What 50 GSA Test & Lab Equipment Analyses Reveal About Blind Auction Risk

Forty-three of the 50 Core analyses contained at least one configuration component recorded as NOT_ESTABLISHED, CONFIRMED_MISSING, or SELLER_EXCLUDED. That does not mean 86% of the lots were unusable. It means the preserved auction evidence did not establish every economically relevant part of the system.

• An RF power meter without an established compatible sensor and sensor cable.
• An analytical instrument without its data station, software, or complete sample path.
• A microscope with motorized hardware but no established camera, acquisition workstation, or full fluorescence path.
• A centrifuge whose rotor and adapters are not established.
• Handheld meters without leads, probes, or calibration documentation.

Identify the configured asset, not just the chassis. The configuration-completeness guide explains why the auction evidence can establish a platform without establishing the resale state being purchased.

Forty-five of the 50 Core analyses—90%—had condition confidence below MEDIUM. Condition confidence is separate from identity confidence. A lot can have a clear manufacturer/model identity while the evidence still does not bound the condition risks that determine a reseller's reachable resale state.

No analysis in the sample had a TESTED condition state. Two analyses had a CALIBRATED condition state, but both require a narrow reading. In this repository, CALIBRATED can preserve historical calibration-label evidence. It is not automatically equivalent to a fresh, independent functional verification of all value-bearing functions.

The September snapshot therefore recorded 48 / 50 analyses, or 96%, without either a TESTED or CALIBRATED condition state.

• A clean exterior establishes appearance.
• An active display establishes power/display evidence.
• Self-test can establish a limited internal check.
• A calibration label establishes historical calibration context.
• A representative functional test establishes a stronger operating state.
• Application-level performance may require still more evidence.

The category determines the right test protocol. The common rule is narrower: do not price the lot as if a stronger evidence state has already been demonstrated. See the testing, calibration, and repair-risk guide.

LotCaliper could calculate a numeric maximum hammer for 47 of the 50 Core analyses. The other three were INSUFFICIENT_DATA: the available evidence did not support a defensible numeric resale-and-bid model. Of those 47 numeric cases, 22 produced a $0 blind maximum hammer—46.8%.

A $0 blind ceiling is not a claim that the equipment has no resale value. It means that under the preserved evidence state and the standard LotCaliper reseller cost-and-return assumptions, no positive blind acquisition price cleared the model.

The maximum-bid guide explains how configuration, function, serviceability, logistics, repair, liquidity, and required return can consume the available spread.

A REVIEW is not a weaker form of BUY. It means the opportunity may become attractive if specific decision-changing evidence is obtained or if buyer-specific economics differ materially from the blind model. INSUFFICIENT_DATA is not a valuation of zero; it means the evidence is too weak for a numeric resale or acquisition ceiling.

Taken together, 74% of the Core sample did not receive an unconditional BUY under the preserved evidence and LotCaliper's standard reseller assumptions.

For the 47 analyses with both a numeric maximum hammer and a positive conservative-resale model, the ratio was:

blind maximum hammer / conservative resale

The zero-heavy distribution reflects the lots where risk and costs consumed the available acquisition spread. These ratios must not be turned into a shortcut such as “bid 15% of resale.” The ratio is an output of the cost stack and evidence state, not an input rule. A fully verified, easy-to-ship instrument with strong market support can support a materially different ratio from a heavy, incomplete, repair-risk analytical system.

This report is useful as a diligence prior, not as an automatic bidding formula.

1. Resolve the economic identity. Confirm exact model, suffix, installed options, modules, accessories, software, and quantity.
2. Classify the condition evidence. Separate appearance, power-on, self-test, calibration history, representative functional testing, and application-level performance.
3. Define the resale state you can support. Do not use tested-working comps for an untested or incomplete evidence state without an explicit bridge.
4. Price the path to the intended resale state. Include logistics, testing, calibration, repair, missing components, selling friction, returns, and uncertainty.
5. Calculate the ceiling from the cost stack. Do not start with an arbitrary percentage of resale.
6. Identify the evidence that could change the decision. A specific test, inventory, software confirmation, rotor check, or logistics quote can remove a decision-dominant uncertainty.

The aggregate data show why this sequence matters. They do not eliminate the need to evaluate each lot individually.

The report uses a frozen LotCaliper research snapshot dated September 19, 2026. The starting corpus contained 64public analyses. The primary sample contains the 50analyses classified CORE; 14 ADJACENT analyses were excluded. All included records are sourced from GSA Auctions and were published on or before the cutoff date.

The underlying auction facts originate from public GSA Auctions evidence. LotCaliper adds its own identification, evidence classification, valuation, risk, and bid-model analysis. Read the LotCaliper methodology and the GSA diligence guide for the underlying framework.

This appendix is rendered from the frozen September 19, 2026 snapshot, not the mutable live corpus. It contains LotCaliper-derived or already-public analysis fields only; it does not reproduce auction text, photographs, credentials, or internal research evidence.

Download the 50-analysis CSV

When citing an individual statistic, include its denominator and preserve the selected-corpus limitation. Cite the row-level appendix or CSV when the aggregate calculations need to be audited or reused.

• U.S. General Services Administration — GSA Auctions API / Data.gov: official public provenance for the federal-auction facts underlying the corpus; aggregate metrics are LotCaliper-derived.
• LotCaliper methodology: evidence-state separation, conservative resale, uncertainty treatment, and risk-adjusted maximum-bid framework.
• The five linked case analyses above support their respective identity, configuration, condition, valuation, and blind-ceiling examples.

• Generalizing the selected sample to all GSA inventory: the report measures the reviewed corpus, not market-wide incidence.
• Reading $0 as “worthless”: the ceiling reflects modeled costs, uncertainty, and required return under the current evidence state.
• Using 4% or 15% as a bidding rule: the ratio is an output of lot-specific economics.
• Treating calibration history as a complete functional test: the two CALIBRATED records do not establish every value-bearing function.
• Using later outcomes to rewrite the study: later lifecycle evidence must not replace the frozen study inputs.
• Comparing recommendations without the evidence standard: decisions reflect both equipment economics and evidence quality at analysis time.