Every claim on this site carries a confidence tag. These tags are not decorative. They determine how a practitioner should weight each data point when making a real decision.
The Four Confidence Tags
// TAG-VERIFIED · Methodology · Mar 2026
Verified -- Independent Source
Manufacturer Claim
A data point qualifies as Verified only when it comes from an official manufacturer filing, government regulatory body, peer-reviewed journal, or credentialled independent laboratory.
Engineering Reality Independently verified
Examples from BYD audit: 1,000V architecture confirmed in BYD product specs. LFP thermal runaway around 270C from NREL thermal analysis. China public network 500kW+ chargers under 0.5% from EVCIPA 2025. Li-ion diffusion 5-10x slower at 0C from Journal of Power Sources.
✓ Verified
Verified data can be cited as independently validated. It is the only category that supports procurement decisions or engineering design choices without additional qualification.
// TAG-CLAIMED · Methodology · Mar 2026
Claimed -- Official Claim, Not Independently Verified
Manufacturer Claim
A data point is tagged Claimed when it comes from manufacturer communications, press releases, or product pages that have not been independently verified at publication.
Engineering Reality Official claim only
Examples from BYD audit: BYD 5-minute charging claim. BYD Megawatt Flash Charging 1,000kW specs. CATL Shenxing 4C / 400km in 10 min. BYD Blade Battery 3,000+ cycle life. Huawei DriveONE 600kW.
⚡ Conditional
Treat Claimed data as stated intent, not confirmed performance. It is the starting point of an audit, not its conclusion.
// TAG-INFERRED · Methodology · Mar 2026
Inferred -- Engineering Inference from Verified Facts
Manufacturer Claim
A data point is tagged Inferred when it is derived by logical analysis from verified facts using established engineering principles. The reasoning is sound. The conclusion has not been independently measured.
Engineering Reality Independently verified
Examples from BYD audit: Effective C-rate for BYD 400km / 5-min claim is approximately 5-6C, not 10-12C as media reported. Derived from published pack capacity. At announced pace, meaningful MFC highway corridor coverage unlikely before Q4 2026.
⚡ Conditional
Inferred data is sound reasoning, not verified measurement. Do not cite as independently validated data. Use as a directional indicator pending actual measurement.
// TAG-DATAGAP · Methodology · Mar 2026
Data Gap -- Withheld or Unavailable
Manufacturer Claim
A data point is tagged Data Gap when the information should exist, has commercial significance, and has not been published by the manufacturer or any credentialled third party.
Engineering Reality Official claim only
Examples from BYD audit: Blade Battery cycle life at sustained 4C or higher. BYD warranty terms for ultra-fast charging degradation. BYD Megawatt Flash Charging deployed station count. BMS throttle curve at low temperatures. Boost converter measured efficiency.
? Data Gap
Absence of data is itself an audit finding. A manufacturer's silence on commercially significant data points is not neutral -- it is a signal. The Blackbox page catalogs all open data gaps by OEM.
NON-NEGOTIABLE
If a data point cannot receive any confidence tag with certainty, this publication writes DATA NOT AVAILABLE. We never estimate. We never round. We never imply. A blank is more honest than a guess.
Version Control and Correction Policy
Every audit report carries a version number and a data cutoff date. When errors are identified, corrections are logged in the version history.
v1.2 Correction Log (BYD Flash Charging Audit)
CATL — DATA GAPS: WITHHELD BY OEM
[?]CORR-01
CATL Shenxing chemistry corrected from NMC to LFP
Original v1.0 stated CATL Shenxing used NMC chemistry. Corrected to LFP in v1.1 following DeepSeek verification review.
IMPACT: The competitive differentiation on safety grounds between BYD Blade and CATL Shenxing is smaller than originally stated. Both use LFP chemistry with similar thermal stability advantages.
TESLA — DATA GAPS: WITHHELD BY OEM
[?]CORR-02
Tesla Supercharger V4 peak power corrected to 350kW typical for passenger vehicles
Original v1.0 cited 500kW for Tesla V4. Corrected to 350kW typical for passenger vehicles in v1.1. Source: Tesla official specifications.
IMPACT: The 500kW figure applies only to Cybertruck configuration. Using 500kW as a comparison point for passenger EVs overstates Tesla charging capability.
BYD — DATA GAPS: WITHHELD BY OEM
[?]CORR-03
BYD effective C-rate corrected to 5-6C for 400km partial charge
Original v1.0 cited 10-12C effective rate. Corrected to 5-6C in v1.1. Source: Engineering calculation from published specifications.
IMPACT: The 10-12C figure implies charging the full pack in 5 minutes. BYD claim is for partial charge of approximately 40-50kWh. Lower C-rate changes thermal load calculations significantly.
Independence Statement
EngiVolt Pro does not accept manufacturer sponsorship, advertising revenue, or commissioned research. No OEM has reviewed any audit prior to publication. No affiliate relationship influences audit conclusions. Paid reports fund independent research — they do not fund favorable coverage.
Data Sources
| Source | Type | Tag |
|---|---|---|
| EVCIPA charging network statistics | Government / Industry body | Verified |
| Journal of Power Sources (LFP degradation 2018-2024) | Peer-reviewed journal | Verified |
| NREL thermal analysis (2022-2024) | Government laboratory | Verified |
| BYD official product specifications | Manufacturer official | Claimed |
| CATL official announcements | Manufacturer official | Claimed |
| Tesla official specifications | Manufacturer (independently confirmed) | Verified |
| Bloomberg / Mysteel lithium carbonate price data | Market data provider | Verified |
| Engineering calculations from published specifications | First principles derivation | Inferred |
// Read the Audit
$49 — Get the Report →
BYD Flash Charging -- Full Engineering Audit Report
The complete audit applying this methodology. 49 pages. Every data point tagged. Three unresolved data gaps identified.