{"id":"2cddbb3d-7b6a-4cbc-8837-636d5780253d","arxiv_id":"2505.20035","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Using 217 small-planet systems and 1167 comparison stars, the authors find cold Jupiter occurrence in small-planet systems (11.1%) matches the field rate (9.8%), with a possible super-solar excess below 3 sigma.","lead":"This paper measures how often stars that host small, close-in planets also host distant giant planets, using the largest sample to date. It finds no strong overall correlation, except a possible excess in massive, metal-rich stars, and a link between giant-planet eccentricity and small-planet multiplicity.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Selection bias in the SP sample: systems with known or suspected outer companions are more likely to have the RV data required by Sect. 2.1, potentially inflating f_CJ|SP and explaining the marginal super-solar correlation.","rationale":"The reader's weakest assumption matches my own: the SP sample's RV-data requirement is a selection on observability of outer companions, not a random draw from SP hosts. This is load-bearing because the central comparison is between f_CJ|SP and f_CJ, and any asymmetry in target selection directly biases the ratio. The paper's own split into transit and RV subsamples (Table E.1) shows f_CJ|SP,RV=13.4% vs f_CJ|SP,transit=9.5%, a gap that, while within errors, is in the direction expected if RV-selected systems are preferentially monitored for additional signals. The discovery-date test is decisive: if many CJs were known before the SP, the sample construction is clearly biased. Even if the null result (no strong correlation) is robust, the specific 2.5 sigma super-solar excess in Sect. 4.3 would not be. Because the reader already flagged this assumption and issued CONDITIONAL, my recommendation is UNCHANGED.","tokens_in":19261,"tokens_out":4693,"duration_ms":76530,"concrete_test":"For each of the 23 SP systems with a CJ in the 0.5-20 M_Jup range, compare the discovery dates of the CJ and the SP from the NASA Exoplanet Archive. If a substantial fraction (e.g., >40%) of the CJs were announced before the SP, then many systems entered the sample because they were already known giant-planet hosts, directly demonstrating selection bias. As a quantitative check, recompute f_CJ|SP after removing systems where the CJ was discovered before the SP; if the super-solar excess in Table D.2 drops below ~1 sigma, the correlation claim is not robust.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing assumption is that the SP sample, defined by requiring at least one public RV dataset with more than 20 measurements and a baseline of at least 1 year (Sect. 2.1), is representative of all SP systems. This criterion is not passive: for transit-discovered SPs, the presence of a suspected outer companion or an RV trend is precisely what motivates extended RV monitoring, so systems with CJs are more likely to satisfy the criterion. The comparison sample (Sect. 2.2) is drawn from blind RV surveys whose targets were chosen without knowledge of planets, so an asymmetric bias would inflate f_CJ|SP relative to f_CJ. The K-S tests in Sect. 2.3 address stellar parameter distributions, not this target-selection effect, and the super-solar excess at 2.5 sigma (Sect. 4.3, Table D.2) rests on exactly the bins where metal-rich stars might be preferentially followed up. If this bias is present, the central null (f_CJ|SP ~ f_CJ) survives, but the claimed mass-metallicity correlation could be entirely spurious.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper measures the occurrence rate of cold Jupiters (CJs, a=1-10 AU, M=0.5-20 MJup) around solar-type stars that host short-period small planets (SPs, P<100 d, 1-20 MEarth), using a sample of 217 SP systems with public RV data and a comparison sample of 1167 stars from the AAT, CLS, and HARPS surveys. Using binomial statistics with injection-recovery completeness, the authors find f_CJ|SP = 11.1^{+2.5}_{-1.8}%, consistent with f_CJ = 9.8^{+0.9}_{-0.8}%, and a possible excess only at super-solar mass and metallicity (2.5sigma). They also find no correlations with SP multiplicity, CJ multiplicity, or SP composition, but report that SP multiplicity is lower around eccentric CJs. The paper concludes that SPs and CJs are mostly independent, with a possible exception in massive metal-rich stars.","tokens_in":19449,"tokens_out":8069,"duration_ms":88019,"significance":"If correct, this is the largest homogeneous comparison to date of CJ occurrence in SP and non-SP hosts, and it strengthens the case that inner small planets and outer giants form largely independently. The use of three blind RV surveys merged into one sample, the consistent application of completeness corrections, and the public data make the analysis reproducible. However, the central comparison is threatened by a selection effect in the SP sample, which the paper does not quantify; the claimed super-solar correlation in particular rests on this assumption.","major_comments":[{"comment":"The SP sample selection in §2.1 requires at least one public RV dataset with more than 20 measurements and baseline ≥ 1 yr. This