The afterlife of a bought-in lot: how buy-ins reprice comparable works
A bought-in lot is more than a lot that fails to meet it's reserve; it is a visible point of disagreement between seller expectations and buyer demand, marking a public failure of price validation that enters the record for comparable works and shapes how subsequent similar works are judged.
The central question is not whether bought-in lots carry information, but where the signal from a major buy-in travels. Major buy-ins rarely create broad auction-room contagion; their effects are selective, local, and comparable-driven, concentrating around lots that share the same artist, category, estimate band, or object profile.
The immediate effect is local, appearing most clearly in the next five to ten catalogue lots before fading across the rest of the sale. The longer-term effect is narrower but more consequential, emerging through same-artist performance, recurrent bought-in lots for a minority of artists, and weaker outcomes where estimates remain stable or continue rising after a major buy-in.
A bought-in lot therefore does not automatically impair an artist’s entire market. Stronger works can still clear successfully, even after a related lot has failed. But where later lots depend on the same fragile price premise, absorption becomes more vulnerable. Major buy-ins rarely break markets in aggregate, but they reveal where the pricing architecture is already weak.
→ Explore auction-market signals with Artscapy's Market Intelligence
Data and scope
A clean buy-in is defined as a lot recorded as passed with no hammer price. This excludes anomalous cases where a lot carries a passed status while still recording a hammer value, preserving a stricter definition of failed public absorption.
The analysis draws on 125,454 bought-in lots across 7,101 sales. The immediate post-buy-in sequencing analysis identifies 71,641 flagged lot-window events, covering 49,404 lots across 6,205 sales from December 19, 2986 to May 7, 2026. The scale of the dataset allows the analysis to move beyond anecdotal examples and test whether bought-in lots produce repeatable aftereffects in subsequent auction performance.
Major buy-ins are identified through several thresholds: artist-level top decile, sale-level top decile, lots above three times the artist’s trailing twelve-month median, and sale top quartile. This captures buy-in events that are material either to the artist’s own market, to the sale context, or to the artist’s recent trading history.
Post-buy-in performance is tested across both local and longer-term windows. The local analysis examines the next five lots, next ten lots, and all remaining lots in the catalogue. The longer-term analysis tracks three-, six-, and twelve-month windows, as well as the next ten comparable lots. Comparable works are defined as lots by the same artist, in the same normalised category where known, and within 0.5x to 2.0x of the bought-in lot’s mid-estimate, with size and date similarity incorporated where possible.
To separate broad auction-room softness from comparable-driven repricing, the analysis also uses a residual “all other” baseline: post-buy-in lots that share none of the same artist, category, or estimate-band characteristics as the bought-in lot. Post-buy-in weakness is also tested against same-sale peers, same-segment peers, broader market-season performance, and the artist’s own pre-trend. This framework distinguishes general market drift from more specific deterioration around related works.
From bought-in lot to pricing signal: why failed auction prices matter
Once a lot is bought-in, the market is left with an unusual form of evidence: not a clearing price, but a failed price point. The reserve remains private, but the estimate range has been publicly tested and rejected. That distinction matters because auction pricing is built through comparable lots. A bought-in lot does not provide a realised value, but establishes a boundary around what buyers are unwilling to validate under public conditions.
The analytical question is where that failed price point remains relevant. It may have little bearing on unrelated material elsewhere in the sale, but where later lots share the same artist, category, estimate band, or object profile, the bought-in lot can become a negative comparable: evidence that a similar valuation premise has already met resistance.
A bought-in lot becomes most consequential when later works depend on the same pricing premise. Its signal does not need to spread across the whole sale to matter; it only needs to alter confidence around the works most closely connected to it. In that sense, the aftereffect of a major buy-in is best understood as selective repricing rather than auction-room contagion.
Immediate buy-in effects: why weakness is localised
If a bought-in lot created broad auction-room contagion, deterioration would be expected across the full remainder of the catalogue; however, the pattern is more concentrated. Post-buy-in weakness is strongest immediately after the failed lot and fades as the sale moves further away from the event.
