Ye Stadium Economics 2026: Capacity, Ticket Prices, Attendance and Gross

On September 1, 2026, three American stadium dates were added to Ye's touring calendar in a single announcement: Reliant Stadium in Houston on October 31, AT&T Stadium in Arlington on November 7, and State Farm Stadium in Glendale on November 21.⁵ This analysis incorporates them as of September 2, 2026. Read against the venues' own published capacity figures — 72,000 as Reliant Stadium's conservative base, 80,000 fixed at AT&T Stadium, 63,400 fixed at State Farm Stadium — the announcement put 215,400 seats of base reference capacity into the market in one day.⁶⁷⁸ That is a large number and an easily abused one. It is not a ticket manifest, not an on-sale, and not revenue. What it is, precisely, is capacity exposure: inventory a promoter has committed to trying to sell.

This article does one thing carefully. It builds a show-by-show economic census of the 2026 stadium run and keeps four quantities apart that are routinely collapsed into one another in touring coverage: paid tickets, attendance, venue capacity and event manifest. The distinction is not pedantry. A stadium's published seating figure can understate a concert crowd by more than half when the field is sold, and overstate it by a third when a stage kills sightlines. Every gross figure that circulates about a run like this rests on which of those four numbers someone quietly substituted for another. Live Index applies the same rule here that it applies to ticket price history and to tour break-even arithmetic: a number is reported or it is modelled, and the two never appear in the same column.

The one hard number: SoFi

The strongest financial datum on the entire run is a pair of shows in Inglewood. Billboard Boxscore reported the April 1 and April 3 SoFi Stadium dates at $32.6 million gross across 149,000 tickets sold.¹² Both halves of that report matter, because together they permit a division that nothing else on this tour permits: $32,600,000 ÷ 149,000 = $218.79 per paid ticket, across 74,500 paid tickets per show. That is an average paid ticket — the amount actually transacted, averaged across every tier from the cheapest upper bowl to the floor — and it is the only such figure the 2026 run has produced.

Reported gross, SoFi pair

$32.6M

Billboard Boxscore, two shows

Reported tickets sold

149,000

Paid tickets, not attendance

Calculated average paid ticket

$218.79

$32.6M ÷ 149,000

Paid tickets per show

74,500

Pair total ÷ 2

Two disciplines follow from this figure and are observed throughout the article. First, $32.6 million is not rounded up to $33 million except where secondary coverage is being described as having rounded it. Second, the SoFi average paid ticket is used as a scenario input and never as an observed value for any other market. It is one market's observation, produced by one price ladder, one fee regime and one local income distribution; carrying it across Almaty, Tirana, Faro or Jakarta would produce a number that looks authoritative and means nothing.

The census: 22 engagements, 26 dates, one gross

The census figure below holds every 2026 stadium engagement Live Index could verify to the official tour schedule, an official venue page or named reporting.⁴ One row is one market/venue engagement; several are multi-night, so the row count and the date count differ and both are stated. As of September 2: 11 engagements performed, representing 14 individual stadium dates; 6 engagements remain ahead, representing 8 dates, of which 2 are venue-disputed; 4 engagements covering 4 dates were removed; and 1 further market was proposed and declined without ever reaching an on-sale. That is 22 engagements and 26 ticketed dates in total. The pattern in the right-hand column is the finding: 26 dates have produced one reported ticket gross and seven published crowd figures. The rest have produced capacity numbers and nothing else — and capacity numbers are the most misleading data on the page if they are read casually.

One scope note. The wider 2026 run opened on January 30 and 31 at Monumental Plaza de Toros México in Mexico City, where the first night was reported sold out with more than 40,000 gathered and organisers expected more than 74,000 across the pair.³⁵ That is a large-format venue but not a conventional stadium, so those dates sit outside this census and outside every aggregate, capacity total and scenario in this article. They are recorded here only as scale context.

One housekeeping note on dates, because it affects what readers will find elsewhere. Some syndicated coverage of the September 1 announcement carried an October 3 date for Houston and a November 24 date for Phoenix. The official tour source and the venues give October 31 and November 21.⁴⁵ This article follows the official schedule and does not reproduce the syndicated errors.

