Conflicting Evidence

Conflicting evidence is normal in collectibles valuation. Two credible-looking prices may differ because they describe different variants, conditions, markets, dates, transaction terms or valuation purposes. The correct response is not to average every number or choose the most convenient one, but to determine what each item of evidence actually proves and how much weight it deserves.

A defensible conclusion follows a visible chain: subject identity, evidence quality, comparability, adjustment, weighting, reconciliation and uncertainty disclosure. Conflict should make a valuation more explicit, not more arbitrary.

Central collector rule

Do not ask only, “Which price is correct?”

Ask what each price describes, whether the transaction is verified, how closely the item matches, which market it represents and which valuation question it answers. Evidence can disagree without either source being false.

What counts as conflicting evidence?

Conflict may concern a factual premise, an interpretation of agreed facts, the market in which the object trades or the final valuation conclusion. Separating these forms matters because each requires a different response.

What is the object?

Factual conflict

Sources disagree about identity, printing, completeness, authenticity, production date, provenance or surviving population. These conflicts must be resolved before price evidence can be compared responsibly.

What do the facts mean?

Interpretive conflict

The facts may be broadly agreed, but specialists differ about whether fading is minor, restoration is acceptable, a signature adds value or a defect is market-critical.

Which market level applies?

Market conflict

Auction results, dealer prices, private sales, overseas transactions and graded examples may support different price levels because they represent different buyers, services and transaction conditions.

What valuation question is being answered?

Conclusion conflict

Two informed valuers may use similar evidence but reach different figures because they use different dates, value definitions, sale periods, market assumptions or adjustment judgements.

Apparent contradiction often disappears after normalisation

A $500 incomplete copy and a $900 complete copy do not necessarily conflict. Nor does a $1,200 dealer price automatically contradict a $750 auction result if the dealer offers authentication, warranty, careful presentation and a longer selling period. Before comparing the figures, align the evidence across the dimensions below.

Comparison axis

Question to ask

Object identity

Same product, edition, printing, state, regional issue and component configuration?

Physical state

Comparable condition, completeness, restoration, grading and eye appeal?

Transaction type

Auction, dealer sale, private transaction, trade sale, asking price or unsold listing?

Market level

Wholesale, collector-to-collector, specialist retail or general marketplace?

Time and place

Close enough to the valuation date, with comparable geography, currency, taxes and buyer pool?

Sale circumstances

Normal exposure, distress, reserve failure, bundled lot, charity sale or related parties?

Included benefits

Authentication, warranty, research, framing, shipping, return rights or immediate availability?

Valuation basis

Market, auction, retail, replacement, insurance, probate or liquidation value?

Why collectible evidence conflicts so often

Thin markets

Rare material may have only one or two visible sales over several years. A result can be heavily influenced by who bid, how the lot was catalogued, reserve strategy, seller reputation or temporary enthusiasm.

Heterogeneous objects

Copies sharing a catalogue title may differ in printing, state, region, packaging, component completeness, signature, provenance, restoration, grading and eye appeal.

Incomplete descriptions

Archived listings often omit internal writing, odour, replaced pieces, detached components, restoration or whether accessories are contemporary. Silence is not proof that a defect was absent.

Survivorship bias

Visible evidence overrepresents successful auctions, exceptional objects and platforms that retain sold records. Withdrawn lots, negotiated discounts and private trades are less visible.

Strategic pricing

An asking price may be an opening position, a test of demand, a carrying-cost decision or a seller's reluctance to part with the item. It records expectation, not exchange.

Market segmentation

General antiques buyers, specialists, dealers, institutions, investors, completionists and grade-focused collectors may value the same object differently.

How the main conflicts should be read

Completed sale versus asking price

A completed transaction proves that a buyer and seller agreed a price under identifiable circumstances. An asking price proves only what a seller seeks.

Dealer retail evidence can still be relevant where it reflects immediate availability, expert selection, authentication, stronger descriptions, return rights and longer exposure. It may answer a retail or replacement question rather than an auction-realisation question.

Public auction versus private sale

Auction evidence is visible but price reporting can be inconsistent. Hammer price, buyer's total and seller's net proceeds are different numbers.

A private sale may reflect excellent specialist matching, but test whether money changed hands, whether the parties were independent, whether other property or trade value was included and whether the quoted figure was gross or net.

Recent outlier versus historical series

A record result may indicate genuine repricing, a newly recognised variant, exceptional condition, important provenance or simply an unrepeatable bidding contest.

One valid high sale may justify raising the upper boundary of a range without resetting the central value of every example.

Older strong evidence versus weak recent evidence

Recency matters, but it is not the only criterion. An older result with exact identification, excellent images, a full condition report and confirmed terms may deserve more weight than a recent screenshot with vague details and no proof of completion.

Expert opinion versus transaction evidence

Transactions show what occurred. Experts help determine whether the object was correctly identified, whether hidden defects or restoration matter and whether the transaction is comparable. Strong valuations use both, while testing expertise, independence, inspection quality and reasoning.

