RaoscaffResearch
Prediction Series · Lock · Issue P-323
Prediction Series · P-323

Two EBITDAs, NZ$61m apart.

90% probability SkyCity's FY2027 underlying EBITDA reaches at least NZ$150m, after NZ$181.6m in FY2026. The same year produced a reported EBITDA of NZ$120.5m. No FY2027 guidance was given, and a NZ$30m cost-out programme is the main visible support.

Type · Prediction Lock · level-anchored floor, casinos and tourism Locked · 2026-09-03 · before the FY2027 result Resolves · ~2027-08-20 · SkyCity Entertainment FY2027 results (skycityentertainmentgroup.com) Scored · binary: reported FY2027 UNDERLYING EBITDA >= NZ$150m yes/no
SkyCity FY2027 underlying EBITDA · our locked floor
NZ$150m
full-year underlying EBITDA · vs NZ$181.6m in FY2026

SkyCity Entertainment Group delivered FY2026 underlying EBITDA of NZ$181.6m, down 22.3% year-on-year and in line with revised May guidance, as flat revenue masked a 5.9% decline in gaming revenue offset by non-gaming growth. Reported EBITDA was NZ$120.5m, down 44.2% or NZ$95.6m, owing to several significant accounting adjustments. No FY2027 earnings guidance was provided due to macro uncertainty, though management targets a material uplift from new online market entry and cost initiatives. A cost-out programme targets NZ$30m in FY2027 and NZ$70m in FY2028, and asset sales are expected to generate NZ$275-300m by December 2026. Source: SkyCity Entertainment Group FY2026 annual results.

— 1 · The Locked Call

SkyCity's FY2027 underlying EBITDA is at least NZ$150m — P = 0.90.

We lock a binary: SkyCity Entertainment Group's reported underlying EBITDA for fiscal 2027 is NZ$150m or higher. Confidence 90%.

There is no FY2027 guidance to anchor to. FY2026 underlying EBITDA was NZ$181.6m, so our threshold sits NZ$31.6m — about 17.4% — below the banked actual. That is a wide buffer, and a business that just fell 22.3% with no guidance published deserves one.

— 2 · The word 'underlying' is doing NZ$61m of work

NZ$181.6m, or NZ$120.5m.

SkyCity's FY2026 produced two EBITDA figures. Underlying EBITDA of NZ$181.6m, down 22.3%. Reported EBITDA of NZ$120.5m, down 44.2%, a fall of NZ$95.6m attributed to several significant accounting adjustments. The gap between the two is NZ$61.1m — a third of the underlying number.

A lock written as SkyCity FY2027 EBITDA above NZ$150m would be unscoreable. If FY2027 produces underlying NZ$165m and reported NZ$130m, the same threshold resolves HIT on one basis and MISS on the other, and whoever scores it a year from now gets to choose. The criterion below names UNDERLYING and rules out reported in terms.

This is the same class of ambiguity as Fletcher Building at P-306, where EBIT before significant items included about NZ$52m of property sales, and the lock deliberately scored the ex-property figure. Where two versions of a number exist, the lock has to say which one it means before the result is known, not after.

— 3 · No guidance, a declining core, and one quantified support

Gaming revenue fell 5.9%. The cost-out is NZ$30m.

SkyCity declined to provide FY2027 earnings guidance, citing macro uncertainty. That places it in the same category as Air New Zealand, which this series scouted and dropped at P-307 for exactly that reason. The difference is that SkyCity left behind quantified, forward, company-stated commitments that a level-anchored lock can lean on, where Air New Zealand's single forward figure was a bespoke disclosure unlikely to be restated in the same form.

The commitments are specific: a cost-out programme targeting NZ$30m in FY2027 and NZ$70m in FY2028, and asset sales expected to generate NZ$275-300m by December 2026. Management also points to new online market entry as a source of uplift.

Set against that, the core is shrinking. Gaming revenue fell 5.9% in FY2026, with total revenue flat only because non-gaming grew to offset it. The NZ$30m cost programme is roughly the size of the buffer we have taken, which is deliberate: the lock does not require the cost-out to succeed, it requires that the business not fall a further sixth even if the cost-out delivers nothing.

The residual 10% is a second consecutive year of gaming revenue decline that the cost programme fails to offset, or the online market entry consuming more investment than it returns inside FY2027.

Locked on 2026-09-03 — scored against SkyCity's reported FY2027 underlying EBITDA.

RAOSCAFF locks P-323 on 2026-09-03, before the FY2027 result. Scored against underlying EBITDA as reported by SkyCity Entertainment Group for the year to 30 June 2027.

Locked
2026-09-03 (commit timestamp on origin/main)
Resolves
~2027-08-20 — SkyCity Entertainment Group Limited FY2027 annual results
Source
SkyCity Entertainment Group Limited FY2027 annual results, underlying EBITDA in New Zealand dollars (skycityentertainmentgroup.com investor centre)
Scored by
Binary: YES if reported FY2027 UNDERLYING EBITDA is NZ$150m or greater; NO if below. UNDERLYING EBITDA on the same basis as the NZ$181.6m FY2026 comparative — explicitly NOT REPORTED EBITDA, which was NZ$120.5m against underlying NZ$181.6m in FY2026, a gap of NZ$61.1m. Also NOT net profit after tax, NOT revenue, NOT proceeds from asset sales (expected at NZ$275-300m by December 2026), NOT the NZ$30m FY2027 cost-out target, and NOT a single property or segment. SkyCity's financial year ends 30 JUNE 2027.

Underlying NZ$181.6m and reported NZ$120.5m for the same year. A threshold between them would resolve both ways. The criterion says which.