How the calculators compute results

Federal ordinary income. Marginal walk across the 2026 IRS brackets (single, married-filing-jointly, head-of-household) published in IRS Rev. Proc. 2025-32. Income added by an exercise or vest is walked from the user's stated baseline through the brackets it crosses.

Federal long-term capital gains. 0% / 15% / 20% brackets per filing status, plus the 3.8% Net Investment Income Tax (NIIT) above the MAGI threshold ($200K single, $250K MFJ).

State income + LTCG tax. Per-state ordinary brackets for 2025 (most states publish 2026 brackets in late 2026). Long-term capital gains follow ordinary brackets in most states; preferential treatment modeled for HI, ND, SC, WI, AR, NM. Washington's 7% LTCG-only tax above $270K modeled separately.

Payroll taxes (FICA). When still employed: 6.2% Social Security up to the 2026 wage base ($184,500), 1.45% Medicare on all wages, plus 0.9% Additional Medicare on wages above $200K (single) / $250K (MFJ).

Source verification. Every 2026 federal constant above — ordinary brackets, long-term capital gains breakpoints, the AMT exemption and phaseout, and the FICA wage base — is asserted against its IRS Rev. Proc. 2025-32 published value by an automated conformance test. A stale or mistyped figure fails the build before it can ship.

Alternative Minimum Tax follows IRC §55: AMTI = ordinary income + ISO bargain element. Federal: 26% up to $244,500 of AMTI (post-exemption), 28% above. Exemption ($90,100 single / $140,200 MFJ for 2026, per IRS Rev. Proc. 2025-32) phases out at 50¢ per $1 of AMTI above the threshold ($500,000 single / $1,000,000 MFJ; the post-OBBBA rate applies as of 2026 per IRC §55(d)(4)).

State AMT modeled for the four states currently in the registry: CA, CO, CT, and MN. CT piggybacks on federal AMT (lesser of 19% of federal TMT or 5.5% of federal AMTI); CA, CO, and MN apply flat rates above a state-specific exemption. Other states: state TMT treated as zero.

The AMT-ISO calculator's optimizer searches for the share count and timing that maximizes net final value (after a qualifying disposition) net of AMT premium. Federal AMT credit recovery is modeled when regular federal tax exceeds federal tentative minimum tax in a future year.

Cash return rate. The optimizer time-values the cash-tax stream at a user-set rate (default 5%, roughly the short-Treasury yield). At rate = 0 the math collapses to a nominal-sum objective; above zero, early AMT premiums get penalized against spread schedules and late credit recoveries get discounted.

Put and call premiums use closed-form Black-Scholes with the underlying's implied volatility. When a tracked options chain is available, σ is read directly from the chain, interpolated to the user's strike (set by their protection level) and tenor.

When no chain is available (bare tools page, pre-IPO companies, untracked public tickers), the protective put tool falls back to a sector-typical realized volatility (e.g. 36% Tech Software, 45% Semiconductors). The concentration tool's hedge math uses the same sector vol scaled by 1.20 to approximate the implied-over-realized premium typical for buying single-name puts, since concentration does not pull chain data directly. This fallback is approximate by design; real quotes vary with skew, tenor, and strike. Each tool surfaces the σ value it used so you can sanity-check before committing.

Single-stock expected return. When a public ticker is selected, the slider seeds from a blend of trailing-window CAGRs (2-year chain history, 5-year, 10-year where available, anchored at a 5-year mid-horizon for the wind-down calculators). On the bare page (no ticker selected) and for pre-IPO companies, the default is 20%/yr. The slider exposes the full range so users with stronger or weaker views can adjust.

Diversified market return. SPY's trailing return seeds the slider, with a haircut applied for volatility drag (½ × variance × T). 10% is the unseeded default.

Volatility drag (haircut). The expected-sale-price haircut maps from σ via the lognormal half-variance correction (1 − exp(−σ² × T / 2)), growing monotonically with the hold period. On chain-equipped public tickers it is seeded from the chain's at-the-money implied σ over the user's hold horizon; without a chain, it defaults to a 20% haircut. The same mapping powers the concentration tool's drag on the held concentrated position (diversified reinvestment is left unhaircut since it already represents a vol-averaged return).

The QSBS checker runs the eight statutory tests against the user's facts (entity type, acquisition method, asset category at issuance, industry, active-business posture, holding period, adjusted basis, expected gain) and returns a verdict, the applicable exclusion percentage, and the federal tax saved.

Four exclusion eras are covered: 50% (acquired 1993-2009), 75% (acquired 2009-2010), 100% (acquired 2010 through 2025-07-04), and the OBBBA tiered regime (50% / 75% / 100% at 3- / 4- / 5-year holds, with caps lifted to $15M / $20M for stock acquired on or after 2025-07-05). State conformity is tracked: non-conforming (CA, AL, PA, MS), partial (HI, MA), and NJ's 2026-01-01 conformity switch are all modeled. Pre-2010 acquisitions retain the 7% AMT preference on the excluded gain.

• Multi-scenario optimization. Each free tool projects one expected case. The full beta product runs bullish, neutral, and bearish in parallel and finds the schedule that performs best across all three.
• Joint optimization across grants. Each tool models a single grant or position. Beta solves the multi-year exercise + vest + sale + hedge schedule across your whole equity portfolio.
• Double-trigger RSUs. Most pre-IPO RSU grants don't settle until a liquidity event. The RSU calculator models single-trigger only.
• $100K ISO limit (§422(d)). Annual ISO grants whose FMV at grant exceeds $100K are reclassified as NSO from the start. The free tool assumes you're within the limit.
• Multi-state moves and dual residency. Single state per scenario; relocation timing not modeled.
• Disqualifying dispositions. ISO projections assume a qualifying disposition (≥1 year from exercise, ≥2 years from grant). Early sales convert to NSO ordinary-income treatment — out of scope.
• §83(b) elections. Out of scope.

Every figure on this page traces to a citable source, and the engine behind the tools is built to be checked, not taken on faith.

• Sourced to IRS authority. Brackets, Alternative Minimum Tax exemptions, and thresholds come from the 2026 IRS Revenue Procedures cited above; the QSBS checker runs the eight statutory Section 1202 tests.
• Deterministic. The same inputs always produce the same result. No language model sits in the calculation path; the recommended schedule is produced by exact deterministic optimization and validated against brute-force results on tractable cases. See the verification page for the math beside the IRS publications and an independent tax engine.
• Private by design. No account and no stored personal data; the numbers you enter are used to compute your result and are never sold, shared, or retained.
• Open source. The calculation engine is public under the MIT license at github.com/AlvisoOculus/optionsahoy-mcp. Audit it yourself.
• Tested. More than a thousand automated tests cover the federal and 50-state tax logic, AMT credit recovery, and the option-pricing model.