HScreen

rendered 2026-08-23 14:15 UTC 1 input MISSING

META · Meta Platforms, Inc. 4d old

conviction3
coverage tierT1-FULL
horizon12 months

Return distribution probability-weighted scenarios, never collapsed to a point

-22%23%
+12%45%
+42%32%
EV +13.78%
why these probabilities: Built forward from growth rate to EPS to multiple to price, off 547 USD (2026-08-18). BASE (+12%, 45%) — the market's expectation delivered. Consensus FY2026 revenue +27.3% and EPS +39.6%, FY2027 revenue +19.9% and EPS +7.2% (fmp:/stable/analyst-estimates?symbol=META, as_of 2026-08-19, annual periods, 41-43 contributing estimates, revenue dispersion 2.6% FY2026 / 7.5% FY2027 — tight enough to treat the point estimate as meaningful, unlike the FY2028 37.3% dispersion which I do not use). On that trajectory 22.8x trailing EPS holds roughly flat as EPS grows and the multiple absorbs the FY2027 EPS deceleration, giving ~610-615, i.e. +12%. That is deliberately BELOW the 723.69 street target (+32%): the street mark requires multiple expansion on decelerating FY2027 EPS, which the capex-discipline regime is currently refusing to grant. I am describing what is priced in, not competing with it. BULL (+42%, 32%) — FY2027 revenue growth materially above the +19.9% consensus, call it ~26-28%, driven by the larger ad ranking models management says are already deployed plus compute explicitly allocated to core-business optimisation. Critically the cost is ALREADY IN the numbers — this is a revenue beat on sunk capex, so it flows to operating income at high incremental margin off a 41.4% base. That both beats EPS and retires the ROI objection the de-rating rests on, so the multiple re-rates
not built — options-implied distribution on the same axis, with the KL number and where the disagreement sits
needs: desk-options implied bands (v1.4 Part 4a) — activates at 60% coverage; currently 33%

Visibility how far ahead this name can be seen

not built — visibility score
needs: visibility_score.py

Trust how much weight this name's own record can carry (credibility Z)

not built — credibility
needs: credibility.py

Sizing survival-first size (half-Kelly) vs the enforced ladder

ladder (enforced)3%
kelly-derived (report-only)0.34%
divergence2.66pp
inputsEV(ann) 13.78% · σ 23.47 · Z

Payout durability does the distribution survive a downturn?

PASS
DPS trajectoryinsufficient_history
interest cover now71.48x
interest cover at FORWARD rates21.21x

Falsifiers what would prove this wrong

Larger AI ranking models materially accelerate core ad revenue above consensus
Data centre capex converts to earnings rather than permanently impairing FCF conversion
The EEA/Switzerland revenue base (16% of worldwide advertising) survives European Commission engagement without material impairment
Advertiser dollars continue rotating toward Meta rather than away
The negative estimate revision tape is a capex-timing markdown, not a demand markdown

Conflicts surfaced, never netted out (T4)


I underwrite a positive probability-weighted expected return (+13.4%) while the estimate revision tape is unanimously negative — EPS 0 up / 4 down, breadth -1.0, median move -4.6%; revenue 0 up / 4 down, breadth -1.0, me		high

Consensus is Buy with 52 of 65 analysts at buy or strong-buy and a 723.69 target implying +32.3%. My expectation is +13.4%, 18.9pp below. The disagreement is specifically about the MULTIPLE, not the revenue line — I use 		high

Internal inconsistency inside the street's own position: rating is overwhelmingly Buy and the target is 32% above spot, yet every recorded revision in the trailing year was DOWN on both EPS and revenue. Ratings and targe		medium

My balance sheet score of 8 rests on FY2025 returns and liquidity, but the pack's third-party evidence says big-tech share count reductions have slowed or reversed as capex absorbs cash flow, and Meta's own disclosure is		medium

The bull leg's central mechanism — larger AI ranking models lifting core ad revenue — is management-stated but UNQUANTIFIED. Meta says the work is underway and that compute is allocated to core-business optimisation; now		high

The case against independent adversary, different model family (P6) weakened

Version history from the research record

archived versions3
archivedhashbytes
2026-08-17T14:27:28b0646e38d44437,022
2026-08-17T15:40:00ba99daebd90237,206
2026-08-19T08:32:01205f3a492c4236,636

What changed in the latest version

-as_of: 2026-08-17
+as_of: 2026-08-19
+provenance:
+ producer: analyst-aisemi
+ generated: 2026-08-19
+ evidence_pack: META evidence pack 2026-08-19T07:35:49Z (build_evidence_pack.py)
+ price_basis: 547 USD close 2026-08-18, fmp:/stable/profile?symbol=META
+ blind_rescore: true
- value: 'Meta Platforms operates the Family of Apps (Facebook, Instagram, WhatsApp, Messenger, Threads)
- monetised almost entirely through auction-based advertising, plus Reality Labs (VR/wearables, structurally
- loss-making). FY2025 revenue $200.97bn, operating income $83.28bn (41.4% operating margin), net income
- $60.46bn, reported EPS $23.98. ROE 27.8%, ROIC 18.0%, current ratio 2.60, EV/EBITDA 16.4x, EV/Sales
- 8.5x on FY2025 metrics. Market cap $1.503tn, beta 1.243, 75,472 employees. Quarterly dividend $0.525
- (raised from $0.500 in Q1 2025), trailing $2.10 — c.0.36% yield, immaterial. The company is mid-way
…diff truncated
Rendered from canonical artifacts only (T1) — no terminal database exists. Every figure is read from the system's own files at render time; where an artifact is missing the panel names the dependency rather than showing an empty box. No execution path exists anywhere in this interface (T5), not even a link.