METHODOLOGY
How ResearchBall scouts, scores, and surfaces academic talent.
DATA SOURCE
All data on ResearchBall is sourced from Dimensions.ai, a comprehensive research information system developed by Digital Science. Dimensions links publications, grants, patents, clinical trials, datasets, and policy documents to create a connected view of the research landscape.
We query the Dimensions DSL API (Dimensions Search Language) in real time. All researcher profiles, publication records, grant histories, citation metrics, and Altmetric scores come directly from Dimensions. No data is fabricated or estimated.
RB SCORE
The RB Score is a weighted composite metric that captures the breadth and trajectory of a researcher's impact. It is calculated from six normalised components:
Average Relative Citation Ratio (RCR) across all publications, normalised against a ceiling of 3.0. RCR measures citation impact relative to field norms -- an RCR of 1.0 means average for the field, 2.0 means twice average.
Dimensions docs >Citation velocity: (recent_citations x 2) / career_years. The "recent_citations" field from Dimensions captures citations received in the most recent 2-year window. Dividing by career length normalises for seniority. Capped at 100 for scoring.
Dimensions docs >Ratio of grant funding secured in the last 3 years vs. total lifetime grant funding. Measures whether a researcher is accelerating or decelerating in funding. A researcher who just landed a major grant scores higher.
Dimensions docs >Average Altmetric Attention Score across publications, normalised against a ceiling of 50. Altmetric tracks mentions in news, social media, policy documents, patents, and other non-traditional impact channels.
Dimensions docs >Count of distinct funding organisations across all grants, normalised against a ceiling of 5. Researchers funded by multiple independent bodies demonstrate broader appeal and lower single-funder risk.
Dimensions docs >Count of distinct top-level Fields of Research (FOR) codes across publications, normalised against a ceiling of 4. Uses the ANZSRC Fields of Research classification. Interdisciplinary researchers score higher.
Dimensions docs >RB Score = round( (heat_norm x 0.30 + velocity_norm x 0.20 + grant_momentum x 0.20 + altmetric_norm x 0.10 + funder_diversity x 0.10 + cross_field x 0.10) x 100 )
RESEARCHER FLAGS
RB Score >= 20 AND the researcher's current_research_org does not match any of the ~30 hardcoded prestige institutions (Harvard, Stanford, MIT, Oxford, Cambridge, Caltech, etc.). Identifies high-impact researchers at institutions that global rankings may overlook.
Has at least one grant with an end_date within 18 months of today. Data sourced from the Dimensions grants data source. These researchers may be open to new positions, collaborations, or funding arrangements.
first_publication_year is within the last 5 years (from Dimensions researchers data source). Early-career researchers who may be on steep growth trajectories.
Recent citation velocity (annualised from the 2-year recent_citations window) exceeds 3x the career-average citation velocity (total times_cited / career years). Identifies researchers whose impact is accelerating sharply.
Publications span 4 or more distinct top-level FOR (Fields of Research) codes from the category_for classification. Indicates genuine interdisciplinary breadth rather than narrow specialisation.
PAGE-BY-PAGE BREAKDOWN
SCOUTING BOARD
Queries Dimensions for researchers where research_orgs is not empty, returning up to 20 results. For each researcher, we run two secondary queries: publications (with metrics, altmetric scores, and FOR codes) and grants (with funding amounts, dates, and funder organisations). The RB Score, flags, and all derived metrics are computed server-side.
Default sort: RB Score descending. Client-side sorting also available by War Chest, Heat Score, Momentum, and Crowd Favourite. Filters (career stage, research field, flags) are applied client-side on the returned set.
ROOKIE REPORT
Queries Dimensions for researchers where first_publication_year >= (current year - 5). Each result is enriched with publication and grant data, then scored. Default sort: citation velocity (Momentum) descending. This surfaces early-career researchers with outsized recent impact relative to their short track record.
FREE AGENTS
Queries Dimensions for grants with end_date between 2026-01-01 and 2027-12-31. Extracts unique researcher IDs from those grants, fetches full profiles, and computes scores. Each researcher gets a "days until contract ends" countdown based on their soonest-ending grant. Sorted by days remaining ascending.
HIDDEN GEMS
Uses the same search endpoint as the Scouting Board, then filters for researchers with the UNDERVALUED flag (RB Score >= 65, not at a prestige institution). Sorted by RB Score descending. The prestige institution list includes ~30 globally recognised universities.
LEAGUE TABLES
Individual rankings: all returned researchers sorted by RB Score. Institution standings: researchers grouped by current_research_org, averaged by RB Score, and sorted by average descending. Total War Chest is summed per institution.
STATS LAB (RESEARCHER PROFILE)
Fetches a single researcher by Dimensions ID, plus their full publication list (up to 100) and grant list (up to 50). Generates: Career Arc (publications per year histogram), Citation Trajectory (SVG line chart of citations by year), Grant Timeline (Gantt chart coloured by funder), and all derived metrics. Distinct research fields and funding sources are extracted and displayed.
DRAFT ROOM
Entirely client-side. Researchers added to the draft room are stored in browser localStorage. Side-by-side comparison of up to 4 researchers with proportional bar charts for each metric. "Export Scouting Report" triggers the browser print dialog.
METRIC GLOSSARY
Total lifetime grant income in British pounds.
Average RCR across all publications. 1.0 = field average, 2.0 = twice average.
Annualised recent citation velocity. Higher = accelerating impact.
Total Altmetric score. Tracks news, social, policy, and patent mentions.
Career length since first recorded publication in Dimensions.
Total number of publications indexed in Dimensions.
LIMITATIONS AND CAVEATS
- 1
The RB Score is an experimental composite metric. It is not peer-reviewed and should not be used as the sole basis for hiring, funding, or tenure decisions.
- 2
Dimensions coverage varies by discipline and geography. Researchers in the humanities, arts, and some social sciences may be underrepresented.
- 3
The current query returns a semi-random sample of 20 researchers from Dimensions. Future versions will support targeted search by name, ORCID, or institution.
- 4
Grant data may be incomplete for researchers funded by organisations not yet indexed in Dimensions.
- 5
The "prestige institution" list used for the UNDERVALUED flag is a rough heuristic and does not reflect a considered quality ranking.
- 6
All currency values are displayed in GBP as provided by the Dimensions funding_gbp field. Exchange rate normalisation may affect comparability.
ResearchBall is built on data from Dimensions.ai, a Digital Science solution.
Dimensions is the most comprehensive research grants database linked to publications, making it uniquely suited for the multi-dimensional researcher assessment that ResearchBall provides.