REALFIT ecosystem layered stack Four tiers: REALFIT as governing authority at top, then Apps layer, then Score Sets layer, then FITDAT open library at the base. REALFIT® — standards authority Governs FITDAT · Issues certification · Maintains versioning · Publishes standards Layer 3 — Apps What athletes, coaches, and organizations use REALFITSCORE.com Flagship · all FITDAT tests ScoreMy40.com REALFIT app · 40yd Developer apps Certified or independent Layer 2 — Score Sets Scoring logic, weights, curves, and reference bands — proprietary or open REALFIT Absolute Score™ Proprietary · 0–1000 REALFIT flagship Public standards Army AFT · Marine PFT Published by REALFIT Developer Score Sets Custom curves + weights Cert. available Layer 1 — FITDAT® open library The open foundation — test definitions, protocols, normalization rules, fitness components Growing library of fully specified tests · 6 core fitness components · expanding replacement groups Open · Versioned · Immutable once published · Free to build on · Governed by REALFIT® FITDAT defines what was measured and how. Score Sets decide what it means. Users Builders Foundation
Layer 0
REALFIT®
The governing authority
Sets the rules for everything below it
+ expand

REALFIT® is the standards and certification authority for objective physical fitness measurement. It does not build apps or scoring engines directly — it sets the rules that everything else must follow.

REALFIT governs FITDAT by defining what qualifies as a test, what constitutes a complete specification, and what versioning rules apply. It issues REALFIT Certification to Score Sets and apps that implement FITDAT correctly and meet reproducibility standards. Certification is auditable and revocable.

What REALFIT does not do: It does not dictate training philosophy, coaching methodology, or how individuals should interpret their scores. That is the job of Score Sets and Reference Bands.

Responsibilities

  • Maintains and versions the FITDAT library
  • Sets specification requirements for new tests
  • Issues and revokes REALFIT Certification
  • Publishes proprietary Score Sets (REALFIT Absolute Score™)
  • Publishes open standard implementations (Army AFT, Marine PFT)

What this means for developers

  • FITDAT definitions will not change without versioning
  • Your Score Set can reference specific FITDAT versions
  • Certification gives your Score Set institutional credibility
  • The library is governed — not crowdsourced
Layer 1
FITDAT®
Open test library — the foundation
Defines what was measured and how. Never interprets results.
+ expand

FITDAT is the open infrastructure layer. It defines tests, protocols, normalization rules, fitness components, and replacement groups. Every definition in FITDAT is fully specified before publication — there are no placeholder entries or partial specs.

The key architectural principle: FITDAT defines facts, not judgments. A measurement of 4.55 seconds in the 40-yard dash is a fact. Whether that is "good" is the job of a Score Set, not FITDAT. This separation is intentional — it means the same FITDAT test definition can be used by a general population Score Set, an NFL Combine Score Set, and a military readiness Score Set simultaneously, each applying their own interpretation without changing the underlying measurement definition.

What FITDAT publishes per test

  • TestDefinition — construct, unit, direction
  • TestProtocol — execution, equipment, precision
  • TestContext — sports, populations, settings
  • NormalizationRule — unit conversion formulas
  • ReplacementGroup — structural equivalence
  • FitnessComponent — primary and secondary

What FITDAT does not publish

  • Scoring curves or points tables
  • Component weights or aggregation logic
  • Reference bands or performance labels
  • Pass/fail thresholds
  • Population-specific interpretations
Layer 2
Score Sets
Scoring logic built on FITDAT
Applies curves, weights, and thresholds to FITDAT measurements
+ expand

A Score Set takes FITDAT test definitions as inputs and applies its own scoring logic: curves, component weights, aggregation rules, and reference bands. Score Sets are where interpretation lives. The same FITDAT measurement means different things to different Score Sets — and that is by design.

REALFIT publishes three types of Score Sets:

Proprietary REALFIT Score Sets — the REALFIT Absolute Score™ is the flagship. Scoring curves and component weights are proprietary to REALFIT and not published in FITDAT. Available via REALFITSCORE.com.

Public standard implementations — REALFIT publishes versioned, reproducible implementations of publicly available standards like the Army Fitness Test and Marine Corps PFT. These use FITDAT test definitions where they exist and flag gaps where new tests need to be specified.

Developer Score Sets — any developer can build a Score Set using FITDAT definitions. Score Sets must declare their FITDAT version references, calculation models, and include at least one Reference Band. REALFIT Certification is available for Score Sets that meet specification requirements.

