Data Source Label
Also known as: DataSource, provenance label, source label
A Data Source Label is the badge shown alongside every metric in SwingVantage that identifies where that data came from. This means Measured, Estimated, AI Interpreted, Placeholder, or Real. This means. As a result, you always know how to weight it.
Every number in SwingVantage carries a provenance label. This system — called DataSource in the codebase — is a core honesty rule, not a cosmetic detail. Labels prevent a natural human tendency: treating all numbers on a screen as equally authoritative. A value labeled 'Measured' was observed by a sensor. 'Estimated' was inferred from video. 'AI Interpreted' emerged from AI reasoning. 'Placeholder' is a demo value that will be replaced with real data. 'Real / Live' confirms the data is from an actual session in real-time sync. No label is ever left off.
Focus first on the [M] or [E] badges — these are the most actionable. Ignore [P] (placeholder) values until you have real session data. They exist to show what a metric looks like before your first upload, not to guide practice decisions.
When exporting data or sharing a report with a coach, always include the data source labels. This means strip them out and a reader may treat estimates as measurements. This misrepresents your data quality and can lead to misdirected coaching decisions.
Example
A report row shows: 'Ball Speed — 147 mph [Measured]' and below it 'Launch Angle — ~12.5° [Estimated]' — the contrast is explicit.
Why it matters
Labels make the difference between an honest tool and a misleading one. You cannot make good decisions from data you do not understand.
How it shows up on video
In Motion Lab, scan the badge column in the metrics panel before reading any values. This means green [M] = measured, yellow [E] = estimated, purple [AI] = interpreted, gray [P] = placeholder, blue [R] = real/live. Checking the badge column first prevents acting on placeholder values as if they were measurements.
Common mistakes
- Ignoring the badge and treating all values in the metrics panel as equally authoritative — the entire purpose of the badge system is to prevent this.
- Sharing report screenshots without the badge visible — recipients will read the numbers as measurements without the label context that defines their reliability.
- Not learning the badge color convention and relying on reading the label text each time. This means the color coding is designed for quick scanning. Learning the colors makes the system faster to use.
- Assuming the badge will always be visible in exported formats. This means verify that data exports preserve provenance labels before distributing analysis results to coaches or external platforms.
In SwingVantage Motion Lab
The data source label system is implemented throughout Motion Lab — on every metric row, in report headers, in retest delta calculations, and on progress charts. Hovering any badge shows a tooltip explaining the specific source and method used to derive that value.
Related terms
- Measured ValueA measured value is a number that came from a real sensor or instrument. This means a launch monitor, radar gun, or wearable. This means and was not inferred or estimated from video or rules.
- Estimated ValueAn estimated value is a number derived from observable cues and encoded rules rather than from direct sensor measurement. This means a confident, grounded inference that is never presented as a measured fact.
- Interpreted ValueAn interpreted value is a judgment or label that emerged from AI reasoning applied to raw data. This means such as 'likely timing issue' or 'probable hip stall'. Instead of from a direct measurement or a deterministic rule.
- Placeholder DataPlaceholder data is a temporary illustrative value shown in a UI before real data is available. This means always explicitly labeled as a placeholder. As a result, it is never mistaken for an actual measurement or estimate.
- Real / Live DataReal / Live data is data captured from an actual user session or sensor. This means not a placeholder, not a demo, not a mock. This means and is labeled accordingly. As a result, you can trust it as genuine input to analysis.
- Evidence LabelAn evidence label separates what SwingVantage measured (a sensor or launch-monitor number) from what it estimated (an inferred likely pattern). As a result, an estimate is never presented as a lab measurement.
Put this into your swing
SwingVantage can spot this in your own swing — free to start.