Ten questions. Five minutes.
One number that reframes everything.
Nobody ever shows you the odds.
"Potential is everywhere. Conditions are not."
No data stored. No account required. Just the math. This model reflects the US structural sorting system and is designed for people who grew up primarily in the United States.
Mapping your chain
Your cumulative odds
Your structural portrait
What set your starting difficulty
Each row multiplies your starting-difficulty score. A higher score means a steeper start. This score sets your percentile and mode.
Where you landed, given that start
The typical path
Continue the conversation
Go deeper with an AI conversation.
Your portrait is a starting point. The deeper experience is a structured conversation that helps you see your specific chain of contingencies — the people, programs, accidents of timing — that account for where you landed relative to where the system predicted you would. It takes 30–60 minutes and costs nothing.*
1
Click the button below to copy your personal prompt to your clipboard. It contains your name and percentile so the conversation starts where your portrait left off.
2
Open a free AI chat. Any of these work — no account required for basic use:
Paste the prompt into the chat and press send. The conversation will begin immediately. Answer honestly — the experience only works if you do.
4
The conversation will end not with a verdict but with a question you carry forward. There's no right conclusion — only a clearer view of the system that shaped you.
Your prompt will appear here after you copy it.
* "Free" isn't free. Running AI models requires significant compute — energy, water for cooling, hardware at scale. The free tiers offered by AI companies are subsidized by investors and infrastructure costs that are real, even when they don't appear on your bill. Using these tools thoughtfully is its own kind of literacy.
Methodology & Sources
How the odds are calculated.
This tool models the structural sorting system of the United States using a chain of conditional probability multipliers — each one derived from peer-reviewed research on the relationship between demographic starting conditions and life outcomes. The evidence base draws from three tiers: administrative and government data (descriptive baselines), top-tier academic research using quasi-experimental and natural experiment designs (causal estimates), and foundational sociological theory (mechanisms). Where sources are descriptive rather than causal, this is noted. The model also applies three empirically-grounded interaction effects where variables are known to moderate each other rather than compound independently — these appear as separate rows in the chain. The model does not claim to be a complete account of any individual life. It claims to make visible what the data consistently show: that starting conditions are among the strongest predictors of where people land, and that those conditions are not equally shared.
The model
Structural multiplier approach
Each starting condition (gender, race and ethnicity, parents' origin and education, neighborhood, household stability, access to guidance) is assigned a multiplier reflecting relative structural difficulty. The multipliers are informed judgment calls anchored in the research below, not fitted statistical estimates. They compound across inputs, with three documented interaction effects. The product is your starting-difficulty score, which sets your percentile and difficulty mode. Birth decade and current income are context only and are not part of the score.
Percentile calibration
The starting-difficulty score is converted to a 1–99 index on a log scale. It is a scaled index of how hard the starting conditions are, not a measured population percentile. As a rough check, using approximate US population shares for each input and treating them as independent, the median American lands near the 50th percentile on this scale. Those shares are rough assumptions, and in reality the inputs are correlated.
Interaction effects
Why variables interact rather than simply compound
The model applies three empirically-grounded interaction adjustments where the research shows variables moderate each other rather than operating independently. These appear as separate rows in the chain display, styled distinctly from the base multipliers.
School quality as partial substitute for parental educational capital
Applied when: neighborhood is middle-class or affluent AND parental education is high school diploma or less.
Wodtke, G.T., Harding, D.J., & Elwert, F. (2011), "Neighborhood Effects in Temporal Perspective," American Sociological Review, 76(5), 713–736. [growing up in the most disadvantaged neighborhood lowers high school graduation by 20 percentage points — implies affluent neighborhoods provide substantial independent educational lift] — Chetty, R. & Friedman, J. (2011), cited in Chetty & Hendren (2018): 40% of the correlation between parental earnings and children's future earnings is attributable to school quality differences. This means school quality — provided by affluent neighborhoods independently of parental education — substitutes for approximately 40% of the parental education effect.
Interaction 2 — Race moderates parental education advantage for Black Americans (× 1.4)
Diminished return on parental educational capital due to persistent discrimination
Applied when: race = Black AND parental education is bachelor's degree or higher.
