The best robot vacuum and mop is not a single machine we ran on a stained floor. WhichVac did not time a mopping test. The combo list is every model a source says can mop, sorted by the cleaning score. Cleaning mixes published suction, mop hardware, navigation, and obstacle sensing. If the source only says the robot mops, and does not name a pad or a roller, the mop input is a neutral 55 and the score is partial. That stop is deliberate. A missing detail must not raise the rank.
Hardware words matter. A fixed pad mostly drags. A sonic pad is cited when a source gives a scrub rate, as with the Shark Matrix Plus. Dual spinning pads are the common Roborock and Dreame description in the reviews we used, sometimes with an extending arm. A roller is a different mechanism. The eufy Omni C28 and the ECOVACS models in this catalog are cited as rollers where the source said roller. The formula gives those mechanisms different points. The gap is the category, not a claim that we watched one erase a spill faster.
Docks are where combo robots either save time or create a new chore. Dust emptying, mop washing, mop drying, and water refill are four facts. A page can confirm one and stay silent on the others. Silence stays Not published. We do not treat the word Omni, Ultra, or dock as proof of a hot wash. Hot wash is scored only when a source gives that temperature story, such as a Dreame dock described at a high wash temperature. Warm air dry and unheated air dry are different scores too. Read the spec row. The badge says cited snapshot and names the page.
Carpet is the failure mode for a mop that cannot lift. Mixed floors are normal. A 10 mm lift and a 12 mm lift are both small. They are still published numbers, so they are scored, with 12 mm as the top of that small weight. A lift with no millimeters uses 70 and is marked an assumption. If you have thick rugs, a published 10 mm may still be too low. The formula does not know your rug pile. It only knows the citation.
Suction still matters on a mopping robot because the same pass is supposed to pick up dry debris. The cap is 25,000 Pa. Above that, the suction subscore does not rise. A model with a higher brochure number is not automatically a better cleaner in this formula. A model with no pascal number contributes nothing on that input. If the rest of the cleaning score is also thin, the model is held off the numbered combo list. You can still open the product page and read every cited fact.
Navigation tells you whether the mop reaches the rooms. LiDAR, camera, LiDAR plus camera, and bump-and-bounce are the buckets. A reviewer's story that the robot "got around the furniture" is not a bucket. Obstacle avoidance is separate. Structured light, infrared, and camera recognition are scored only when named. A home test that says the robot avoided a chair leg, without naming a sensor, stays out of the obstacle ranking.
Water control is easy to over-read. One Roborock page we cited lists many flow levels. That is an app fact. It does not mean the robot is safe on unsealed wood. The same review says unsealed wood still needs care. WhichVac does not convert that into a score. It leaves the warning in the extra facts when the source wrote it.
Price is the trap on combo robots. The sticker in a September or October 2026 roundup is a snapshot. Two sources can disagree. The Dreame X50 Ultra is the example in this catalog: a review snapshot near one thousand dollars and a higher list price from another table. Both are stored. The lowest street snapshot is what "from" uses. The list price is labeled list. Neither is a live Amazon price. The Amazon button says price not recorded until the Creators API or PA-API fills it. That is an Associates rule, not a missing feature.
Value is a different question from cleaning. A combo with a strong cleaning score and a high snapshot price can lose the value list to a weaker, cheaper robot. The value page divides a performance blend by the recorded price and scales the best raw result to 100. It is marked partial while the price is a snapshot. Do not quote the value score as a lab result. Quote it as math on the inputs shown.
Maintenance after mopping is the part owners feel. Pads need washing if the dock does not wash them. Rollers need the dock path the maker designed. Tanks need filling if refill is not cited. The low-maintenance score tries to capture those cited chores. Vacuum-only models skip the mop chores instead of being punished. A combo that does not wash its own pads should not get those points. Check the wash row before you assume the dock is hands-off.
Small homes pay for dock bulk. A mop station is tall. When height is unpublished, a self-empty dock is assumed bulky and that assumption is labeled. The apartment ranking is the place to see that penalty. A combo can be the right cleaner and the wrong object for a hallway. Both pages can be true.
Compare is the tool for the last cut. Pick the combo leader, a roller model, and a spinning-pad model. The table will show blanks. Those blanks are the specs you still need from the source link. If noise matters, look at the noise row. In this catalog a decibel figure is published for the eufy 11S Max and not for most combos. Do not invent a quiet ranking from that single number. The site does not.
This guide is a draft. It is marked noindex and it is not in the sitemap. The best robot vacuum and mop order you see in the table is computed from the catalog at render time. If a mop type is wrong, send the source URL through the contact on the editorial standards page. Do not treat the draft as a final buying order until the owner approves it.
A practical way to use the page is to write down the job first. If the job is dry debris plus a light damp pass on sealed hard floors, a cited dual-spinning pad with a wash dock is enough to compare. If the job is dried spills, the formula still will not tell you which roller won a clay test, because we did not run one. Good Housekeeping described its own mopping results for one ECOVACS model. Those percentages stay in the extra facts as their lab note. They are not copied into the WhichVac score. Mixing their test into our arithmetic would pretend we measured the same floor.
Edges are the other limit. An extending pad or side brush is an extra fact when a source names it. It is not its own score. Round robots miss a strip along baseboards unless the hardware reaches. If that strip is your complaint, filter the spec table for FlexiArm, MopExtend, or an extending brush, and ignore models where that sentence was never published. Then check obstacle avoidance. A mop that swings outward can also swing into a cord. The obstacle ranking is the cross-check.
Finally, re-read the Amazon row before you click. Empty means we do not have a Creators API or PA-API price. The from-price above it is the recorded snapshot. Both can sit on the same screen without one pretending to be the other. That split is the editorial rule for every combo in the catalog, including whichever model happens to lead today.