What attendance actually tells us

Seven completed engagements have a published crowd figure. Istanbul's Atatürk Olympic Stadium show on May 30 drew 118,000 attendees, per Reuters.³ Madrid's Riyadh Air Metropolitano on July 30 drew 65,000, with more than 40,000 of them reported to have arrived from abroad, per RTVE.¹² Faro's Estádio Algarve drew approximately 34,000 on August 7, per Euronews.¹⁵ Almaty's Central Stadium drew 31,000 on August 15 — moved a day for weather — including roughly 7,700 visitors from more than fifty countries.¹⁹ San Antonio's Alamodome drew more than 60,000 on July 4.¹¹ New Orleans' Caesars Superdome drew 65,000 or more on August 28.²⁰²¹ And SoFi sold 149,000 tickets.¹

Adding those together gives roughly 522,000 admissions, of which the four highest-confidence anchors — SoFi, Istanbul, San Antonio and New Orleans — account for more than 392,000 on their own. Live Index publishes that aggregate under a deliberately awkward name: reported audience floor. It is a floor because Tampa's two nights, Tbilisi, Tirana and the two sold-out GelreDome shows published no count at all, so the true completed-run audience is materially higher. And it is not a paid-ticket total, because six of the seven components are attendance figures, which may include comps, guests, credentialled staff and industry. Only one component, SoFi's 149,000, is confirmed paid.

The Arnhem pair illustrates the limit of what a sold-out notice supports. GelreDome's own agenda archive marks both June 6 and June 8 Uitverkocht, and the venue's official information page gives a maximum of 41,000 in the round against 34,000 in the north-end concert configuration.²⁸ The honest inference is a configuration range of 68,000 to 82,000 across the two nights. The dishonest one is to pick 41,000, double it, and file it as attendance. This article publishes the range and excludes it from every aggregate.

Why fixed venue capacity lies

Istanbul is the cleanest demonstration on the run. The stadium's fixed seating is approximately 75,145; the reported crowd was 118,000.³ The difference is the field, sold as event inventory, and it means the bowl figure understates the crowd by about 57 percent. Faro shows the same direction at smaller scale: roughly 30,305 fixed seats against approximately 34,000 attendees.¹⁵ Madrid runs the other way — 65,000 attendees against a stated maximum concert capacity of 70,700, a utilisation ratio of 91.9 percent.¹²¹³ That ratio is a reference figure only. The denominator is the venue's maximum, not this event's manifest, so it cannot be reported as sell-through.

The Alamodome makes the point structurally rather than by example. The same building publishes 43,672 for a full-stadium concert configuration and 73,086 in the round — roughly 29,000 seats apart, in one venue, depending on where the stage goes.¹¹ A 360-degree central stage of the kind this run has used repeatedly sits at the high end of that spread. Without the event map, sellable inventory for any given night simply cannot be asserted, which is the structural argument Live Index has made about announced capacity and real supply and mapped across the country in the venue capacity atlas.

Prices: seven ladders, one currency each

Primary price ladders are published in enough markets to see the shape of the pricing strategy, though not enough to weight it. Portugal's official seller listed seated tickets from €139 to €209 by section, standing floor at €209, VIP Regular at €269, VIP Gold at €349 and Golden Circle at €450.¹⁶ Tirana ran standing regular at €120, bronze at €177, silver at €219 and golden circle at €313, with seated premium tiers into the low-to-mid €300s.¹⁷ Madrid's official tourism listing gave tickets from €150.¹⁴ Tbilisi's official event site gave tickets from GEL 400 against a stated 55,000 capacity.²³ Istanbul's official ticketing platform listed tickets starting at ₺5,400, and the Turkish business daily Dünya published the primary range as 5,400–35,000 TL.³⁶ St. Petersburg's disputed pair was announced with reported face values from about 9,000 RUB to about 160,000 RUB.²⁷³¹ New Delhi's removed date had reported ranges of ₹7,500 to ₹30,000.²⁹

These ladders are held in native currency and are not converted or compared. A €120 entry in Tirana and a €150 entry in Madrid are not the same economic object, and neither is comparable to 9,000 RUB without an event-date exchange rate and a purchasing-power adjustment. More importantly, none of them supports an average paid ticket. An average requires the allocation mix — how many seats sat in each tier and how many of those sold — and no promoter on this run has published one. The tiers show a strategy of steep vertical price discrimination in every market; they do not show what the average buyer paid.