criterion is not selection-neutral: for transit-discovered SPs, extended RV monitoring is often triggered by suspected outer companions or RV trends, so systems with CJs are more likely to satisfy the criterion. The K-S tests in §2.3 compare stellar metallicity and mass distributions between the SP and comparison samples, but they do not address this target-selection effect. If the bias is present, f_CJ|SP is inflated, and the 2.5σ excess at super-solar mass and metallicity reported in §4.3 (Table D.2) could be entirely spurious. The authors should demonstrate that the SP sample is representative by, e.g., comparing f_CJ|SP between transit-detected SP systems with and without the required RV data, or by explicitly modeling the probability of having such data as a function of known companion status.","section":"§2.1, §2.3, §4.3, Table D.2"},{"comment":"The comparison in §4.3 is made without matching the stellar parameter distributions of the SP and merged RV samples. Section 2.3 reports a K-S p-value of 3e-9 for stellar mass and p<0.01 for metallicity within the 1.0-1.2 Msun bin, indicating that the two samples are not drawn from similar parent distributions. Because f_CJ increases with both stellar mass and metallicity, the higher f_CJ|SP in the super-solar mass and metallicity bin could reflect these distributional differences rather than a genuine SP-CJ correlation. The paper should either reweight the comparison sample to match the SP sample's stellar parameter distribution or restrict the comparison to a matched subsample, and report the mean [Fe/H] and M_star within each bin to allow the reader to assess this.","section":"§2.3, §4.3, Table D.2"}],"minor_comments":[{"comment":"The description of the merged comparison sample says '1167 stars, 213 of which are common to two or all three surveys'; the paper should state explicitly whether the merged sample is the union of unique stars and whether each star is counted once in the binomial statistics, since double-counting common stars would distort the effective sample size.","section":"§2.2"},{"comment":"The binomial formula uses N_star,eff = N_star * C with an average completeness; this ignores the star-to-star variance in completeness, and the authors should comment on whether a hierarchical treatment would change the quoted uncertainties.","section":"§3, Eq. (1)"},{"comment":"The manuscript quotes a 1.7σ significance for the difference between f_CJ|SP and f_CJ at super-solar metallicity, but it does not specify how the significance was computed; please state the test statistic (e.g., a two-proportion z-test or a bootstrap).","section":"§4.2"},{"comment":"The color bars are difficult to read in the printed version; please increase the font size or adjust the layout so that the completeness scale is legible.","section":"Figure B.1"},{"comment":"The caveat that completeness values are optimistic because they ignore uncorrelated jitter is useful, but a quantitative estimate of the typical jitter effect on f_CJ would help the reader gauge the impact.","section":"§5.2"}],"recommendation":"major_revision","confidential_remarks":"The authors should be asked to address the selection-bias concern before publication, perhaps by a sensitivity analysis restricted to transit-detected SPs with uniform RV follow-up. The paper is otherwise within the scope of A&A and would be a useful contribution once the selection issue is clarified."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my read. The paper does the field a service: it recomputes the cold-Jupiter occurrence rate around small-planet hosts with a larger, more carefully vetted sample (217 SP systems vs 184 in BL24) and a merged comparison sample of 1167 stars from AAT, CLS, and HARPS. The integrated f_CJ|SP = 11.1% is consistent with f_CJ = 9.8%, which effectively retires the old claims of ~40%. The completeness injection-recovery is done properly and the consistency across the three RV surveys is reassuring. The tables are thorough and the authors are appropriately cautious about the sub-3-sigma super-solar excess; they don't oversell it.\n\nTwo soft spots. First, the selection-bias concern raised by the stress-test is legitimate, though not fatal. Requiring public RVs with >20 points and baseline >1 yr plausibly enriches for systems already suspected of hosting outer companions, which would inflate f_CJ|SP. That wouldn't threaten the central null—it would strengthen it—but it could produce a spurious excess at super-solar mass/metallicity if metal-rich stars get preferentially followed up. The K-S tests address stellar parameter distributions, not this target-selection effect, and the paper doesn't discuss it. Second, the eccentricity-multiplicity trend in Sect. 4.6 rests on 27 systems and is presented as a dynamical expectation. It's visually suggestive but there's no statistical test, so claiming SPs are 'not indifferent' to CJs on this basis is overreach for now.\n\nNone of this changes my overall view: the central result is probably correct, and the paper is transparent about its limitations. It deserves peer review. The authors should be asked to add a sensitivity analysis or at least a discussion of the RV-data selection effect before publication.