The strongest deterioration appears immediately after the bought-in lot. In the five lots immediately following a major bought-in lot, 30% to 33% of testable cases showed measurable deterioration relative to the comparable pre-buy-in window. The effect remains visible across the next ten lots, where roughly a quarter of testable cases deteriorate. The effect falls off sharply across the remainder of the catalogue, where the share declines to roughly 13% to 15%.
Figure 1. Post-buy-in deterioration is most visible in the next five to ten catalogue lots and fades across the rest of the sale, suggesting a temporally concentrated rather than sale-wide effect.
Source: Artscapy | Visualisation: Darden Gildea
This does not mean that the bought-in lot contaminates the auction as a whole, rather it destabilises the pricing field immediately around it. The next few lots may remain exposed to the psychological and comparative pressure created by the buy-in event; later lots are less consistently affected as the sale moves away from the immediate valuation context.
Aggregate medians reinforce this interpretation. Median hammer-to-mid-estimate performance and median hammer price generally soften after major buy-ins, but the deterioration is not dramatic. Sell-through rates remain largely flat in the five- and ten-lot medians, while secondary measures such as the share sold below low estimate and the share sold above high estimate remain mostly stable in short windows and only mildly worsen across all remaining lots.
The immediate sale-window evidence therefore establishes a bounded effect: major bought-in lots are followed by measurable deterioration nearby, but not by a general collapse in sale performance. The remaining question is whether that nearby weakness reflects catalogue sequence alone, or whether it is concentrated among lots that share the same artist, category, estimate band, or valuation premise.
Comparable-driven repricing: how buy-in effects travel through similar lots
The local-window results establish the timing of post-buy-in weakness: deterioration is most visible near the failed lot. But timing alone does not explain transmission. Further analysis therefore has to ask a more precise question: does weakness follow catalogue order or does it follow similarity, defined are shared artists, category, estimate band, or valuation premise?
Unrelated lots provide the clearest test. If a major bought-in lot weakens each subsequent lot, then post-buy-in lots with no shared artist, category, or estimate-band characteristics should weaken too. They do not. In this residual group, sell-through remains between 80% to 100% for the next five lots, between 75% to 80% for the next ten lots, and 73% to 77% across all remaining lots. Further, median hammer-to-mid performance remains comparatively stable, ranging from 0.91 to 0.96 in the next five lots, 0.89 to 0.95 in the next ten lots, and 0.86 to 0.91 across all remaining lots.
This residual baseline sets the bar against which contagion claims are measured. Within the post-buy-in sample, unrelated lots remain materially stronger than matched lots. The signal becomes most visible where later lots that share the same artists, category, estimate band, or valuation premise as the bought-in lot demonstrated deterioration.
The sharpest immediate effect appears when another lot by the same artist comes up to the auction block soon after the bought-in lot. In the next five lots, these same-artist examples perform materially worse than unrelated lots: sell-through rates are 33 to 50 percentage points lower and hammer-to-mid ratios are 0.071 to 0.112 points lower. When the same artist reappears shortly after a major buy-in, performance weakens materially relative to unrelated lots. Although a causal relationship in buy psychology cannot be drawn, it is consistent with the failed price point becoming directly relevant for the same artist’s near-term pricing. The caveat is that same-artist post windows can be sparse, with a median of only two offered lots, so the effect is sharp but relatively narrow.
Figure 2. Matched lots weaken more than unrelated post-buy-in lots, with same-artist lots showing the sharpest immediate hammer-to-mid deterioration.
Source: Artscapy | Visualisation: Darden Gildea
The broader and more persistent effect appears in lots in the same category. In the ten-lot window, same-category lots show weaker post-buy-in performance in 28% to 32% of testable cases, across the four major-buy-in definitions. Even across all remaining lots, the pattern remains visible, where the share is 18% to 22%. This suggests that a bought-in lot can unsettle confidence beyond the artist alone. When one lot fails in a particular category, later lots in the same category may face more selective demand.
Lots in a similar estimate band also show persistent weakness, which shows that the signal is price-level specific. In the ten-lot window, lots in a similar estimate band underperformed in 24% to 32% of cases. Across all remaining lots, the share remains elevated at 21% to 26%. This matters because it shows that the market is not only to the artist and category, but to price levels, even for aesthetically and art historically unrelated lots. When a lot fails at a given estimate range, later lots priced in a similar range become more exposed.