Boundary with authentication and provenance

Do not treat value as settled while identity or authenticity remains materially disputed

Where authenticity controls value, report the conclusion subject to confirmation, provide alternative scenario values or defer a final conclusion pending specialist examination.

Provenance should be graded as established, strongly supported, plausible, reported, disputed or disproved. Any premium should follow the evidential level, not the appeal of the story.

A practical hierarchy for weighting evidence

No single hierarchy works for every category, but collectors need a disciplined default. Relevance, verification and transparency usually matter more than the sheer number of data points.

Tier 1

Direct, verified and highly comparable

A recent arm's-length completed sale of the same variant in similar condition and completeness, with transparent terms and clear photographs.

Usual role: Usually primary evidence.

Tier 2

Strong but adjustable

A closely related printing, adjacent grade, slightly older sale, another region or a confirmed dealer transaction where differences can be explained.

Usual role: Strong supporting evidence after reasoned adjustment.

Tier 3

Indirect market evidence

Active asking prices, unsold listings, auction estimates, related-series sales or partially verified private transactions.

Usual role: Useful for boundaries, direction and sentiment more than a precise point value.

Tier 4

Contextual evidence

Rarity guides, price guides, collector discussions, polls, historical catalogues and broad trend commentary.

Usual role: Explains context but should rarely dominate good transaction evidence.

Tier 5

Unsupported assertion

Unverifiable claims, seller-supplied stories, undated screenshots, unexplained automated estimates or copied values.

Usual role: Minimal or no weight.

The nine evidence-weighting questions

Identity match

Is it the same product, printing, state, region and configuration?

Condition match

How do structure, surfaces, fading, restoration, packaging and completeness compare?

Transaction reliability

Was the sale completed, paid, arm's length and exposed to the market?

Market relevance

Did it sell in the market where the subject would normally be offered?

Temporal relevance

How close is the evidence to the valuation date, and did the market change?

Transparency

Can the date, lot, price basis, photographs, condition and completion be verified?

Independence

Does the source benefit from a higher or lower value?

Repeatability

Is the result supported by a cluster or isolated?

Adjustability

Can material differences be explained without a chain of speculative assumptions?

Reconciliation workflow

The sequence matters. Fine price adjustments are premature while identity, authenticity or completeness remain unresolved.

Define the valuation question

State the item, valuation date, currency, value type, intended use, relevant market, assumed sale period and major assumptions.

Confirm the subject

Record exact identity, variant, condition, completeness, alterations, restoration, provenance and authenticity status. Many price conflicts are actually object conflicts.

Separate evidence by type

Keep confirmed sales, reported private transactions, dealer evidence, asking prices, unsold listings, estimates, guides and opinions in distinct groups.

Normalise the numbers

Use a consistent currency and price basis. Clarify hammer price, buyer total, seller net, premiums, taxes, shipping and bundled property.

Compare material attributes

Mark each comparable as superior, broadly equivalent or inferior on the features that actually drive value.

Explain the conflict

Identify the likely cause before making adjustments: condition, completeness, venue, timing, market level, provenance, authenticity or sale circumstances.

Rank the evidence

Assign qualitative weight such as primary, strong supporting, secondary, contextual or rejected. Avoid numerical scores that imply false precision.

Reconcile rather than average

Use means or medians only when observations are genuinely comparable. Do not blend incompatible markets or object states into one convenient figure.

Test alternative interpretations

Ask what follows if the high sale is representative, if it is an outlier, if the condition adjustment is larger or if provenance remains unverified.

State an appropriately precise conclusion

Collector meaning and risk

Publication lag, market change or variant mismatch may explain the gap. A guide is a dated reference source, not a valuation by itself.

When evidence should be excluded

Exclusion is appropriate when the source cannot answer the valuation question with reasonable reliability. Rejected evidence should remain in the audit trail with an explanation.

×The object cannot be identified adequately
×The transaction cannot be confirmed or was later cancelled
×The price includes unidentified additional property
×Related-party or trade terms are unknown
×Fraud, manipulation or copied reporting is suspected
×Condition is materially undisclosed
×The date is too remote for the relevant market
×Distress conditions conflict with the valuation basis
×Authenticity is unresolved where it materially controls value
×The figure answers a different valuation question

Outliers: reject, retain or limit?

Retain

Verified transaction
Accurate identification
Difference can be explained
Shows a real upper or lower market segment
May signal genuine market change

Give limited weight

Exceptional lot or provenance
Unusual venue
Condition incompletely disclosed
Bidding appears unrepeatable
No later evidence confirms the level

Reject

Sale did not complete
Price was misreported
Object was misidentified
Manipulation is evident
Terms make the result irrelevant

Ranges, scenario values and confidence

Conflict often supports a range, but the range should express probable variability and unresolved uncertainty rather than simply joining the lowest and highest observations.

Narrower range

Several recent close comparables, settled authenticity, good condition documentation, active trading and tightly clustered results.

