EntryLib: the issues that lower the reliability scores of entrylib:2026.000001@v1, by the EntryLib rubric 0.2.
For the authors and their agent: each issue lowers the scores of the claims it names, by the points given. Address what you can in a correction (a new version of the claimset) and submit it; the scores are computed again when it is published. Before submitting it, `claimset score` (or the MCP tool score) shows the scores it would have.
Scores are computed by rules from the claimset file (EntryLib rubric 0.2): they check the record, not the science, and may change as the rubric and reviews develop.
Rubric: https://github.com/entrylib/rubric/blob/main/RUBRIC.md
Page: https://entrylib.org/p/2026.000001@v1#score · Scores as JSON: https://entrylib.org/p/2026.000001@v1.score.json

PAPER: reliability 60 (medium), the score of its lowest principal claim, C6. Raising the lowest principal claim raises the paper.

ISSUES, those of the principal claims first, most costly first:

P1. Rests on argument only.
  Check: support_kind (group: support); costs up to 30 points.
  Lowers: C6 (-30).
  To fix: Support the claim with evidence the claimset records, or have its proof checked.

P2. Limitation L4 (could bias magnitude, not mitigated): smooth 1.0.8 ran from a separate worktree at tag v1.0.8-py with its own Python environment, so the versions also differ in the environment around the package.
  Check: limitation (group: open); costs up to 12 points.
  Lowers: C4 (-12).
  Concerns: L4.
  To fix: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

P3. Limitation L6 (could bias magnitude, not mitigated): The thresholds of plan PL15 were fixed after the exploratory benchmark run of auto.adam (RN45) was seen, on M4 series that run did not use.
  Check: limitation (group: open); costs up to 12 points.
  Lowers: C5 (-12).
  Concerns: L6.
  To fix: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

P4. A principal claim related to no earlier work.
  Check: no_prior_work (group: context); costs up to 5 points.
  Lowers: C3 (-5), C4 (-5), C5 (-5), C6 (-5).
  To fix: Relate the claim to the earlier work it extends, addresses or departs from.

P5. The lab record flags the test: code changed after plan.
  Check: plan_flags (group: plan); costs up to 5 points.
  Lowers: C2 (-5), C3 (-5), C4 (-5).
  To fix: Rerun the test as planned, or explain the flag in a limitation.

P6. The plan's time is anchored by a git push only.
  Check: plan_anchor (group: plan); costs up to 3 points.
  Lowers: C1 (-3), C2 (-3), C3 (-3), C4 (-3), C5 (-3), C7 (-3), C13 (-3), C14 (-3), C15 (-3), C16 (-3), C17 (-3), C18 (-3).
  To fix: Anchor plans with the platform, a timestamp authority (RFC 3161) or a registry.

P7. Assumption C24 is not checked: The coverage of a set of series is well summarised by first averaging the holdout observations within each series and then averaging over series, all horizons together.
  Check: unchecked_assumption (group: open); costs up to 3 points.
  Lowers: C1 (-3).
  Concerns: C24.
  To fix: Test the assumption, or analyse how sensitive the claim is to it.

P8. Limitation L2 (reduces precision, not mitigated): The lab record flags plans PL11, PL12 and PL13 with code changed after the plan; the only commit between those plans and their runs updated the claimset skills under .claude/skills, not the analysis scripts.
  Check: limitation (group: open); costs up to 5 points.
  Lowers: C2 (-5), C3 (-5), C4 (-2).
  Concerns: L2.
  To fix: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

P9. Limitation L3 (reduces transferability, not mitigated): ETS(MNN), the comparator whose coverage stays flat over the level terciles on M4, is badly centred there (q50 of 0.372) and has an MCE of 0.0849, so its flat cov80 does not make it well calibrated.
  Check: limitation (group: open); costs up to 1 points.
  Lowers: C2 (-1), C3 (-1).
  Concerns: L3.
  To fix: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

P10. Limitation L5 (reduces transferability, not mitigated): The auto.adam comparison ran in R, because Python's AutoADAM removes dlnorm from the candidate distributions of a pure ARIMA whatever the user passes, while R's auto.adam removes it only from its default list.
  Check: limitation (group: open); costs up to 2 points.
  Lowers: C5 (-1), C20 (-2).
  Concerns: L5.
  To fix: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

P11. Limitation L7 (reduces precision, not mitigated): The series where the selected ETS(ANN) and ARIMA(0,1,1) give the same forecast and width are 114 of 5,315.
  Check: limitation (group: open); costs up to 5 points.
  Lowers: C6 (-5), C10 (-5).
  Concerns: L7.
  To fix: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

P12. Assumption C25 is not checked: The ratio of the mean of the last third of a series to its overall mean indexes how much the scale of its errors at the forecast origin exceeds the average scale over the sample.
  Check: unchecked_assumption (group: open); costs up to 3 points.
  Lowers: C2 (-3), C3 (-3), C8 (-3).
  Concerns: C25.
  To fix: Test the assumption, or analyse how sensitive the claim is to it.

P13. A predictive claim tested without a plan recorded beforehand; the lab record shows every run.
  Check: exploratory_inference (group: plan); costs up to 5 points.
  Lowers: C21 (-5).
  To fix: Test it again under a lab-record plan or a registration, or publish the lab record that shows every run.

P14. Limitation L1 (could bias magnitude, not mitigated): The 12 wrong models of case W were chosen by the authors, not drawn from the mistakes AutoMSARIMA makes, so the size of the coverage loss they produce depends on that choice.
  Check: limitation (group: open); costs up to 12 points.
  Lowers: C16 (-12).
  Concerns: L1.
  To fix: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

P15. Limitation L8 (could bias magnitude, not mitigated): The prediction of coverage at the benchmark lengths uses Gaussian processes and interpolates the simulated coverage over log length, with lengths outside 15 to 300 at the nearest simulated size.
  Check: limitation (group: open); costs up to 12 points.
  Lowers: C21 (-12).
  Concerns: L8.
  To fix: Mitigate it with evidence, or narrow the claim's scope so that it no longer applies.

CLAIMS: score (band): issues
C1: 94 (high): P6 (-3), P7 (-3)
C2: 83 (high): P5 (-5), P8 (-5), P12 (-3), P6 (-3), P9 (-1)
C3: 78 (medium): P4 (-5), P5 (-5), P8 (-5), P12 (-3), P6 (-3), P9 (-1)
C4: 73 (medium): P2 (-12), P4 (-5), P5 (-5), P6 (-3), P8 (-2)
C5: 79 (medium): P3 (-12), P4 (-5), P6 (-3), P10 (-1)
C6: 60 (medium): P1 (-30), P11 (-5), P4 (-5)
C7: 97 (high): P6 (-3)
C8: 97 (high): P12 (-3)
C9: 100 (high): none
C10: 95 (high): P11 (-5)
C11: 100 (high): none
C12: 100 (high): none
C13: 97 (high): P6 (-3)
C14: 97 (high): P6 (-3)
C15: 97 (high): P6 (-3)
C16: 85 (high): P14 (-12), P6 (-3)
C17: 97 (high): P6 (-3)
C18: 97 (high): P6 (-3)
C19: 100 (high): none
C20: 98 (high): P10 (-2)
C21: 83 (high): P15 (-12), P13 (-5)
C22: 100 (high): none
C23: 100 (high): none