.670±0.020) agrees with the observed 0.690±0.050, and the bulk classical term is underestimated by a factor of about 16. The general conclusion that bulk searches must be followed by target-specific stellar modeling is therefore supported even if the CH Ind case is downgraded to an upper limit. The paper explicitly acknowledges the limitations, and the negative result is robust at 95% credibility under the paper's own error model. I would not change the verdict, but the CH Ind agreement should be read as a weak consistency rather than a confirmation.","tokens_in":13765,"tokens_out":13806,"duration_ms":155992,"concrete_test":"Obtain Thornton et al.'s selected CH Ind eclipse times (or re-fit the paper's 45 published eclipse times treating each event as independent with formal errors, without the common sector jitter) and recompute the six D-nodes and the slope. If the resulting observed rate is close to 1.30±0.10 and differs from the stellar prediction 2.05+0.43/-0.34 by more than 3 sigma, the CH Ind consistency claim fails; if it remains near 1.8 with ~1×10^-3 uncertainty, the no-excess claim is robust but is only an upper-limit statement.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The CH Ind branch of the central claim depends on comparing a new low-precision observed rate (Eq. 7: 1.82+0.99/-0.94 × 10^-3 deg/cycle) with a target-specific stellar prediction (Eq. 8: 2.05+0.43/-0.34). But the measurement that actually motivated the candidate, Thornton et al. (2026)'s 1.30±0.10 × 10^-3 deg/cycle, is never reconciled. The paper states it cannot reconstruct that value because the selected CH Ind timings are unpublished (Section 3; SI 1.3). With its own uncertainty of about ±1.0, the observed rate is consistent with both the stellar prediction and a null excess, so the comparison has little discriminating power. The coeval MESA fit used for k2 is also imperfect: chi^2 = 8.79 for 3 dof (p=0.032) and the more massive star is inferred to be the smaller one (Section 3; SI 2). If the true CH Ind rate is closer to Thornton's value, and the true k2 pair is lower than the inferred values by about 0.5 dex (outside the tested 0.05–0.20 dex envelope), the residual excess could become significant. The paper's no-excess conclusion may still survive, but the specific claim that target-specific stellar modeling explains the CH Ind apsidal motion is not established at the precision needed to test the original candidate.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper revisits two circumbinary-planet candidates from the bulk apsidal-motion survey of Thornton et al. (2026), using target-specific stellar models. For CH Ind, the author obtains a TESS sector-level apsidal rate of 1.82 +0.99/-0.94 x 10^-3 deg/cycle and a coeval MESA-based stellar prediction of 2.05 +0.43/-0.34 x 10^-3, giving an extra prograde rate of -0.26 +1.06/-1.04 x 10^-3. For SW CMa, the adopted observed rate 0.690 +/- 0.050 matches the published target-specific prediction 0.670 +/- 0.020, with extra rate 0.020 +0.063/-0.066. The author concludes that neither candidate shows a significant prograde excess and that the bulk survey's classical stellar terms were too small by factors of 24 and 16; a radius-sensitivity screen f_R is introduced for all 27 candidates.","tokens_in":14086,"tokens_out":14656,"duration_ms":141438,"significance":"The paper's main contribution is a convincing demonstration that bulk apsidal-motion CBP searches can misattribute classical stellar contributions, because the (R/a)^5 scaling and adopted k2 matter for evolved binaries. The SW CMa branch closes the apsidal budget almost perfectly and is strong. The CH Ind branch is a useful consistency check supported by multiple timing-robustness tests. Strengths include the explicit formula audits, injection tests, machine-readable data and reproducibility statements, and honest discussion of limitations, including the unpublished Thornton timings and the imperfect coeval stellar fit. If accepted as a validation study, it should change how such candidates are followed up.","major_comments":[{"comment":"The CH Ind comparison has limited discriminating power: the 95% non-negative upper limit of 2.28 x 10^-3 deg/cycle exceeds the nominal excess of about 1 x 10^-3 that originally motivated the candidate, so the CH Ind sector-level data alone cannot rule out a companion-induced excess at that level. The paper should state this explicitly and, for completeness, note that taking the published 1.30 +/- 0.10 x 10^-3 rate at face value would give an even more negative residual relative to the stellar