Wider range

Rarely traded item, major adjustments, incomplete condition evidence, disputed provenance or authenticity, fragmented markets or genuinely dispersed prices.

Scenario values can be more honest than one broad range

$1,200-$1,500 if a signature is authenticated.
$550-$700 if the signature remains unverified.
$300-$400 if the signature is non-genuine and materially defaces the object.

High confidence

Multiple recent verified comparables, close identity and condition match, transparent terms, a stable market and little dependence on assumptions.

Moderate confidence

Some relevant sales, manageable adjustments, limited dispersion and only minor uncertainty over condition or market selection.

Low confidence

Few transactions, major differences, opaque private evidence, volatile markets or disputed authenticity and provenance.

Worked collector scenario: a rare boxed role-playing-game module

Observed evidence

$1,850: specialist auction sale, complete and near mint.
$1,100: general auction result described as complete, but photographs show a missing insert.
$2,400: dealer asking price for a sealed copy.
$900: private sale reported on a forum, with no photographs.
$1,500: price-guide figure published two years earlier.

Weak method

Average all five figures and report $1,550

This treats a sealed retail listing, an incomplete auction lot, an unverified report and a dated guide as though they were equivalent completed sales.

Stronger method

Anchor to the strongest comparable and explain adjustments

Use the $1,850 complete near-mint sale as the primary comparator, the $1,100 incomplete result as a lower condition/completeness boundary and the $2,400 sealed listing as a higher retail boundary. Give little weight to the forum claim and contextual weight to the guide.

For a complete, unsealed example in very good rather than near-mint condition, the market-value conclusion would likely sit below the $1,850 primary comparator. The valuation should explain the condition adjustment instead of presenting an arithmetic average.

Documentation checklist

A collection-management record should preserve the evidence and the reasoning, not only the final number.

Subject record

Exact catalogue identity, edition, printing, state and variant
Dimensions, component count and completeness
Condition by material axis rather than one broad grade
Restoration, alteration and authentication status
Provenance status, photographs and inspection date

Evidence record

Source, archive reference, seller or auction house and lot number
Sale date, transaction status and price basis
Hammer, premium, total price, currency and tax treatment
Description, photographs, condition, completeness and provenance
Notes on reliability, independence and market relevance

Reconciliation record

Why the evidence conflicts and which attributes require adjustment
Weight assigned to each source and reasons for exclusions
Selected market, valuation method and key assumptions
Unadjusted range, reconciled range and central conclusion
Confidence, unresolved conflicts and review triggers

Myth versus reality

Myth

The latest sale is always the best evidence

Reality

A recent but poorly identified result may be weaker than an older sale with exact variant details, clear images and confirmed terms.

Myth

Completed sales are automatically reliable

Reality

A completed transaction can still involve a bundle, related parties, distress, mistaken identity or abnormal bidding.

Myth

More data always improves the valuation

Reality

Large volumes of irrelevant evidence can obscure the few genuinely comparable transactions.

Myth

Conflicting prices should be averaged

Reality

Averaging incompatible objects or markets can produce a number unsupported by any real transaction.

Myth

A professional should eliminate uncertainty

Reality

Good practice is to analyse, manage and disclose uncertainty, not hide it behind a precise-looking number.

Myth

Different appraisals prove one appraiser is wrong

Reality

The reports may use different dates, markets, value definitions or assumptions. Align those terms before comparing the conclusions.

Review triggers

Revisit the valuation when a conflict is resolved or the market evidence changes materially.

↻A disputed variant is definitively identified
↻Authentication or specialist examination is completed
↻Missing provenance documents emerge
↻Restoration or substituted components are discovered
↻Several new sales establish a different market level
↻A record result is repeated or later cancelled
↻A grading census changes materially
↻The relevant valuation purpose or market changes

When specialist intervention is warranted

Seek a specialist appraiser, authenticator, conservator, archivist or category expert where:

authenticity materially controls value;
the object may be unique or exceptionally rare;
conflicting opinions concern a subtle production state;
restoration cannot be identified visually;
provenance could create a substantial premium;
private-market evidence dominates the available data;
the valuation is for tax, probate, litigation, insurance settlement or another regulated purpose; or
the financial consequence of error is substantial.

Key takeaways

Conflict is expected in heterogeneous, thin and imperfectly documented collectibles markets.
Normalise identity, condition, market, timing and price terms before deciding that evidence contradicts.
Weight evidence by comparability, reliability, transparency and relevance, not by volume or convenience.
Reconcile strong evidence; do not average incompatible observations.
Use ranges, scenarios and confidence ratings when the evidence does not justify a single precise figure.
Preserve the reasons for adjustments, exclusions and unresolved uncertainty.

Continue learning

Adjusting Comparables

Learn how to translate differences in condition, completeness, market and timing into reasoned adjustments.

→

Back to Valuation Evidence

Return to the evidence chapter and its full sequence of valuation topics.

→

Evidence Quality

Continue with a deeper framework for judging source reliability, transparency and comparability.

→