Assari, S. & Mistry, R. (2018), "Educational Attainment and Intergenerational Mobility; Role of Race and Gender," Behavioral Sciences, 8(11), 107. [significant interaction between race and parental education: parental education produces smaller intergenerational mobility gains for African Americans than for white Americans] — Mechanism: high social status increases contact with white-dominated institutions, which increases exposure to discrimination. As a result, returns to parental educational attainment are systematically smaller for Black Americans. Chetty et al. (2024, "Changing Opportunity"): racial convergence in mobility occurred largely through improved chances for Black children to escape poverty, not through increased access to the top of the income distribution — confirming that upper-end advantages are differentially accessible by race.
Interaction 3 — Low-income neighborhood compounds with low parental education via peer community effects (× 1.2)
Social network homophily amplifies co-occurring disadvantages
Applied when: neighborhood is low-income or frequent moves AND parental education is less than high school.
Chetty, R., Friedman, J., et al. (2024), "Changing Opportunity: Sociological Mechanisms Underlying Growing Class and Shrinking Race Gaps in Economic Mobility," Opportunity Insights Working Paper. [children's outcomes most strongly associated with parental employment rates of peers in same race and class group; effects decay sharply for adjacent cohorts, consistent with social network homophily — low-income children in low-parental-education households face compounded peer community effects beyond simple resource addition] — Crowder, K. & South, S. (2003): for Black adolescents from single-parent households and white adolescents from low-income households, the association between neighborhood disadvantage and dropout risk is particularly strong, consistent with compounding rather than additive effects.
Sources by variable
Race & ethnicity multipliers
Racial wealth gap, educational attainment, incarceration disparities, and hiring discrimination
Descriptive: Bhutta et al. (2020), "Disparities in Wealth by Race and Ethnicity in the 2019 Survey of Consumer Finances," Federal Reserve Board. — NCES (2022), "Status and Trends in the Education of Racial and Ethnic Groups." — Carson (2020), "Prisoners in 2019," Bureau of Justice Statistics. — Pew Research Center (2016), "On Views of Race and Inequality" [note: this source documents perceptions of inequality, not structural inequality itself].
Causal: Kline, P., Rose, E.K., & Walters, C.R. (2022), "Systemic Discrimination Among Large U.S. Employers," Quarterly Journal of Economics, 137(4), 1963–2036. [83,000 fictitious résumés; Black names received 2.1 percentage point lower callback rates; firm-level discrimination identified causally] — Bohren, J.A., Hull, P., & Imas, A. (2025), "Systemic Discrimination: Theory and Measurement," Quarterly Journal of Economics, 140(3), 1743–1799. [iterated audit design showing how direct discrimination compounds across hiring rounds]
Descriptive: BLS (2023), "Highlights of Women's Earnings in 2022." — Goldin, C. (2014), "A Grand Gender Convergence: Its Last Chapter," American Economic Review. [note: Goldin documents why convergence slows, not raw gap size — still directionally supports the multiplier].
Causal: Correll, S., Benard, S., & Paik, I. (2007), "Getting a Job: Is There a Motherhood Penalty?" American Journal of Sociology. [audit experiment; mothers penalized relative to non-mothers and men] — Bohren, J.A., Imas, A., & Rosenberg, M. (2019), "The Dynamics of Discrimination: Theory and Evidence," American Economic Review, 109(10), 3395–3436. [field experiment showing how initial gender bias compounds over evaluation history]
Parental education multipliers
Intergenerational transmission of educational attainment and economic mobility
Descriptive: Torche, F. (2011), "Is a College Degree Still the Great Equalizer?" American Journal of Sociology. — NCES (2019), "First-Generation Students: College Access, Persistence, and Postbachelor's Outcomes." — Pew Research Center (2015), "The American Middle Class Is Losing Ground."
Causal: Chetty, R., Friedman, J., Saez, E., Turner, N., & Yagan, D. (2020), "Income Segregation and Intergenerational Mobility Across Colleges in the United States," Quarterly Journal of Economics. [administrative tax data; parental income rank strongly predicts children's college attendance and adult earnings, capturing both education and class transmission]
Immigrant parent multipliers
Immigrant generation effects on education, income, and institutional navigation
Descriptive: Portes, A. & Rumbaut, R. (2001), Legacies: The Story of the Immigrant Second Generation. — NCES (2019), data on first-generation and language-minority students.
