Heston vs Black-Scholes

Side-by-Side

When to Use Black-Scholes

• Implied volatility extraction. The standard convention for quoting and comparing options is via BS-implied IV. To compute IV from a market price, you invert BS analytically. Even when you intend to price with Heston, you'll still read and report results in BS-implied IV space.
• Liquid ATM options. For a SPY 50-delta option 30 days out in a normal vol regime, BS pricing is within a few cents of any smile-aware model. The complexity premium of Heston is wasted on these strikes.
• Real-time Greek aggregation. Computing portfolio gamma, vega, and theta across thousands of positions in real-time requires speed. BS analytical Greeks are 100-1000x faster than Fourier-method alternatives, and the precision difference is small for ATM books.
• Baseline comparison. Every smile-aware model is most cleanly evaluated by computing the BS-implied vol it produces and comparing to the BS-implied vol of the listed market price. The deviation tells you the model's smile-fitting quality.
• Quick mental math. The Brenner-Subrahmanyam ATM straddle ≈ 0.8 × S × σ × √T identity is BS-derived and useful for on-the-fly expected-move calculations.

When to Use Heston

• Smile-aware pricing. When the strike is OTM and the smile matters (deep ITM puts, far OTM calls, tail-protective structures), Heston produces materially better prices than BS. The error from using BS at the 10-delta SPX put can exceed 30-50% of the option's theoretical value.
• Exotic options. Variance swaps, cliquets, forward-start options, and barrier options are sensitive to forward-smile dynamics. BS produces zero forward smile; Heston produces realistic forward smile that matches market behavior of these exotic structures.
• Volatility surface calibration. If you need a single consistent model that prices the entire surface (all strikes, all expirations), Heston is the standard 5-parameter starting point. SABR is faster per-expiration but doesn't directly produce term-structure dynamics.
• Long-dated options. Mean reversion of variance matters for options expiring in 1+ years. BS implicitly assumes vol stays at today's level; Heston explicitly mean-reverts toward the long-run level θ, producing more realistic long-dated pricing.
• Volatility-of-volatility products. VIX options, VVIX-based structures, and any vol-of-vol-sensitive position requires a model with stochastic volatility. Heston is the canonical choice.

Where They Agree

For ATM strikes at intermediate maturities (typically 7-90 DTE) in calm regimes, Heston and Black-Scholes produce nearly identical prices, often within 1-2% of each other. The ATM IV term structure of Heston converges asymptotically toward the BS-implied IV at long horizons (the long-run mean variance √θ).

Both produce identical Greeks structure: delta, gamma, theta, vega all exist and behave consistently. The numerical values differ at OTM strikes (especially for vega and second-order Greeks), but the conceptual interpretation is the same in both models.

Both models are arbitrage-free under their respective assumptions. Both rely on the same risk-neutral pricing framework: option value equals the discounted expected payoff under a probability measure that makes the discounted underlying a martingale.

Where They Diverge

• OTM put pricing. A 5-delta SPX put 30 DTE: BS prices ~$0.40, Heston (calibrated to surface) prices ~$0.85. Market price typically matches Heston within 5%; BS underprices by 50%+.
• Term structure of IV. BS produces a flat term structure (one σ for all expirations). Heston produces an explicit term structure determined by κ, θ, and v₀. Practical impact: pricing of calendar spreads and term-structure trades.
• Forward-smile shape. BS produces a flat forward smile (smile at all future dates is flat). Heston produces a non-trivial forward smile that persists. Practical impact: cliquet pricing, forward-start option pricing, and reset structures all require Heston-class models.
• Vol-of-vol pricing. BS prices vol-of-vol at zero. Heston explicitly prices vol-of-vol via ν. Practical impact: VIX options, butterfly structures, and any product whose payoff depends on the variability of vol cannot be priced under BS.

Why They're Often Confused

Both models are usually quoted in BS-implied IV space, which obscures the underlying model differences. A trader looking at "current IV is 14%" doesn't see whether that came from BS or Heston: it's the same number. The difference shows up only at non-ATM strikes or in vol-sensitive structures.

Heston is sometimes described as "Black-Scholes with stochastic vol," which is technically correct but misleading: Heston is a fundamentally different stochastic process with five parameters versus one. The two models share Brownian-motion structure for the spot process, but Heston's variance dynamics introduce qualitatively new behavior.

The real practitioner question is rarely "Heston or Black-Scholes?" Heston is used in addition to BS, not instead of. BS handles fast Greek computation and IV reporting; Heston handles smile-aware pricing and surface-consistent modeling. Both run side-by-side in production systems.

Further Reading

• Black-Scholes Model Documentation
• Heston Model Documentation
• Pricing Model Landscape (overview map)
• What Is Volatility Skew?
• What Is Vol of Vol?
• SABR vs Heston

References

• Black, F. and Scholes, M. (1973). "The Pricing of Options and Corporate Liabilities." Journal of Political Economy, 81(3), 637-654.
• Heston, S. L. (1993). "A Closed-Form Solution for Options with Stochastic Volatility with Applications to Bond and Currency Options." Review of Financial Studies, 6(2), 327-343.
• Carr, P. and Madan, D. (1999). "Option Valuation Using the Fast Fourier Transform." Journal of Computational Finance, 2(4), 61-73.
• Gatheral, J. (2006). The Volatility Surface. Wiley.

This is one of the model-vs-model comparison pages. For the full landscape of pricing models and their relationships, see the Pricing Model Landscape.