These results point to a selective form of contagion. Weakness does not wash indiscriminately through subsequent lots, but concentrates where the market can plausibly read one lot as information about another. Same-artist, same-category, and similar-estimate-band lots are pulled into the failed lot’s orbit, while unrelated works continue to clear on their own fundamentals. In practice, major buy-ins act less as a general drag on the room and more as a targeted repricing signal for the specific clusters that share the same valuation premise.
Artist risk vs segment risk: when a buy-in signals selective weakness
Although same-artist lots offered near a major bought-in lot show the sharpest immediate deterioration, the risk can exist at the segment level, not the artist level. The artist’s name exposes later lots to the negative signalling, but is not determinant for the sale outcome.
The market is rarely responding to the artist’s name alone. It is responding to a specific combination of price, object quality, category, medium, period, scale, provenance, freshness, and demand depth. One lot can fail because it sits in a vulnerable segment of the artist’s market, while another, even at a materially higher estimate, can clear because it occupies a more defensible position within the artist’s hierarchy.
This distinction is especially important in mature markets, where a single artist often contains multiple pricing structures. Paintings may behave differently from works on paper; unique works from editions; early periods from late periods; major works from minor works; fresh-to-market supply from material that has appeared repeatedly at auction. A bought-in lot in one segment may expose weakness in that segment without undermining demand for the artist’s best works.
The relevant risk is therefore not always artist risk. More often, it is segment risk. The question is not simply whether the artist’s market has weakened, but whether later lots depend on the same pricing premise as the bought-in lot. Where they do, the failed lot can become a negative comparable. Where they do not, the artist’s market may continue to absorb stronger or more differentiated works.
This distinction is critical because a bought-in lot may create opportunity if the failure reflects an unrealistic estimate or a weak object rather than exhausted demand. But it may signal deeper risk when multiple comparable lots begin to fail around the same category or estimate band.
A major bought-in lot does not necessarily cap an artist’s market. It reveals where the market is drawing the line between object quality, price level, and demand depth.
Slow-motion repricing: the three-, six-, and twelve-month aftereffect
While the immediate sale-window analysis shows that major buy-ins can create local pressure, a longer-term analysis reveals a subtler story: a failed price point can enter an artist’s pricing history and continue to shape performance well after the sale has ended.
Across all major buy-in events, same-artist performance softens over three-, six-, and twelve-month windows. In the three-month window, sell-through rates fall slightly from 78% to 77% and median hammer-to-mid falls from 0.92 to 0.91. In the six-month window, sell-through rates remain stable at 76%, but median hammer-to-mid declines from 0.92 to 0.90. Interestingly, the twelve-month window shows the clearest deterioration: sell-through fall 2% to 75.5% while median hammer-to-mid falls from 0.93 to 0.89.
Figure 3. Same-artist hammer-to-mid performance declines across all three windows, with the clearest deterioration over twelve months.
Source: Artscapy | Visualisation: Darden Gildea
Further, the twelve-month window shows a weaker distribution of price outcomes. The share of works selling below low estimate rises from 29% to 33%, while the share selling above high estimate falls from 29% to 26%. Post-buy-in softness is therefore not limited to lots passing. Even when works sell, they are more likely to clear with weaker bidding intensity relative to expectations.
The next 10 comparable auction lots by the same artist adds a narrower but important perspective. In that test, comparable works are defined as the same artist, same normalised category where known, and mid-estimate within 0.5x to 2.0x of the bought-in lot. Only a quarter of examined buy-ins had enough comparable lots of either side of the buy-in event to support a before/after comparison. Of those qualified events, sell-through rate remains flat at 80% across the 21 auction results, but median hammer-to-mid falls from 0.94 before the buy-in event to 0.91 after.
The aftereffect of a bought-in lot is not always visible as a sharp fall in sell-through rates. It often appears first as weaker pricing intensity: comparable works still sell, but they clear less strongly against estimates. The longer-term effect of a bought-in lot is not always visible in headline sale rates; it often appears more clearly in the relationship between estimates and realised prices.
The bought-in lot’s afterlife is therefore subtle but consequential. It becomes integrated in the comparable record as a failed price point. Future works continue to sell, but if estimates do not absorb the signal, the same pricing premise remains vulnerable.