Causal context: Chetty, R. et al. (2017), "The Fading American Dream: Trends in Absolute Income Mobility Since 1940," Science, 356(6336), 398–406. [note: this establishes baseline mobility trends for all Americans, not immigrant-specific effects — used here as mobility context] — Abramitzky, R. & Boustan, L. (2022), Streets of Gold: America's Untold Story of Immigrant Success. [historical and quasi-experimental evidence on immigrant mobility patterns across generations]
Neighborhood multipliers
Place-based causal effects on mobility and opportunity — strongest causal domain in the model
Causal: Chetty, R. & Hendren, N. (2018), "The Impacts of Neighborhoods on Intergenerational Mobility I & II," Quarterly Journal of Economics, 133(3). [quasi-experimental; childhood exposure effects on adult income causally estimated] — Chetty, R., Hendren, N., & Katz, L. (2016), "The Effects of Exposure to Better Neighborhoods on Children," American Economic Review. [Moving to Opportunity RCT; randomized housing vouchers show causal income gains for children who moved] — Chyn, E. (2018), "Moved to Opportunity," American Economic Review, 108(10), 3028–3056. [public housing demolition as natural experiment; improved long-run outcomes for displaced children]
Theoretical foundations: Wilson, W.J. (1987), The Truly Disadvantaged. — Sampson, R. (2012), Great American City: Chicago and the Enduring Neighborhood Effect.
Economic stability multipliers
Income volatility as distinct from income level; consumption shocks and child outcomes
Descriptive / mechanistic: Morduch, J. & Schneider, R. (2017), The Financial Diaries: How American Families Cope in a World of Uncertainty. — Duncan, G. & Brooks-Gunn, J. (1997), Consequences of Growing Up Poor.
Causal: Ganong, P. & Noel, P. (2019), "Consumer Spending during Unemployment: Positive and Normative Implications," American Economic Review, 109(7), 2383–2424. [de-identified bank account data; spending drops sharply at predictable income decline — causal evidence that income shocks drive consumption instability independently of baseline income level]
Educational attainment: start-to-outcome odds
Credential attainment by family background; returns to education by credential level
How the odds are computed: The "1 in X" figure is the estimated probability that someone starting where you did ends with exactly the schooling you reported. It comes from an ordered-logit (proportional-odds) model over seven credential levels, from less than high school to multiple graduate degrees. The model is driven by the starting conditions that bear on credentials, with gender (other than non-binary) and parents' origin excluded and the race multiplier damped to its square root. Its two parameters (slope and thresholds) are calibrated so that (1) the overall mix of credentials matches Census 2022 attainment of adults 25+, and (2) the bachelor's-by-24 gap between the top and bottom family-income quartiles matches the Pell Institute estimate (about 58% vs. 16%). The population shares used for calibration are rough approximations, and the model treats inputs as independent. Treat the output as an order-of-magnitude estimate, not a precise probability.
Calibration data: U.S. Census Bureau (2022), Educational Attainment in the United States (adults 25+). — Pell Institute, "Indicators of Higher Education Equity in the United States" (bachelor's attainment by age 24 by family income quartile).
Descriptive: BLS (2023), "Education Pays," Occupational Outlook Handbook. — Hout, M. (2012), "Social and Economic Returns to College Education in the United States," Annual Review of Sociology.
Causal: Autor, D. (2014), "Skills, Education, and the Rise of Earnings Inequality Among the 'Other 99 Percent,'" Science. [documents skill-biased technological change driving credential premiums; partially causal via structural labor market analysis]
Institutional knowledge / social capital multipliers
First-generation navigation, information access, and college opportunity
Theoretical / mechanistic: Putnam, R. (2015), Our Kids: The American Dream in Crisis. — Jack, A. (2019), The Privileged Poor: How Elite Colleges Are Failing Disadvantaged Students. — Lareau, A. (2003), Unequal Childhoods: Class, Race, and Family Life.
Causal: Hoxby, C. & Turner, S. (2015), "What High-Achieving Low-Income Students Know about College," American Economic Review, 105(5), 514–517. [RCT; personalized college information intervention significantly raised low-income students' applications to, enrollment in, and persistence at selective colleges — information access is causally consequential]
Limitations
What this model does not capture
Disability status, sexual orientation and gender identity beyond the binary, undocumented immigration status, regional variation within the US, specific state-level policy environments, the role of individual contingency (a teacher, a program, a moment of luck), and the interaction effects between variables beyond what the multiplicative model approximates. The model is a simplification. It is designed to make structural forces legible, not to provide a complete account of any individual life.