\n\nWho is it for: people working on occurrence rates, planet formation, or RV survey design. I'd cite it.","headline":"A careful, larger-sample recomputation that settles the cold-Jupiter occurrence rate around small-planet hosts near ~11%, with a plausible but sub-3-sigma excess at super-solar mass/metallicity and an under-powered eccentricity-multiplicity trend.","tokens_in":20000,"tokens_out":3570,"would_cite":true,"duration_ms":39907,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Cold Jupiters and small planets are mostly unrelated, census shows","keywords":["cold Jupiters","small planets","exoplanet occurrence rates","planet formation","radial velocity surveys","stellar metallicity","planetary system architecture","exoplanet demographics"],"falsifier":"A decisive test is to recompute $f_{\\rm CJ|SP}$ using only systems in which the radial-velocity monitoring was planned without any prior hint of an outer companion—for example, targets selected purely by transit detection. If the rate in that unbiased subsample matches $f_{\\rm CJ}$ in every mass-metallicity bin, the tentative super-solar excess is a selection artifact; if it exceeds $f_{\\rm CJ}$ at super-solar mass and metallicity at more than $3\\sigma$, the correlation is real.","tokens_in":19080,"feed_emoji":"🪐","tokens_out":12061,"duration_ms":116416,"temperature":0.7,"pith_summary":"Does a star's short-period small planets tell you anything about whether it also hosts a distant cold Jupiter? This paper answers with a large homogeneous comparison: the cold-Jupiter rate in 217 small-planet systems is $f_{\\rm CJ|SP}=11.1^{+2.5}_{-1.8}\\%$, statistically the same as the $9.8^{+0.9}_{-0.8}\\%$ rate measured in a comparison sample of 1167 solar-type stars from blind radial-velocity surveys. The only deviation appears in stars with super-solar mass and metallicity, where the rate is higher but the excess is below $3\\sigma$. The authors conclude that small planets and cold Jupiters are mostly indifferent to each other, with a possible exception in massive metal-rich stars, and that the two populations' occurrence rates cannot be used to argue for strong formation coupling.","feed_headline":"Cold Jupiters and small planets are mostly unrelated, census shows","feed_subtitle":"The largest homogeneous comparison finds no correlation, except a tentative one in massive metal-rich stars.","key_machinery":"The analysis rests on paired samples and a per-star completeness correction. The small-planet sample is drawn from public catalogs using fixed criteria (confirmed planet mass, solar-type host, public radial velocities with more than 20 epochs spanning at least one year), and the comparison sample is a merged set of stars from three large radial-velocity surveys selected under the same stellar criteria. For each star, the sensitivity to cold Jupiters is computed by injecting synthetic Keplerian signals on a $30\\times30$ grid in planet mass and semimajor axis, drawing eccentricities from a $\\beta$ distribution, and counting recoveries with a model-selection criterion; the average recovery rate converts the raw star count into an effective number of stars per bin. Occurrence rates are then derived from binomial statistics, and differences between rates are evaluated across metallicity and mass bins, with two-sample distribution tests used to check that the samples being compared are not too dissimilar.","core_discovery":"On its own terms, the paper's discovery is that the occurrence of cold Jupiters does not depend on whether a system also has short-period small planets, once selection effects are handled homogeneously. Using 217 systems with confirmed small planets and a merged comparison sample of 1167 stars from three radial-velocity surveys, it obtains an integrated rate $f_{\\rm CJ|SP}=11.1^{+2.5}_{-1.8}\\%$ at average stellar metallicity and mass, fully consistent with $f_{\\rm CJ}=9.8^{+0.9}_{-0.8}\\%$ in the comparison sample. Broken down by stellar mass and metallicity, only the bin with $M_\\star\\ge1.0\\,M_\\odot$ and $[\\mathrm{Fe/H}]>0.1$ shows an excess of cold Jupiters in small-planet systems, at $2.5\\sigma$ for the $0.5$--$20\\,M_{\\rm Jup}$ mass range. The paper further finds no correlation with small-planet multiplicity, cold-Jupiter multiplicity, or small-planet composition, but does find that the multiplicity of inner small planets drops when the cold Jupiter is eccentric, as dynamical theory predicts.","pith_inferences":["If future larger samples confirm the indifference, formation models that predict a strong dynamical coupling between inner small planets and outer giants—such as the pebble-isolation barrier—would need to be revised or restricted to specific disk conditions.","A testable extension the authors do not pursue: apply the same completeness machinery to finer composition classes (rocky with thin envelopes versus water worlds), since the bulk-density degeneracy they note could hide an architecture-composition link.","The current $11.1\\%$ rate is a lower bound in practice: the analysis counts only confirmed cold Jupiters, and three short-baseline small-planet systems show long-term radial-velocity trends