Artist-level buy-in recurrence: when bought-in lots become more frequent
A major bought-in lot is not usually followed by a broader run of failures across the artist’s market. For most artists, the effect stops short of recurrence. But for a meaningful minority, bought-in rates rise after the initial event, and the pattern becomes more visible over longer windows.
Of the 36,205 artists examined, 2,691 or 7.4% showed a meaningful post-buy-in deterioration in at least one of the three tested windows: three months, six months, or twelve months after a bought-in lot. Within that group, 70% showed a broad increase in bought-in rates after the event, while 50% showed a more specific pattern: buy-ins rose more sharply among lots that resembled the original bought-in lot than among less comparable lots by the same artist.
The recurrence pattern strengthens over time. Among artists with sufficient pre- and post-event data, 5.8% show an overall material increase in bought-in rates over three months, rising to 8.4% over six months and 12.7% over twelve months. The median pre-to-post bought-in rate moves from 0.0% to 10.3% in three months, 0.0% to 16.7% in six months, and 0.0% to 20.0% in twelve months.
Where recurrence does appear, it is rarely evenly distributed across the artist’s market. It is more often tied to shared characteristics with the original bought-in lot. Estimate range is the clearest channel. Later lots priced in a similar range to the original bought-in lot show the strongest increase in buy-in risk relative to less comparable works: +7.8 percentage points after three months, +9.5 percentage points after six months, and +9.1 percentage points after twelve months.
Figure 4. At twelve months, later buy-ins are most strongly linked to works offered in a similar estimate range, suggesting that recurrence is more a pricing problem than an artist-wide market event.
Source: Artscapy | Visualisation: Darden Gildea
Subject similarity applies to more cases, but the effect is less specific. By twelve months, subject-related lots had a 2.6% lower sell-through rate, compared to the 9.1% decrease for the estimate range. Medium, size, subject, date, and estimate range all matter, but the strongest recurring signal is price level.
The implication is measured rather than universal. A major bought-in lot does not usually lead to a broad run of failures, but when subsequent buy-ins occur, they cluster around lots that share similarities with the original, especially those offered at similar price levels. The recurrence is therefore best understood as a pricing problem before it is an artist problem: the market is not necessarily rejecting the artist, but questioning whether that segment can support the estimates being asked of it.
Buy-in case studies: Jammie Holmes and Sigmar Polke
A buy-in cascade is most convincing when the bought-in lot does not remain a single failed outcome. The strongest cases are those where weakness continues to appear in comparable works, subsequent estimates, recurrence patterns, and control-adjusted performance.
Jammie Holmes and Sigmar Polke show two different versions of that pattern. Holmes illustrates the vulnerability of a thinner, momentum-driven contemporary market, where pricing can depend heavily on recent comparables and confidence in continued absorption. Polke shows that comparable-driven pressure can also emerge in a higher-value, more established market, where the artist’s broader status remains intact but specific segments become more exposed.
Figure 5. Both case studies show weaker post-buy-in absorption, with a sharper decline in Holmes’s thinner contemporary market and a more selective but still material weakening in Polke’s established market.
Source: Artscapy | Visualisation: Darden Gildea
Figure 6. After the bought-in lot, hammer performance weakened for both artists. Holmes moved from selling above mid-estimate to materially below it, while Polke’s post-buy-in performance also moved further below expectation.
Source: Artscapy | Visualisation: Darden Gildea
Jammie Holmes: fragility in a thinner contemporary market
Jammie Holmes is one of the clearest examples of a bought-in lot exposing fragility in a recent contemporary market. On June 4, 2022, Holmes’s Child Soldier Series – 5 Cents a Day failed to sell in Ravenel’s Spring Auction against an estimate of $66,100 to $97,400.[1] In the twelve months after the sale, Holmes’s sell-through rate fell from 90.0% to 40.9%, while the median hammer-to-mid-estimate ratio declined from 1.20 to 0.80.
The deterioration is not simply a small-sample effect. The pre- and post-event windows include 30 works by Holmes before the event and 22 after, including 18 comparable works before and 11 after. That provides enough depth to treat the post-buy-in weakness as more than an isolated result.
