Eight categories, fixed weights
The weights are the editorial position: what you can eat without paying extra matters most, and it isn't close. They never change from ship to ship, which is what makes any two scores on this site directly comparable.
Half the scorecard — Included Dining, Value, and half of Variety — is about what every passenger gets without spending another dollar. A ship with a dazzling paid tier and a weak included one scores like what it is: a great dining program for people willing to pay twice for dinner.
Two layers: venues and categories
Every ship gets both, and they are not derived from each other. That's deliberate, and it's the single most misunderstood thing about these scores.
Venue scores
Every restaurant, counter, food hall, and dinner show on board is scored individually, on its own merits, on that specific ship — the same venue can score differently on two sisters if the execution differs.
These power the venue explorers, the head-to-head venue tables on showdown pages, and the dining maps.
Category scores
Each of the eight categories is scored as an ecosystem judgment — how the whole roster works together — never as an average of venue scores. A category asks a different question than any single venue can answer.
These eight numbers, weighted, produce the ship's overall Seabound Dining Score.
Because the layers measure different things, a ship can win every shared venue in a head-to-head and still trail overall — it has happened, and when it does, we show you exactly why instead of hiding the tension. Both layers are always published, so every gap can be traced to its source.
Scored once, locked, and never retyped
Each ship gets one scoring pass, and the numbers it produces are locked. Every other page that mentions that ship — its class guide, every showdown it appears in, every comparison table — reuses the exact published figures. Nothing is re-derived, re-rounded, or "adjusted for context" downstream.
The overall score is pure arithmetic from there: each category score times its weight, summed, rounded to one decimal. When two ships publish the same rounded score, the underlying weighted totals usually differ by a few hundredths — and on showdown pages we print those full decimals rather than pretending a 7.640 and a 7.600 are the same number.
Re-scoring happens when the ship changes, not when the calendar does. A drydock that swaps a venue, a new evening concept, a program-wide fleet change — those trigger a fresh pass. A new marketing campaign doesn't.
Language proportional to the gap
Every comparison on this site describes score gaps with the same four words, tied to the same four ranges — checked programmatically on every page, so the prose can never inflate the math.
Identical published scores. We say so plainly — and if the full decimals differ underneath, we print them.
A real but small edge. You would be unlikely to notice it across a single sailing.
A gap you'd plausibly feel — a stronger roster, a better night, a noticeably higher ceiling.
Not a close call. One ship simply does this better, and the page says so without hedging.
We never manufacture a bigger gap than the evidence shows — and never flatten a real one to seem diplomatic. If a matchup is a coin flip, the page says coin flip. If it's a rout, the page says rout.
What a 1.0 actually means
A 1.0 is an absence floor, not a grade. When a ship genuinely has no product in a category — no suite-restricted restaurant of any kind, for example — the category locks flat at 1.0. It represents that absence directly instead of pretending to evaluate something that doesn't exist.
It sits outside the normal scale on purpose. Every other score on a ship's page reflects venues actually on board. A 1.0 doesn't — and the ship's page always explains the specific absence in plain terms, including whatever the line offers instead (suite perks, priority booking windows, a private lounge) that simply isn't a restaurant.
It is never an editorial dig at the cruise line. Some lines skip whole product categories deliberately — an egalitarian one-class dining model is a design choice, and the page says so. The floor exists so a genuinely absent product can't quietly masquerade as a mediocre one.
A worked example, from our own pages
The clearest test of a methodology is the result it can't fudge. Here's the strangest one on the site, exactly as published.
Both ships publish 7.6. Scarlet wins five of the six non-tied categories, and every single one of the thirteen identically-named venues the two ships share. By any casual read, Scarlet is the stronger ship — and yet Resilient holds the real underlying edge.
Resilient's only category win — Special Experiences, 6.9 to 3.5, because she runs an exclusive dinner show and Scarlet runs none at all — is worth more, weighted, than all five of Scarlet's wins combined. The methodology produced an uncomfortable, counterintuitive result, and we published it with the full arithmetic showing rather than rounding the tension away. Read the full showdown →
The editorial process behind the numbers
Every guide carries a visible "Last reviewed" date and notes when pricing was checked. Those dates are real — they change when a human actually re-reads the page against current facts, not on a schedule.
Claims are attributed. When a cruise line's marketing counts "20+ eateries" and our roster counts fourteen scored venues, the page says both numbers and tells you whose counting convention each one is. Site-voice claims are ours; marketing claims are labeled as the line's.
Comparisons are checked programmatically. Every gap-band label, every venue count, every weighted total on a showdown page is verified against the locked source numbers by script before publication — the prose isn't allowed to drift from the math.
And prices still change constantly. Menus, covers, packages, and reservation rules vary by sailing and change without notice. The scores tell you which ship's program is stronger; your specific sailing's details always need a current check — every page says so, because it's true.
Quick answers
Where do Seabound Dining Scores come from?
Every ship gets one scoring pass against the same locked, eight-category weighted scorecard — Included Dining (25%), Specialty Dining (20%), Variety (15%), Quality Ceiling (10%), Value (10%), Casual & Quick Dining (10%), Special Experiences (5%), Suite & Restricted Dining (5%). Category scores are ecosystem judgments made in that pass and then locked; every page that mentions the ship reuses the exact published numbers.
What does a 1.0 category score mean?
An absence floor, not a quality judgment. When a ship genuinely has no product in a category, the score locks flat at 1.0 to represent the absence directly instead of pretending to grade something that doesn't exist. The ship's page always explains the specific absence in plain terms.
Why do showdown pages call some gaps “marginal” and others “decisive”?
Language is proportional to the gap by rule: 0.0 is a tie, 0.1–0.2 is marginal, 0.3–0.5 is meaningful, 0.6+ is decisive. Every label is checked programmatically against the actual score difference, so a tenth of a point is always described as a tenth of a point.
Are category scores calculated from venue scores?
No. Venues are scored individually; categories are scored separately as judgments about the whole ecosystem. That's why a ship can win every shared venue in a head-to-head and still lose the overall — the two layers measure different things, and both are always published so any gap can be traced to its source.
How often do scores change?
When the ship changes — a drydock, a venue swap, a new evening concept — it gets a fresh scoring pass. Every guide shows a “Last reviewed” date and notes when pricing was checked, and every page reminds you to verify current details for your specific sailing, because itineraries and pricing change constantly.