prediction, also implying no prograde excess. Adding this sentence would directly address the otherwise unresolved difference between the two observed rates and would make the CH Ind branch a fairer consistency check rather than a precision test.","section":"Section 3, Eq. (7)-(8); Conclusion (i)"}],"minor_comments":[{"comment":"The f_R values quoted in the text (1.71 for CH Ind, 1.97 for SW CMa) are computed from Thornton et al.'s catalogue rates (CH Ind 1.300, SW CMa 0.850), not from the rates adopted elsewhere in the main analysis (1.82 and 0.690); the table and figure should state this clearly, or the screen should be recomputed with the adopted rates, since the current presentation is confusing.","section":"Section 5, Eq. (10), Table 5, Fig. 3"},{"comment":"The statement that the nominally more massive component is observed to be the smaller one is important; a short main-text sentence quantifying how this mass-radius reversal affects the k2 posterior would be helpful, even though the separate-star fit already gives a consistent rate.","section":"Section 3 and SI Table 2"},{"comment":"The SW CMa simplified fit has reduced chi-square 4.42 and is used only as a consistency check; the main text should say one sentence explaining why the historical subset is not combined with the adopted Claret et al. rate, since the reader otherwise sees two observed values.","section":"Section 4 and SI Section 4"},{"comment":"There are formatting placeholders such as 'MNRAS000' in the preprint, and the reference to Torres et al. 2010 should be formatted as A&ARv; these should be corrected in the final version.","section":"General"}],"recommendation":"minor_revision","confidential_remarks":"This is a credible single-author validation study with open data and transparent robustness tests. The main caveat is that the CH Ind branch is a low-precision consistency check, so the revised version should avoid any impression that it is a decisive test of the Thornton candidate. The paper is well suited to MNRAS if the requested clarifications are made."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this paper deserves a serious referee and will probably be accepted after minor revision. It does something useful and does it honestly. The SW CMa re-analysis is the strongest part: using measured dimensions and the published stellar k2, the classical term is 16x larger than the bulk estimate, and the predicted apsidal rate (0.670±0.020) agrees with the observed (0.690±0.050). The excess is 0.020±0.065, i.e. nothing. That alone is a cautionary result for bulk searches.\n\nThe CH Ind part is newer but less precise. The author measures a sector-level rate of 1.82+0.99/-0.94 from six independent TESS nodes, and predicts 2.05+0.43/-0.34 from a coeval MESA fit. The two overlap, and the excess is -0.26+1.06/-1.04. That is consistent with no planet, but it is also consistent with the original 1.30±0.10 rate from Thornton et al. The paper cannot reconstruct that rate because the underlying timings were not published, so the discriminating power is low. The stellar fit itself is imperfect too: χ²=8.79 for 3 dof (p=0.032), and the models want the more massive star to be the bigger one, contrary to observation. The author flags all of this explicitly and runs robustness checks (leave-one-out, different detrending, joint model). None of that rescues the precision, but it does mean the negative conclusion is not overclaimed.\n\nThe f_R radius-sensitivity screen is a nice lightweight addition, and the paper ships code and data on Zenodo, which is good practice. The SW CMa result is not new in the sense that Claret et al. already published it, but the independent reproduction is useful for validating the method.\n\nSoft spots, proportionately: the CH Ind measurement would need a longer baseline or published individual timings to decide whether the original candidate is truly dead. The paper is honest about this. The Thornton et al. tidal equation typo is a side observation, correctly not central to the argument.