What makes the case especially revealing is that estimates did not meaningfully adjust. Post-event estimates remained roughly flat, rising only 2.4%. Yet the next ten comparable lots weakened, the underperformance survived matched controls, and later buy-ins rose by 43 percentage points.
Holmes therefore illustrates how quickly a pricing structure can become exposed when demand thins but estimates remain sticky. The bought-in lot did not simply mark one failed outcome; it revealed that the surrounding estimate framework had become more difficult to defend. Because pricing did not adjust, later comparable works were left to test the same premise under weaker conditions.
The Holmes case is especially revealing in the context of a thinner contemporary market, where price levels often depend heavily on recent comparables and confidence in continued absorption. Once that confidence weakens, there may be few long-established reference points to stabilise pricing. The cascade is not simply a reaction to one failed lot; it is the process by which a recent pricing framework loses credibility.
Sigmar Polke: repricing pressure in a blue-chip market
Sigmar Polke provides a more established version of the same mechanism. Here, the market is neither thin nor speculative. On May 9, 2022, Polke’s Filzschleife was bought in at Christie’s The Collection of Thomas and Doris Ammann Evening Sale against an estimate of $1.5 million to $2.0 million.[2] In the twelve months following the bought-in lot, Polke’s sell-through rate fell from 91% to 60%, while the median hammer-to-mid-estimate ratio declined from 0.98 to 0.81.
The pattern also appears among the Polke lots most similar to Filzschleife. Across the next ten comparable lots, sell-through fell from 100% to 80%, while the hammer-to-mid-estimate ratio fell from 1.18 to 0.76. The typical comparable sale therefore shifted from clearing above the mid-estimate to selling materially below expectations. Among Polke works at a similar estimate level, the tension is even clearer: auction houses raised expectations after the bought-in lot, with the median mid-estimate increasing by 25%. Buyers did not meet those higher expectations. Hammer-to-mid-estimate performance fell from 1.39 to 0.93, and the share of comparable works selling below low estimate rose from 0% to 50%. Later comparable Polke works were being estimated more aggressively, but bidding no longer supported those estimates.
The Polke case extends the argument beyond thinner contemporary markets. The issue is not the disappearance of demand for the artist, but the weakening of confidence around a specific pricing level. In a blue-chip market, a bought-in lot is less likely to erase the artist’s broader auction track; however, it can still expose pressure around works dependent on similar valuation assumptions. Filzschleife did not make the Polke market collapse. It made a particular estimate structure harder to defend.
Read against Holmes, the contrast is instructive. Holmes shows how quickly a thin contemporary market can lose pricing confidence when estimates remain anchored to recent momentum. Polke shows the same mechanism operating more selectively in an established market, where the artist’s broader status remains intact but comparable works become harder to price and absorb.
Buy-in effects vs market drift: separating signal from broader auction weakness
Post-buy-in weakness cannot be interpreted in isolation. A bought-in lot may occur during a weak sale, within a soft category, or in a difficult market season. Underperformance in subsequent lots cannot be solely attributed to the failure to sell, as it may be one visible expression of a broader environment.
To separate these effects, post-buy-in performance is tested against several benchmarks: other lots in the same sale, comparable lots in the same category and price band, broader market-season performance, and the artist’s own pre-event trend. On median, major buy-in events often move roughly in line with controls, suggesting that broad sale and market conditions explain much of the average softness.
But the control tests do not eliminate the buy-in signal. Around a quarter of buy-in events underperform each external control individually. Lots by the same artist that underperformed against their own trend were then measured against two of three external benchmarks: same-sale peers, same-segment peers, and broader market-season performance. Even under that test, 9.6% of lots underperformed within three months, 11.8% in six months, and 13.2% within twelve months. When the controls were tightened to all three benchmarks, the shares that underperformed were 8.3%, 10.4% and 11.5%, respectively. Further, across the board, 21.3% of works by the same artist as the bought-in lots demonstrated subsequent artist-specific weakness.
This is an important constraint on interpretation. Buy-ins do not operate outside market conditions. Broader softness matters, and in many cases it explains a substantial portion of post-buy-in weakness. Yet the persistence of artist-specific flags after controls shows that the buy-in signal cannot be dismissed as simple market drift.