that could turn out to be cold Jupiters; if so, the integrated rate would rise.","The tentative super-solar excess predicts that a future sample restricted to $M_\\star>1.0\\,M_\\odot$ and $[\\mathrm{Fe/H}]>0.1$ will show $f_{\\rm CJ|SP}$ rising with metallicity; if it stays flat, the excess was a fluctuation."],"forward_implications":["The previously reported $\\sim40\\%$ occurrence of cold Jupiters in small-planet systems is not reproduced; the new integrated rate is $11.1^{+2.5}_{-1.8}\\%$.","Any SP-CJ correlation, if real, is confined to super-solar mass and metallicity and is currently below $3\\sigma$; the default conclusion is that the two populations form independently.","The absence of SP-CJ trends with small-planet multiplicity, cold-Jupiter multiplicity, and small-planet composition weakens anticorrelation models in which cold Jupiters block the inward migration or growth of small planets.","The strong drop in inner-planet multiplicity around eccentric cold Jupiters means that even at equal occurrence rates, the architecture of a small-planet system is shaped by its outer giant."],"supporting_citations":[{"why":"Supplies the sample-selection criteria and the earlier 2.7σ claim of an SP-CJ correlation at super-solar metallicity that this paper re-tests and extends.","marker":"BL24"},{"why":"Provides the injection-recovery completeness method and a previous lower estimate of fCJ|SP that this work compares against.","marker":"B23"},{"why":"Provides one of the three blind radial-velocity surveys merged into the comparison sample for fCJ.","marker":"Rosenthal et al. 2021"},{"why":"Provides a second blind radial-velocity survey used to estimate fCJ in the comparison sample.","marker":"Wittenmyer et al. 2020"},{"why":"Provides the third blind radial-velocity survey used to estimate fCJ in the comparison sample.","marker":"Mayor et al. 2011"},{"why":"Supplies the stellar mass and metallicity values used to bin both the small-planet and comparison samples.","marker":"Hinkel et al. 2014"},{"why":"Reported a ~40% occurrence rate of cold Jupiters in small-planet systems, the high estimate this paper's 11.1% rate contradicts.","marker":"Bryan et al. 2019"},{"why":"Reported a similarly high rate and introduced the conditional-probability framework fSP|CJ used here.","marker":"Zhu & Wu 2018"},{"why":"Provides the Generation 3 Bern model predictions of weak or no SP-CJ correlation and an architecture-composition link that the paper tests.","marker":"Schlecker et al. 2021"},{"why":"Reports a positive SP-CJ correlation at super-solar mass and metallicity above 2σ; this paper's 2.5σ excess in the same bin is compared directly to it.","marker":"BL25"}],"fun_headline_variants":["Cold Jupiters and small planets: no strong link in largest census","Cold Jupiters don't favor systems with small planets, study finds","Only in massive metal-rich stars do cold Jupiters co-occur with small planets","Largest homogeneous sample finds cold Jupiters indifferent to small planets","Eccentric cold Jupiters shape inner small-planet systems"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the 217 small-planet systems are representative of all small-planet systems, with no bias toward stars already known or suspected to host an outer cold Jupiter; if the sample is biased that way, the measured $11.1\\%$ rate is inflated and the comparison against blind surveys is unfair.","fun_headline_variants_meta":{"raw":{"variants":["Cold Jupiters and small planets: no strong link in largest census","Cold Jupiters don't favor systems with small planets, study finds","Only in massive metal-rich stars do cold Jupiters co-occur with small planets","Largest homogeneous sample finds cold Jupiters indifferent to small planets","Eccentric cold Jupiters shape inner small-planet systems"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000942,"raw_usage":{"total_tokens":4203,"prompt_tokens":1303,"completion_tokens":2900,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":919,"completion_tokens_details":{"reasoning_tokens":2804}},"tokens_in":919,"tokens_out":2900,"duration_ms":21129,"temperature":1.0,"reasoning_tokens":2804,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T14:00:16.614172+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A decisive test is to recompute $f_{\\rm CJ|SP}$ using only systems in which the radial-velocity monitoring was planned without any prior hint of an outer companion—for example, targets selected purely by transit detection. If the rate in that unbiased subsample matches $f_{\\rm CJ}$ in every mass-metallicity bin, the tentative super-solar excess is a selection artifact; if it exceeds $f_{\\rm CJ}$ at super-solar mass and metallicity at more than $3\\sigma$, the correlation is real.","supporting_citations":[{"cited_title":"A., Wang , S., Horner , J., et al","cited_arxiv_id":null,"evidence_quote":"Provides a second blind radial-velocity survey used to estimate fCJ in the comparison sample."},{"cited_title":"2021, , 656, A71","cited_arxiv_id":null,"evidence_quote":"Provides the Generation 3 Bern model predictions of weak or no SP-CJ correlation and an architecture-composition link that the paper tests."}],"review_version":1}