Resale is a demand signal, not revenue

The two Soldier Field nights on September 3 and 4 offer the clearest live demand reading on the run, and the most dangerous numbers in the article. As of September 2, standard primary inventory for both nights was reported sold out, with resale get-ins around $432 to $466 for the third and around $658 to $687 for the fourth, depending on marketplace.¹⁰ Soldier Field's official capacity is 63,500, so the pair carries 127,000 seats of nominal fixed-seat reference capacity before any concert configuration.⁹

Those resale figures are not multiplied by anything anywhere in this article. A get-in is the cheapest listing a reseller has posted at a moment in time; it reflects reseller expectations, not buyer behaviour, and it is not promoter revenue in any part. The same wall applies to the Istanbul platform average of 14,142 TRY observed around that date: it is a listing-platform average, not a paid average, and it is excluded from all arithmetic here.³⁷ The economics of what resale prices do and do not measure are treated at length in the resale market audit.

New Orleans deserves the same time-stamped treatment rather than a resolution. Axios reported tickets widely available on August 27, the day before the show.²² WDSU and XXL reported a crowd of 65,000 or more, with XXL describing the show as sold out and produced by IKON Presents.²⁰²¹ Both are dated observations and both may be accurate: late demand, production holds released close to doors, walk-up movement and differing internal definitions of sold out would each produce this pattern. Live Index records the sequence and declines to pick a mechanism without evidence for one.

Gross: what is reported, and what can only be modelled

The remaining route has to be split before it can be modelled, because one market's status is genuinely contested. Reuters reported on August 17 that a pair of St. Petersburg dates had been announced for October 10 and 11 at Gazprom Arena, roughly 70,000 seats a night, with face values from about 9,000 RUB to about 160,000 RUB.²⁷ On August 25 the stadium said it would not host the two shows, while the promoter, Say Agency, said they had not been cancelled; the official tour schedule continued to list them.³¹⁴ Live Index records all three facts and classes the pair schedule-listed / venue disputed rather than resolving a dispute the sources have not resolved. The venue-aligned portion of the route — the Chicago pair, Jakarta, Houston, Arlington and Glendale — therefore carries 419,593 seats of nominal reference capacity: 127,000 in Chicago, 77,193 at Gelora Bung Karno's Main Stadium, 72,000 in Houston, 80,000 in Arlington and 63,400 in Glendale.^4678926 The venue-disputed pair would add a further 140,000 seats, taking schedule-listed inclusive exposure to 559,593 seats. Either figure is a capacity exposure metric: it is not a manifest, it does not account for stage kills or added floor inventory, and it assumes nothing about sell-through.

Applying the SoFi average paid ticket to the venue-aligned base produces the scenario matrix published below. The formula is stated in full and has exactly three inputs: reference capacity, an assumed sell-through, and an assumed average paid ticket. At 75 percent and $175, the modelled gross capacity is about $55.1M. At 85 percent and the SoFi-derived $218.79, about $78.0M. At 95 percent and $250, about $99.7M. On the schedule-listed inclusive base of 559,593 seats, the same three corners give about $73.4M, $104.1M and $132.9M; that is a bridge scenario conditional on the disputed pair proceeding, not the base case. The same arithmetic on the three new U.S. dates alone gives about $47.1M at full base inventory and about $40.1M at 85 percent.

The same restraint applies backwards. Istanbul's 118,000 attendees can be multiplied by illustrative averages — 8,000 TRY gives 944 million TRY, 12,000 TRY gives 1.416 billion TRY, 14,142 TRY gives about 1.669 billion TRY — but the top of that range uses a listing average and would materially overstate actual gross if the listed sample skewed premium, which listings routinely do. Faro's approximately 34,000 attendees at illustrative averages of €160, €190 and €220 give roughly €5.4 million, €6.5 million and €7.5 million. Both are sensitivity exercises, not estimates, and neither is entered into any total.