The implication is not that bought-in lots mechanically cause later weakness. In many cases, the bought-in lot is better understood as a visible symptom of an already soft artist segment, price pocket, or auction season. But in the strongest cases, weakness persists even after controlling for sale context, segment performance, market season, and pre-existing artist trends. Those cases suggest that the bought-in lot becomes more than a failed outcome; it becomes part of the subsequent pricing record.
Estimate discipline after buy-ins: how pricing adjustments contain risk
The clearest practical implication of the analysis concerns estimates. After a major lot is bought-in, estimates adjust, but often only modestly. Median mid-estimates drop 1.5% in the three-month period, 1.9% in the six month period, and 3.0% in the twelve month period. For the next ten comparable lots, the median mid-estimate drops 1.1%. The average mid-estimates decline slightly, reaching a 3.3% decline over the twelve month period and the average range width narrows slightly.
Realised performance weakens more clearly than estimates. The median hammer-to-mid estimate ratio falls by 1.7% in three months, 2.3% in six months, 3.7% in twelve months, and 3.3% across the next 10 comparable lots. The market therefore appears to be adjusting through realised performance more than through aggressive estimate revision.
That gap matters. Across all major buy-ins, more than half of the lots events still show stable or rising estimates after the bought-in lot: 54.6% in three months, 54.4% in six months, and 52.8% in twelve months. Yet, weaker sell-through rates are more common when estimates are stable or rising than when they are revised downward. In the twelve-month window, weaker sell-through rates appeared in 40.5% of downward-revised cases, compared with 46.2% of stable-or-rising cases and the average sell-through rate moves in opposite directions: +2.0 percentage points when estimates are cut, compared with -3.3 percentage points when they are stable or rising.
Figure 7. Estimates decline only modestly after major bought-in lots, while hammer-to-mid performance weakens more visibly. The market appears to adjust first through weaker bidding against estimates rather than through immediate estimate resets.
Source: Artscapy | Visualisation: Darden Gildea
The implication is not that every post-buy-in estimate should be cut aggressively. Rather, a major bought-in lot creates a new pricing constraint. It is public evidence that a comparable price level has met resistance. When subsequent estimates adjust, the market appears more capable of absorbing the signal. When estimates remain unchanged or continue rising, later sell-through weakens more often.
For specialists and consignors, estimate discipline is therefore the clearest control lever. A bought-in lot does not have to become a cascade. But when later comparable works are priced as though the failed lot revealed nothing, the risk of further weakness increases.
Figure 8. In calendar windows, weaker sell-through is more common when estimates remain stable or rise after a major buy-in than when estimates are revised downward.
Source: Artscapy | Visualisation: Darden Gildea
→ Assess your artwork's auction-marker position with Artscapy
Market implications: buy-ins as selective pricing signals
Major buy-ins do have an effect, but their effect is specific rather than systemic. The data does not support a broad auction-room contagion thesis in which one bought-in lot weakens everything that follows. Unrelated lots often remain resilient, and stronger works by the same artist can still sell successfully when they occupy a more desirable segment of the market.
The effect is strongest where the connections are real. It appears locally in the next five to ten catalogue lots, becomes more visible among works by the same artist, in the same category, or in a similar estimate band, and can persist over three-, six-, and twelve-month windows. For a minority of artists, it is also followed by a measurable increase in further buy-ins, especially where later lots share the same pricing structure.
The central implication is that a bought-in lot should be read as a pricing signal. It is not just a failed transaction. It is evidence that the market has rejected a particular valuation premise under public conditions. The more closely future lots depend on that same premise, the more vulnerable they become.
The buy-in cascade therefore operates through comparability. The bought-in lot is not the domino itself; it is the signal that shows where the dominoes are connected.
A major bought-in lot does not break the market. It exposes whether the pricing story behind comparable works can still hold.
→ Read more art market analysis from Artscapy
Resources
[1] Ravenel - Spring Auction 2022, June 4, 2022 - Lot 074: Jammie Holmes, Child Soldier Series - 5 Cents a Day
[2] Christie’s - The Collection of Thomas and Doris Ammann Evening Sale, 9 May 2022 - Lot 8A: Sigmar Polke, Filzschleife
Continue Your Market Research with Artscapy
