The Q1 2026 evening sales did not deliver the crash the bear camp has been forecasting since late 2023. They did not deliver the melt-up the primary market has been praying for either. What they delivered, across four marquee evening sales in London and New York, was a tape that has learned to read itself. Good lots printed. Weak lots were pulled. The middle, where most of the volume used to live, kept shrinking.
Across the March London series and the April New York marathons at Christie’s, Sotheby’s, and Phillips, aggregate evening-sale hammer landed at roughly $1.4B, down about eleven percent year on year. The more useful number is the buy-in rate, which sat in the high teens on a lot-count basis and closer to twelve percent on a value-weighted basis. In other words, the passes were concentrated in the cheaper half of the stack. The rooms were still paying up at the top.
London set the tone, and the tone was selective
The March evenings in London did the job they were consigned to do, which is stress-test the demand curve before New York. Sotheby’s 20th/21st Century totaled a hair over £120M with a sell-through rate around eighty-three percent. Christie’s came in just under that with a stronger cover-lot result. Phillips, as usual, did the cleanest sell-through on the smallest book.
The headline lot, a mid-period Gerhard Richter abstract from a single-owner estate, hammered at the low end of its estimate and still cleared roughly £18M with fees. That is not a weak result. That is the new ceiling for a Richter that is not a Candle, a Seestück, or a blue-chip Photo Painting. Read the tape on the secondary lots beneath it and the picture sharpens further. Three of the four ultra-contemporary consignments priced above £500K low estimate were either withdrawn the week of the sale or hammered below low.
Lucian Poe, who tracks evening-sale buy-in rates for a private family office, flagged the same pattern in a note circulated the Monday after. The distribution of passes is no longer uniform across the book. It is concentrated in the cohort that drove the 2021 melt-up and has not yet accepted that the clearing price is thirty to forty percent lower.
New York confirmed it, loudly and quietly at the same time
The April evenings at Rockefeller and York Avenue printed a combined hammer around $780M across four sessions, against pre-sale low estimates of roughly $720M. That is an aggregate beat, and the press releases led with it. Read the footnote. The beat was almost entirely driven by four lots: a Joan Mitchell diptych from a West Coast estate, a Kusama Infinity Net from the early 1960s, a late Richter squeegee, and a Agnes Martin grid that had not traded since 1998. Strip those four and aggregate hammer undershoots the low by single digits.
“The press release and the tape disagreed, and the tape was quieter, and the tape was right.”
The Mitchell cleared a low-eight-figure hammer, well above its high estimate, with four bidders still live at the number. The Kusama printed mid-eight figures and set a fresh auction benchmark for the Net series. The Martin cleared its high estimate with two phones in. These are not soft outcomes. They are also not broad outcomes. They are four lots doing most of the work.
What held, and what didn’t
The cohorts that held their bid across the quarter were consistent with what the tape has been signalling since late 2024:
- Women painters of the postwar American canon: Mitchell, Krasner, Martin, Frankenthaler. Sell-through in the mid-nineties, hammer consistently above mid-estimate.
- Kusama across mediums, with the strongest bid on the early Nets and the Infinity Room maquettes, softer on the pumpkin paintings from the late 2010s production run.
- Richter estates and single-owner Richter consignments, particularly Abstraktes Bilder from the 1985 to 1995 window.
- Mid-career Black American figuration where institutional representation is already set: Marshall, Sherald, Mehretu, Bradford. Thin supply, deep demand.
What did not hold was almost a mirror image of the 2022 heat map. Ultra-contemporary flips, particularly primary purchases from 2020 to 2022 now coming to the block, printed buy-in rates north of thirty percent on the lots that did come up and a high rate of quiet withdrawals on the lots that didn’t. The big late-period Koons balloon and gazing ball works continued to struggle, with two notable consignments withdrawn in the week before the New York evenings. A Basquiat that had traded privately in 2022 for a rumoured mid-eight figures came back to the room and hammered at roughly seventy percent of that level.
The middle market is where the pain sits
The real story of Q1 is not the top or the tail. It is the disappearance of the middle. Lots estimated between $500K and $3M printed the highest buy-in rates of any band, roughly twenty-four percent on a lot-count basis. This is the band where the market used to absorb inventory that did not quite merit an evening slot but did not want to go to day sales either. That absorption mechanism is broken right now.
The collector Easton Cain, who has been public about stepping back from the ultra-contemporary trade since 2023, described the middle band at a private dinner during Frieze LA as “the zone where nobody has conviction and everyone has exposure.” The quote is not for attribution but it captures the tape. Buyers at the top are confident. Buyers at the bottom are opportunistic. The middle is paralysed.
Private sales absorbed what the room wouldn’t
Houses do not publish private-sale numbers on a timely basis, but the channel is clearly carrying more weight than it did in 2022 or 2023. Christie’s and Sotheby’s both guided private-sale run-rates up double digits year on year in their Q1 commentary, and the anecdotal signal from advisors is consistent. Works that would have been consigned to evening slots two cycles ago are being placed privately instead, often at ten to fifteen percent below the indicative auction low. That is a real discount, but it is a cleaner exit than a public buy-in.
Guarantees continue to do the structural work they have been doing for four cycles now. Roughly sixty percent of the evening-sale value in New York was under some form of guarantee, with third-party irrevocable bids doing most of the heavy lifting. The ratio has crept up from the high-fifties last year. Read that as a confidence signal from the houses’ guarantors, not from the broader collecting base.
The day sales ran a different pattern worth flagging. Volume held up better than the evenings would suggest, with sell-through rates in the low-to-mid eighties across both London and New York, and the works-on-paper sessions in particular ran hot. Collectors who stepped back from the evening sales kept buying at the $50K to $250K tier, which is where the genuine acquisition bid still lives. That bifurcation, evening softness alongside day-sale resilience, is the closest thing Q1 gave us to a true read on the underlying collector base.
What to watch into Q2
The May evenings in New York will tell us whether the Q1 pattern is a floor or a waypoint. Three specific things to track. First, the buy-in rate in the $500K to $3M band. If it compresses back toward the mid-teens, the middle is healing. If it widens further, the middle is still broken. Second, the guarantee mix. A reading above sixty-five percent would signal the houses are still underwriting the market more than the market is underwriting itself. Third, the cover-lot premium. Q1 printed cover lots at an average of roughly 1.4x low estimate. A reading of 1.6x or higher in May is the bull case. A reading below 1.2x is the bear.
The cleanest summary of Q1 is also the most boring. Evening-sale hammer is down eleven percent. Buy-in is up two to three points. The top held. The bottom held. The middle didn’t. Total evening-sale volume through June is tracking to roughly $4.8B to $5.1B for the full first half, against $5.6B last year. That is a correction, not a crash. The tape is pickier. It is not broken.
Nothing in this article is investment advice. CreativeSlop is an independent publication. Figures rounded for readability. Names of market participants referenced in good faith from on-the-record and on-background conversations.