\n\nWho is this for: anyone doing apsidal-motion searches or using eclipse timings to claim circumbinary planets. It is a proper cautionary paper that belongs in the literature. I would accept it for peer review.","headline":"A careful, honest re-analysis that removes two apsidal-motion planet candidates, with SW CMa decisive and CH Ind limited by timing precision.","tokens_in":14628,"tokens_out":2730,"would_cite":true,"duration_ms":63855,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Computing stellar apsidal motion from measured binary dimensions removes the excess that suggested circumbinary planets around CH Ind and SW CMa.","keywords":["apsidal motion","circumbinary planets","eclipsing binaries","stellar interiors","apsidal-motion constant","TESS","CH Ind","SW CMa"],"falsifier":"A longer TESS baseline providing more than six independent sector nodes for CH Ind would test the observed slope: if it moved outside $1.82^{+0.99}_{-0.94}\\times10^{-3}$ deg cycle$^{-1}$ by more than the method scatter, the stellar-prediction comparison would change. Separately, independent $k_2$ values from asteroseismic modeling of CH Ind's 46 pulsation frequencies would test the adopted internal-structure constants: if the true $k_2$ pair differs from the inferred values by more than the adopted systematic, the stellar prediction could shift enough to either create a false excess or conceal a real one.","tokens_in":13544,"feed_emoji":"🪐","tokens_out":7742,"duration_ms":68906,"temperature":0.7,"pith_summary":"A bulk search of TESS eclipsing-binary timings reported 27 non-transiting circumbinary-planet candidates from excess apsidal motion, but its stellar corrections used catalogue temperatures, main-sequence relations, and a fixed $k_2=0.01$. This paper re-checks two of those candidates, CH Ind and SW CMa, whose component masses and radii have been measured in dedicated studies, by computing the stellar contribution directly. For CH Ind the target-specific prediction of $2.05^{+0.43}_{-0.34}\\times10^{-3}\\,\\mathrm{deg\\,cycle^{-1}}$ is consistent with the observed rate of $1.82^{+0.99}_{-0.94}\\times10^{-3}\\,\\mathrm{deg\\,cycle^{-1}}$, and for SW CMa the published prediction agrees with the observed rate, leaving neither system with a significant extra prograde component. The paper's message is that bulk search algorithms are useful for selecting candidates but cannot replace target-specific binary models when deciding whether a residual is planetary.","feed_headline":"Planet candidates vanish under target-specific stellar tides","feed_subtitle":"Bulk searches underestimated classical apsidal motion by factors of 16 and 24; measured binary dimensions leave no excess.","key_machinery":"The load-bearing machinery is the first-order apsidal-motion decomposition $\\dot{\\omega}_{\\rm obs}=\\dot{\\omega}_{\\rm GR}+\\dot{\\omega}_{\\rm rot}+\\dot{\\omega}_{\\rm tide}+\\dot{\\omega}_{\\rm extra}$, with the classical contribution $\\dot{\\omega}_{\\rm CL}=360\\,\\sum_i k_{2,i}(c_i^{\\rm rot}+c_i^{\\rm tide})$ containing terms proportional to $(R_i/a)^5$ and to the apsidal-motion constant $k_2$ (the internal-structure constant tabulated by stellar models, twice the fluid Love number). The differential timing variable $D(t)\\equiv[(O-C)_{\\rm pri}-(O-C)_{\\rm sec}]/2$ isolates the apsidal signal from common-mode ephemeris or LTTE trends, and the paper also introduces a radius-sensitivity screen $f_R=[(\\dot{\\omega}_{\\rm obs}-\\dot{\\omega}_{\\rm GR})/\\dot{\\omega}_{\\rm CL,bulk}]^{1/5}$ that flags candidates where an incorrect luminosity class or evolutionary state could dominate the claimed excess.","core_discovery":"The central claim is that the two candidate circumbinary planets identified by the bulk search are not supported once the stellar contributions are computed from measured binary parameters. For CH Ind, a coeval fit to the measured component masses, radii, temperatures, and rotation yields $\\dot{\\omega}_{\\rm stars}=2.05^{+0.43}_{-0.34}\\times10^{-3}$ deg cycle$^{-1}$, compared with the observed $\\dot{\\omega}_{\\rm obs}=1.82^{+0.99}_{-0.94}\\times10^{-3}$ deg cycle$^{-1}$; the extra component $\\dot{\\omega}_{\\rm extra}=-0.26^{+1.06}_{-1.04}\\times10^{-3}$ deg cycle$^{-1}$ is consistent with zero. For SW CMa, the observed rate $0.690\\pm0.050\\times10^{-3}$ deg cycle$^{-1}$ matches the target-specific prediction $0.670\\pm0.020\\times10^{-3}$ deg cycle$^{-1}$. The paper therefore concludes that the classical stellar terms in the bulk search were too small by factors of 24 and 16, because the candidate stars are evolved with radii of 2.5–3.0 $R_\\odot$ rather than unevolved main-sequence values.","pith_inferences":["Inference: The radius-sensitivity statistic $f_R$ could be used as a ranking metric for follow-up of the entire candidate list, with resources directed first to systems where a modest radius revision could erase the excess (e.g., $f_R\\lesssim1.2$); the paper suggests this prioritization but does not fully develop it.","Inference: An independent test of the CH Ind $k_2$ values via asteroseismology (the system pulsates with 46 frequencies) could