Travel demand, and the impact figure that is not revenue

Two markets published something rarer than attendance: where the audience came from. RTVE reported that more than 40,000 of Madrid's 65,000 attendees arrived from outside Spain.¹² Almaty's organisers reported roughly 7,700 of 31,000 attendees travelling from more than fifty countries.¹⁹ Those shares — well over half in Madrid, roughly a quarter in Almaty — describe a run functioning as destination inventory rather than local inventory, which is the same demand structure that drives festival tourism economics.

Kazakh reporting also put the concert's city and tourism effect at approximately 5 billion KZT.¹⁹ That figure is local economic impact: hotel nights, restaurants, transport, retail. It is not ticket revenue, it does not accrue to the artist or the promoter, and it is not converted or added anywhere in this article. Conflating destination spending with box office is one of the most common inflation mechanisms in live-event coverage, and Live Index treats the two as separate ledgers by rule.

Route attrition as an asset-utilisation problem

4 announced 2026 stadium dates never happened, and a further market was proposed and declined. New Delhi on May 23 at Jawaharlal Nehru Stadium — 60,000 fixed seats per the Sports Authority of India — was postponed indefinitely following security directives, after primary prices of ₹7,500–₹30,000 had been reported.²⁹ Marseille's Orange Vélodrome date on June 11 was postponed until further notice, with the notice carried on the official venue page; no capacity figure is entered for it because none was verified.³² Chorzów's Superauto.pl Stadion Śląski date on June 19 was cancelled in a statement published by the stadium itself; the venue's official configuration page gives 54,378 fixed stand seats with a concert floor of roughly 30,000, and those two figures are not summed here into an event manifest.³³ Basel's St. Jakob-Park never became a ticketed date at all: the club declined to host a proposed June engagement, and it is recorded as a proposal rather than a cancellation.³⁴ Reggio Emilia's RCF Arena date on July 18 was cancelled after the performance was prohibited; the venue's own pre-event page had stated an intended audience of 103,000, which is an announced event plan rather than sold inventory or attendance, and is the largest single block of removed capacity exposure on the route.³⁰ Almaty's August 14 date moved a day for weather and performed.¹⁹

Read as economics rather than as controversy, this is an asset-utilisation pattern. Stadium touring commits fixed production and routing costs against dates that can be removed by authorities, insurers, venues or weather with little notice, and a removed stadium date is a large block of committed capacity earning nothing. The same run that added 215,400 seats of U.S. capacity in a single September announcement lost 4 dates across four countries earlier in the year and had a fifth market decline it outright. Attrition runs in both directions and is a real cost of scale, not a footnote to it.

What this says about stadium-scale economics in 2026

The precise finding is this: a single artist generated verified stadium-scale demand across 14 performed stadium dates spanning North America, Europe and Asia in 2026 while producing exactly one public box-office observation, and that observation — $32.6 million against 149,000 paid tickets at SoFi Stadium — is the only point in the entire dataset where price and quantity are both known. Everything else in public is a crowd count, a bowl capacity or a resale listing. No gross outside the SoFi pair has been located anywhere in the public record audited here. Stadium touring at this scale is therefore structurally unmeasurable from outside: it runs on market-specific price ladders that are never weighted, oversized central-stage configurations that make published capacities unusable, promoter reporting that stops at a single Boxscore entry, and routing volatile enough that announced inventory is not a reliable predictor of performed inventory.

The measurable consequence for anyone modelling this business is narrow and firm. Reported audience on the completed run is at least 522,000 admissions and rising with every unpublished count; venue-aligned remaining reference capacity is 419,593 seats, or 559,593 if the disputed St. Petersburg pair proceeds as listed; and the honest range of modelled gross across the venue-aligned base spans roughly $55.1M to $99.7M depending on two assumptions neither of which is observable. A range that wide, derived from one observed average paid ticket, is not a failure of the model. It is an accurate measurement of how little of stadium-scale touring economics is public.