Methodology · Published in full

The Cars-Data Score™.

A transparent 1.0–5.0 rating computed from technical specifications for over 100,000 car variants. Segment-relative, formula published, no editorial overrides, no paid placement. This page is the formula.

1. Five principles

Segment-relative, never absolute

A city car and a supercar are never scored on the same axis. Every car is ranked only against its peers: same body type × same fuel type × same price band (quintiles within the body–fuel group). If a segment has fewer than 8 members, it widens — first dropping the price band, then the fuel type — so no car is ever "best of one".

Percentile-in-class normalization

Each criterion becomes a percentile rank inside the segment (the approach recommended by the OECD–JRC Handbook on Constructing Composite Indicators). That is why a score can honestly say "top 12% of its class" — the number IS the rank. Percentiles are also immune to spec-sheet outliers.

Missing data is never guessed

If a car is missing one criterion (e.g. no boot-capacity figure), the remaining criteria of that dimension are re-weighted to sum to 100% — the gap is not filled with an estimate. If an entire dimension has no data, it is excluded and the overall re-weights across what remains. A car with no usable data gets NO score, never a fabricated one.

Weakest-link cap

Borrowed from Euro NCAP and Green NCAP: a car cannot reach the top of the scale if a core dimension is bottom-tier. If Efficiency or Value sits below the 20th percentile of its segment, the overall score is capped at 4.2. A 700 hp car with terrible economy cannot out-score everything anyway.

Popularity is never an input

Page views, search demand and brand popularity have zero weight. A viral car is not automatically a good car. The score is computed from technical specifications only.

2. Dimensions and published weights

The overall score blends five dimensions. Each dimension is itself a weighted blend of percentile-ranked criteria within the car’s segment. Criteria where lower is better (price, consumption, CO₂, 0–100 km/h time) are inverted before ranking, so a higher percentile always means better.

DimensionWeight in overallCriteria (weight within dimension)
Performance22%0–100 km/h acceleration (40%), top speed (30%), power-to-weight (30%)
Efficiency20%combined consumption l/100km; for EV & plug-in hybrid: kWh/100km derived from battery capacity ÷ range
Eco20%CO₂ emissions g/km (tailpipe — see honesty note below)
Value20%performance + practicality composite divided by new price, ranked within segment
Practicality18%seats (40%), doors (30%), boot capacity (30%)

overall = 22%·performance + 20%·efficiency + 20%·eco + 20%·value + 18%·practicality

Displayed score = 1 + 4 × overall, rounded to one decimal. If Efficiency or Value is below the 20th percentile of the segment, the overall is capped at 4.2 (the weakest-link rule). Internal-combustion and electrified cars are compared on a common energy axis: EV and plug-in hybrid efficiency is expressed as kWh/100km derived from battery capacity and range, the same approach Green NCAP uses.

3. How segments work

Every score you see includes a sentence like "top 12% of 214 diesel wagons in its price band". That is the segment: body type × fuel type × price band, where price bands are quintiles computed within each body–fuel group. Segments with fewer than 8 members widen automatically — first the price band is dropped, then the fuel type — so every ranking has enough peers to be meaningful. The segment and its size are printed next to every score; a number without its peer group would be marketing, not measurement.

4. What the score deliberately does not cover

Safety

Crash safety cannot be computed from spec sheets, and we refuse to fake it. Euro NCAP integration (official ratings with sub-scores) is on the roadmap as a separate, clearly-attributed dimension.

Reliability

Long-term reliability requires inspection or breakdown statistics (TÜV Report, ADAC Pannenstatistik). Planned as an objective, source-attributed dimension — not derivable from specs, so not in the current score.

Comfort, noise, driving feel

Subjective by nature. We will only ever cover these through transparent owner feedback, never through spec-derived proxies.

No credible overall car rating can be computed from manufacturer specifications alone — which is why this one does not claim to be an overall verdict on the car. It is the objective, spec-computable part, published with its limits stated.

5. Questions we get

Why is the scale 1.0–5.0?

The raw overall is a 0–1 weighted percentile blend, displayed as 1 + 4 × raw, rounded to one decimal. Half the scale is not wasted below "average": a mid-pack car in its segment lands near 3.0.

Why do two similar cars have different scores?

Because they compete in different segments. A diesel wagon is ranked against diesel wagons in its price band. Crossing a price-band boundary changes the peer group — which is the honest comparison, and also why we publish the segment and its size next to every score.

Is the CO₂ figure lifecycle or tailpipe?

Tailpipe only, as reported in homologation data. It deliberately does not model electricity generation, battery production or fuel supply chains — a lifecycle model would require assumptions we cannot verify per-car. The Eco dimension is labeled accordingly.

Can a manufacturer pay to improve a score?

No. The score is computed by a published formula over the same specification database everyone sees. There is no editorial override, no paid placement, and the like/dislike feedback on the badge never changes the number — it only tells us where to look for data errors.

How often is it recomputed?

On every data refresh cycle. Because the score is segment-relative, adding new cars to a segment can slightly move existing scores — exactly like class standards rising over time.

6. Version

Methodology v1 (spec-computed dimensions only), in production since June 2026. Material changes to the formula — new dimensions, weight changes, cap changes — will be documented on this page with dates. Data sourcing and editorial standards are described on the about page.

Cars-Data Score™ is a trademark of cars-data.com. The score is an independent, algorithmic rating; it is not endorsed by, or affiliated with, any vehicle manufacturer.