⚠️ Beta Version — This dashboard is under active development. Data, scores, and methodology are drafts. Do not treat any output as final until the dashboard is explicitly labeled as a stable release. Check back for updates.
Terrain

When fascists and billionaires are rigging elections, electoral organizing gets hard. But there is one force that can never be stopped.

The ability for workers to cause a crisis — to deny their labor — is the most durable form of political power ever created. No gerrymandering touches it. No dark money buys it. The real question is: what is the fastest and most strategic way to build that power? This project offers some ideas. But it is designed to inform — not to direct real on-the-ground organizing.

Urgency requires strategy and prioritization.

We don't have the time or the luxury to not build the strongest labor movement capable of transforming this country. That requires an important distinction. Every worker deserves a union — the hospital worker, the farm worker, the software engineer. But not all workers are equally strategic for large-scale political change. That does not mean they don't deserve democracy in their workplace. It means that if we want to win fast, we need to prioritize some sectors in some locations.

We score every U.S. county on five variables. We score every sector on its strategic value.

The result is a map of where power can be built — and what kind.

VariableWhat it measures
Strike CapacityCan workers cause a crisis?
Political AlignmentIs terrain winnable electorally?
Community EmbeddednessAre workers embedded in community relationships?
Existing InfrastructureAre unions already organized here?
Electoral AlignmentIs the county competitive?
SectorSVS Score
Hospitals100
K-12 Education100
Ports95
Utilities85
Retail10
Tech0
→ Full methodology

Different workers are powerful in different ways. We've divided leverage into two buckets.

A port strike chokes capital flow — economic pressure forces a response. A teachers strike chokes the community — parents, students, churches mobilize as active allies. Jane McAlevey called this whole-worker organizing: workers aren't just employees, they're embedded in relationships that can be mobilized into power. Same county, completely different organizing strategy.

Logan County
WV · Type B
19.21
OPP SCORE · Sectoral: 2.2 · Scale: 10

Once the largest coal-producing county in WV. Nearly 28% poverty rate today. Capital extracted the wealth and left. High chokepoint potential, low community infrastructure.

Coal Reignites A Mighty Battle of Labor History ↗
Kanawha County
WV · Type B
40.6
OPP SCORE · Union Culture: 80 · Scale: 60

2018 wildcat strike shut down all 55 counties. Rank-and-file refused back-to-work orders from their own union leaders. Launched the national Red for Ed movement.

Biggest Wildcat Strike in Decades Hints at New U.S. Labor Unrest ↗

Your goal shapes what you see. Your terrain type tells you what to do with it.

A Senate race in 2026 needs you to activate existing organized power fast. A decade of power building in Texas means finding places where new organizing can take root. Switch the goal — the map responds.

Three intervention types. No rankings — just different strategic situations:

Type A — Low existing infrastructure, high potential. Build from scratch. Find the organic leaders.

Type B — Organizations exist but aren't moving politically. The power is there. It needs activation.

Type C — Strong, aligned organizations ready for coordinated action. Best terrain for electoral leverage.

Examples: Clark County NV is Type C — Culinary Union knocked on 900,000 doors in 2024. Cook County IL is Type B — CTU's 2012 strike launched a national movement but the organized power still isn't fully activated politically. Kanawha County WV is Type B — 20,000 workers shut down 55 counties without leadership support.

Why Terrain is different from traditional labor strategy.

Since the 1970s, much of the labor movement has used union density — how many workers have unions in a given sector or region — as the primary goal and metric for success. We push back against that.

The real power of workers is not in their technical membership. It's whether the actual workers on the ground are actively organized and ready to ultimately force the hand of those in power. A union more focused on increasing membership numbers than on strikes, contract negotiations, and competitive elections is doing little to change the country.

We focus on political impact. And this is a good moment to be clear: just because a sector isn't strategic for large-scale change doesn't mean those workers don't matter. Hospitality workers often have some of the worst working conditions in the country and have built some of the strongest organizations — Clark County NV is proof. But a hotel strike is unlikely to create the kind of community or capital crisis that moves elections. Terrain helps you see the difference.

3,144 counties. Every one scored. Start with a goal.

Explore the map or dig into case studies below.

Kanawha County
WV · Type B: Political Activation
40.6
OPP SCORE
Union Culture: 80 Scale: 60

2018 wildcat strike shut down all 55 counties. Rank-and-file refused back-to-work orders from their own union leaders. Launched the national Red for Ed movement.

Logan County
WV · Type B: Political Activation
19.21
OPP SCORE
Sectoral: 2.2 Scale: 10

Once the largest coal-producing county in WV. Nearly 28% poverty rate today. Capital extracted the wealth and left. High chokepoint potential, low community infrastructure.

Clark County
NV · Type C: Partnership
86.5
OPP SCORE
Union Culture: 80 Scale: 100

Culinary Union Local 226. 60,000 hospitality workers. 900,000+ doors knocked in 2024. Pivotal in flipping Nevada Democratic every cycle since 2008.

Cook County
IL · Type B: Political Activation
86.5
OPP SCORE
Scale: 100

CTU's 2012 strike launched a national movement and directly inspired the WV wildcat. Organizers were studying McAlevey's book about CTU when they started planning. Bargaining for the common good.

