Setup

FICO printed Q2 FY2026 on April 28. Stock is down 47% over twelve months and trades at $1,044.81 — 13% of its 52-week range, 19.5x forward P/E against a 91%-margin Scores monopoly and a Platform business growing +49% YoY. Sell-side mean target sits at $1,598 (+53%). The filing showed acceleration in the metrics that test the thesis, not deceleration.

What the filing says

Platform ARR reached $348.8M, +49% YoY — accelerating from +33% the prior quarter. Platform DBNRR jumped 14 points sequentially to 136% (was 122% in Q1, 110% one year ago). At trajectory, Platform ARR crosses non-platform ARR ($440M, -8% YoY) by Q4 FY2026. ACV bookings $66.1M H1 vs $43.0M prior year (+54%) — forward pipeline leading reported ARR.

Scores revenue $475M (+60% YoY), B2B Scores +72%, segment operating margin 91% (up from 89%). Driver disclosed verbatim: "higher unit price and an increase in volume of mortgage originations." Bureau revenue concentration jumped to 64% (from 52% in prior year Q2) — all three credit bureaus individually exceed 10% of FICO revenues. Total revenue $691.7M (+39%), Q2 diluted EPS $11.14 vs $6.59 (+69%).

CEO William Lansing terminated his 10b5-1 trading plan March 26 — for ≈55,000 shares running through December 2026, established November 2025. Form 4 verification confirms zero personal code-S sales between plan establishment and termination. Stock was $1,183 on the termination date; ≈55,000 shares of authorization went unused.

Software segment margin compressed to 29% (vs 31% PY) — second consecutive quarter at this level, attributed to cloud infrastructure costs and mix shift toward ratable SaaS.

Cross-ticker math conserves

TRU 10-Q (filed 4/28) cites "FICO mortgage royalties" as primary cost driver, -100bps consolidated EBITDA drag. EFX 10-Q (filed 4/21): USIS margin 21.1% → 20.2% "primarily due to higher royalty costs in USIS." Combined bureau cost increase ≈ $180M YoY ≈ FICO's B2B Scores revenue increase of $175M. Math is conserved across three independent SEC filings, same calendar quarter. The royalty flywheel ($4.95 → $10 per-score, 20x since 2018) is documented from supply and demand sides simultaneously.

What the market thinks

Trailing-250d factor regression: 93.4% idiosyncratic variance, SPY β +2.97, QQQ β -1.43 — tape priced FICO as AI-loser, trading inversely to tech. Residual α = -70.7% annualized — that's the compression in price.

Options: ATM IV 58.5% (59th percentile of 52-wk range), max pain $1,040 at spot. P/C OI 0.79 (bullish at the open-interest level); today's flow P/C 1.46 with 29 unusual put strikes (bearish flow into May 15 expiry). Positioning is mixed.

Sell-side: 15 buy/strong-buy vs 5 hold, 1 sell. The most recent print was the fourth consecutive earnings beat (+13.9%).

Where the gap is

Reverse-solving market-implied scenario probabilities at $1,044 against fundamentally derived targets (≈$1,915 bull / ≈$1,359 base / ≈$1,004 muddle / ≈$644 bear) yields a Bull probability of ≈15-25% depending on scenario-midpoint assumptions. Filing-grounded reading suggests 35-45%. The gap is roughly +15-25pp on the bull case — wider than scenario-precision noise.

EV at $1,468 implies +40% over 12-18 months, Sharpe 0.76 (under the 0.5-1.0 plausibility ceiling). Notably, my EV is below sell-side mean target $1,598 — alpha here is not "see what analysts miss" but "tape hasn't priced what analysts already see."

Why the gap exists

AI-disruption narrative is sticky; the QQQ -1.43 beta says the market processed FICO as Zoom-analog. VantageScore overhang carries an uncertainty premium even though GSE submission share has held <5% nine months after FHFA authorization. Hedge fund net software exposure is at record lows (GS Prime data) — flow positioning, not fundamentals. The cross-ticker math conservation — bureaus literally naming FICO as their cost driver — is buried in TRU/EFX 10-Q footnotes that retail and most sell-side don't synthesize.

Risks (ranked)

1. VantageScore GSE share breaks 15% by year-end — forces LLPA grid reconciliation, moat erodes (low probability: 9mo at <5%, but 85% is not 99%)
2. Sherman §2 antitrust pricing remedy — claim survived dismissal Nov 2024, in discovery, no schedule disclosed
3. Mortgage cycle reversal — rate spike kills refi volume, flywheel stalls
4. Software margin compression below 27% — cloud opex drag becomes structural
5. Leverage approaches 3.5x covenant — $3.64B debt, coverage 9x today, but $1.5B issued in 12mo at 6.0-6.25%; buyback flywheel breaks if credit tightens

Catalysts

Cluster point: late July through early August 2026 — three predictions resolve in a two-week window.

What would change our mind

• Platform DBNRR drops below 120% in Q3 FY2026 — structural deceleration confirmed
• FHFA report shows VantageScore GSE submission share at 15%+ in any quarter — regulatory moat eroding
• Sherman §2 ruling or settlement forces structural pricing concession
• Mortgage origination volume drops >30% YoY for two consecutive quarters
• Bureau revenue concentration crosses 70% AND one bureau publicly threatens defection
• Stock breaks $850 on volume without identifiable news — tape priced something the filings haven't disclosed

Evidence