confirm or refute the adopted internal-structure constants without waiting for a longer timing baseline.","Inference: Extending the $f_R$ screen to a full grid in $k_2$ and rotation could turn it from a screening statistic into a prior for Bayesian model comparison of third-body versus stellar hypotheses."],"forward_implications":["CH Ind and SW CMa should be removed from the list of secure (or even probable) circumbinary-planet candidates unless new data revive the excess.","The remaining 25 candidates need the same target-specific treatment: measured masses, radii, rotations, and evolutionary-state-dependent $k_2$ values before a third-body interpretation.","The $f_R$ screen (median 1.97, with 14 of 26 finite cases requiring radius scale factors above 2.0) identifies which candidates are most vulnerable to stellar-radius errors.","A validation hierarchy emerges: component dimensions first, rotation bounds second, a coeval pair of $k_2$ values third, then a covariance-aware timing analysis whose independent units are observing sectors, not individual eclipses.","The reported CH Ind LTTE (2.0 min, 10056 d period) cannot account for the nominal apsidal excess: a circular, nearly coplanar companion at that radius would contribute only about $1.6\\times10^{-6}$ deg cycle$^{-1}$, roughly 600 times below the claimed excess."],"supporting_citations":[{"why":"The bulk search whose candidate list and classical estimates are re-evaluated here.","marker":"Thornton et al. (2026)"},{"why":"Provides CH Ind component masses, radii, eccentricity, and pulsation frequencies used for the coeval fit and the observed-rate comparison.","marker":"Liakos (2025)"},{"why":"Gives SW CMa component masses, radii, and rotations, plus the coefficient-weighted theoretical $\\bar{k}_2$ value.","marker":"Torres et al. (2012)"},{"why":"Supplies the published target-specific apsidal prediction and observed rate for SW CMa, and the stellar-model grid from which its $\\bar{k}_2$ is drawn.","marker":"Claret et al. (2021)"},{"why":"The evolutionary tables from which CH Ind's individual $k_{2,1}$ and $k_{2,2}$ are inferred at a shared age and metallicity.","marker":"Claret (2023)"},{"why":"The classical apsidal-motion formula defining the tidal and rotational contributions.","marker":"Sterne (1939)"},{"why":"Pseudo-synchronous rotation prescription checked for SW CMa when testing rotation assumptions.","marker":"Hut (1981)"}],"fun_headline_variants":["Stellar tides erase two circumbinary planet candidates","Target-specific tides cancel planet signals in binaries","Excess apsidal motion vanishes with accurate stellar models","No planets after correcting binary tidal models","Bulk planet candidates fail stellar-tide test"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"That the two apsidal-motion constants $k_2$ inferred for CH Ind's components are accurate within the adopted 0.05–0.20 dex common uncertainty, even though the coeval stellar model fit does not perfectly reproduce the measured radii and temperatures ($\\chi^2=8.79$ for 3 degrees of freedom, $p=0.032$, and the more massive star is observed to be the smaller one).","fun_headline_variants_meta":{"raw":{"variants":["Stellar tides erase two circumbinary planet candidates","Target-specific tides cancel planet signals in binaries","Excess apsidal motion vanishes with accurate stellar models","No planets after correcting binary tidal models","Bulk planet candidates fail stellar-tide test"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000564,"raw_usage":{"total_tokens":2753,"prompt_tokens":1101,"completion_tokens":1652,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":717,"completion_tokens_details":{"reasoning_tokens":1581}},"tokens_in":717,"tokens_out":1652,"duration_ms":12474,"temperature":1.0,"reasoning_tokens":1581,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:57:58.486958+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A longer TESS baseline providing more than six independent sector nodes for CH Ind would test the observed slope: if it moved outside $1.82^{+0.99}_{-0.94}\\times10^{-3}$ deg cycle$^{-1}$ by more than the method scatter, the stellar-prediction comparison would change. Separately, independent $k_2$ values from asteroseismic modeling of CH Ind's 46 pulsation frequencies would test the adopted internal-structure constants: if the true $k_2$ pair differs from the inferred values by more than the adopted systematic, the stellar prediction could shift enough to either create a false excess or conceal a real one.","supporting_citations":[],"review_version":1}