Every Score Set must declare

  • Which FITDAT tests and versions it uses
  • The calculation model (points table, formula, hybrid)
  • Component weights and aggregation method
  • At least one Reference Band
  • Its own version number

Score Set types

  • REALFIT Absolute Score™ — proprietary flagship
  • Army Fitness Test (AFT) — public standard
  • Marine Corps PFT — public standard
  • NFL Combine — professional benchmark
  • Developer custom Score Sets
Layer 3
Apps
What users actually touch
Consumer apps, developer tools, institutional deployments
+ expand

Apps are the user-facing implementations of Score Sets. They collect measurements, apply Score Set logic, and present results. Apps may be built by REALFIT or by third-party developers.

REALFIT apps are built and maintained by REALFIT directly. REALFITSCORE.com is the flagship — a web app where anyone can score all 14 FITDAT tests online and receive their REALFIT Absolute Score. ScoreMy40.com is a focused single-test app for the 40-yard dash.

Certified developer apps implement FITDAT definitions accurately, declare their Score Set version, and reference StandardSets correctly. REALFIT Certification is auditable and revocable — it signals to users that the app's results are reproducible and governed.

REALFIT apps

  • REALFITSCORE.com — full battery, Absolute Score
  • ScoreMy40.com — 40-yard dash with race comparison
  • iOS / Android apps — in development

For developers building apps

  • Implement FITDAT test definitions exactly
  • Declare Score Set version in your UI
  • Reference StandardSets correctly
  • Apply for REALFIT Certification via realfit.com

How a measurement flows through the stack

From a physical result to a scored, reproducible output — every step is declared and versioned.

01

Test is performed using a FITDAT protocol

The athlete runs 40 yards. The administrator uses proto_40yd_dash_v1 — standing start, electronic timing, nearest 0.01 seconds. The protocol is fully specified and versioned. Any administrator anywhere using this protocol produces a comparable result.

02

Result is recorded as an immutable measurement

4.55 seconds. The measurement is a fact — it references test_40yd_dash, proto_40yd_dash_v1, and the canonical unit (seconds). It cannot be changed retroactively.

03

Normalization rule is applied if needed

The 40-yard dash input is already in seconds — no normalization needed. For a Bench Press result entered in lbs, norm_lbs_to_kg converts it to the canonical unit before scoring.

04

Score Set applies its scoring curve

The REALFIT Absolute Score Score Set maps 4.55 seconds to a score using its proprietary scoring table. The Score Set version is declared in the output. A different Score Set (e.g. NFL Combine) applies its own positional benchmarks to the same 4.55-second measurement.

05

Reference Band provides human interpretation

The score maps to a Reference Band label — "Near elite", "NFL starter speed", "Above D1 average". Reference Bands are non-scoring interpretive layers. They do not change the underlying score.

06

Output is reproducible and auditable

Any system using the same FITDAT test version, the same protocol, and the same Score Set version will produce the same result for the same input. That's the whole point — fitness data that compounds and scales.

Building on FITDAT as a developer

Four steps from idea to certified Score Set.

Step 1

Browse the library

Choose which FITDAT tests your Score Set will use. Each test has a full specification — definition, protocol, normalization rule, and context tags.

Step 2

Build your Score Set

Define your scoring curves, component weights, aggregation method, and at least one Reference Band. Declare which FITDAT test and protocol versions you reference.

Step 3

Test with the demo

Use the FITDAT scoring demo to verify your implementation produces correct canonical values and that your Score Set logic applies correctly.

Step 4

Apply for certification

Submit your Score Set specification to REALFIT for certification review. Certified Score Sets are listed in the REALFIT ecosystem and carry the REALFIT Certified mark.

REALFIT Certification

REALFIT Certification is available to any Score Set or application that correctly implements FITDAT definitions, declares its versions, and meets reproducibility standards. Certification is not an endorsement of scoring philosophy — it is a technical attestation that the implementation is accurate and auditable.

Certification requirements: Implement FITDAT test definitions exactly as specified. Declare Score Set version, FITDAT version references, and protocol IDs in all outputs. Include at least one Reference Band. Pass REALFIT's reproducibility audit.

Certification is revocable if an implementation is found to be inaccurate or non-reproducible. The certification status of all Score Sets and apps is publicly auditable via REALFIT.com.