METHODOLOGY & DATA STANDARDS

Numbers a scout can trust

A development report is only as good as the data underneath it. Here is exactly how SwingInsight turns sensor readings into coaching — and the provenance standard that lets college recruiters and pro scouts rely on what they see.

The data we trust

We read the sensors coaches already own — and give each one the job it does best.

HitTraxBall flight & contact
Exit velocity, launch angle, distance, hard-hit %

Our primary source for outcome metrics — when two sensors report the same measure, HitTrax wins.

Blast MotionBat sensor
Bat speed, attack angle, on-plane efficiency, rotational timing

The swing-mechanics layer: what the barrel is doing, independent of the ball.

K-Vest / K-MotionBody kinematics
Kinematic sequence, pelvis–torso separation, speed-gain ratios

The hardest-to-fake layer — how the body sequences energy into the swing.

RapsodoSpin & trajectory
Spin rate, spin direction

Scoped to what it measures best — spin — rather than duplicating contact metrics.

One source of truth per metric. When two devices report the same measure, a fixed priority order decides which wins — so a number never quietly changes depending on which sensor happened to be plugged in.

How a report is built

Four steps from raw capture to a report a parent can read — with a human in the loop.

1

Capture

Sessions arrive from the sensors above. Every reading is stored with its source, so the system always knows which device produced which number.

2

Validate

Readings are range-checked and de-duplicated. When sensors overlap, a fixed priority order decides the source of truth — no silent averaging of disagreeing devices.

3

Narrate

The AI writes the report against the validated numbers only. It is constrained to cite measured values — it cannot invent a metric the session did not produce.

4

Coach review

A coach reviews and can refine the plan before it reaches a player or parent. The numbers are the machine’s; the judgment stays human.

Provenance is the standard

Borrowed from how the pros handle data: every capture carries a trust tier, so the context behind a number travels with it.

Verified

A coach confirmed the capture in person. Highest trust — the number is locked and tamper-anchored.

Supervised

Captured under staff supervision in a known context (cage eval, showcase BP).

Self-reported

Submitted without supervision. Shown honestly as unverified — never dressed up as more.

Flagged

Inconsistent or out-of-range. Surfaced for review rather than buried.

Context, not just a number

A 95 mph exit velo off a tee in showcase BP is not the same as 95 in a game. Captures are labeled by context, so inflated showcase numbers can't masquerade as game performance.

Tamper-anchored on verify

When a coach verifies a capture, its raw data is fingerprinted and locked. A verified number can be trusted to be the number that was actually measured.

What we don't do

See it on a real profile

Every number on a SwingInsight report carries its source and trust tier.

See plansEstimate your ROI