Allegheny County
PA · Type C: Partnership
86.5
OPP SCORE
Union Culture: 100

Deep union history. Strong infrastructure. But labor leaders told state lawmakers in 2023 they're worried about declining membership despite widespread public support. Classic Type B: power exists, not yet activated.

Maricopa County
AZ · Type C: Partnership
75.25
OPP SCORE
Union Culture: 20 Scale: 100

Arizona teachers struck in 2018 inspired by WV. Today it's swing Senate terrain. High organized scale, low union culture score — built for deployment, not organic solidarity. Goal matters here.

Philadelphia County
PA · Type C: Partnership
86.5
OPP SCORE
Union Culture: 100

High OOS, concentrated healthcare and education sectors, dense community embeddedness. The potential is here. The organizing infrastructure is not yet at scale.

Multnomah County
OR · Type B: Political Activation
70.31
OPP SCORE
Scale: 80

Portland. Strong progressive infrastructure, community-facing sectors well represented. Type B — organized power looking for the right campaign.

Viewing: Default lens
Senate competitiveness tiers based on NPR analysis by Domenico Montanaro (May 2, 2026). Source: NPR — 2026 Senate Races to Watch. Reflects structural competitiveness, not current cycle polling.

Metro Area

States
Organizing
Opportunity
Organizing
Sectoral
Presidential
Statewide
Congressional
Infrastructure

Metro Area

Counties
Population
Union Culture
Organized Scale
Swing State(s)

Counties in Metro

Default
Community
Capital
Intervention Type
Show Intervention Types
# County ST OPP Type Pres State Cong Org Sect Infra 2024 Swing

The Model, Explained

Every score in this model is grounded in social science theory, built from public data, and documented here. Click any component to expand the full explanation — from peer-reviewed framework to organizer's field logic.

YOUR GOAL— Electoral objective set by the user
ORGANIZING OPPORTUNITY SCORE= Sectoral Value (55%) + Organizing Potential (45%)
+ INFRASTRUCTURE SCORE→ drives Intervention Type (A / B / C)
ELECTORAL SUB-SCORESPresidential · Statewide · Congressional — goal-dependent
COMPOSITE RANKING= Opportunity × Electoral weight
Scoring Components
🎯
Organizing Opportunity Score
The composite terrain score — how favorable is this county for building power?
Primary output · 0–100
Social Science

A composite index measuring the structural preconditions for collective action formation, operationalizing McAlevey's (2016) whole-worker organizing framework as a weighted combination of sector-level marketplace bargaining power (Silver, 2003) at 55% and unorganized labor reserve at 45%. Scores are normalized against an absolute national benchmark — not relative rankings — preserving cross-county comparability independent of filter state.

OOS = (Sectoral Value Score × 0.55) + (Organizing Potential Score × 0.45)

Electoral scores and Infrastructure Score are intentionally excluded from the OOS formula. Combining organizing potential with electoral value would conflate two analytically distinct questions, corrupting both measurements.

Plain Language

A 0–100 score answering one question: how good is this county's ground for building worker power? It combines the leverage of the workers here (what sector they're in) with how many of them are still unorganized. Both have to be true at once to score high.

Organizer's Take

This is your terrain map before you commit resources. High score means you've got leverage and people to organize — the combination that turns a drive into a power-building project. A high score in a county with no organizers is a call to act. A low score in a county you're already in is a question you need to answer.

🏭
Sectoral Value Score (County)
Does this county have strategically powerful workers — and enough of them?
55% of OOS
Social Science

County-level weighted summation of sector-specific Strategic Value Scores (SVS), normalized to an absolute benchmark. Implements Silver's (2003) marketplace bargaining power thesis: workers' structural position in production — specifically their capacity for disruptive action — is the primary determinant of leverage, independent of organizational density or political will. A county's sectoral composition is thus the foundational structural variable.

County SVS = Σ (sector_SVS × county_employment_in_sector) → scaled 0–100

Known gap: County Business Patterns excludes government workers by design. Government-heavy counties (DC, state capitals) are systematically understated.

Plain Language

Two things have to be true at once — it's a multiplication, not an addition. A county scores high only if it has both strategically important workers AND enough of them. A county full of retail workers scores low even in a swing state. A county with a single large hospital system in a rural region scores high.

Organizer's Take

One hospital system beats five strip malls. It's not headcount — it's who can actually stop something when they walk off. Before you open a campaign, look at what sector you're entering. This score tells you if the terrain has leverage baked into it, before you've talked to a single worker.

Strategic Value Score (Sector)
Boolean rubric: how much structural leverage does this sector have?
Feeds County SVS
Social Science

Sector-level boolean rubric operationalizing chokepoint theory (Clawson & Clawson, 1999; Silver, 2003) and McAlevey's (2016) community embeddedness thesis. Points are assigned for: Chokepoint Potential (25 pts) — capacity to disrupt essential flows; Crisis-Creating (20 pts) — a stoppage creates visible public harm; McAlevey Priority (20 pts) — explicitly identified in the whole-worker framework; Community-Facing (15 pts) — workers are embedded in the social fabric beyond the workplace; Non-Offshoreable (10 pts) — cannot be geographically displaced. A ×1.15 multiplier applies when Chokepoint=YES AND Community-Facing=YES, reflecting empirical evidence that these two dimensions produce non-additive leverage when combined.

SectorChokeCrisisMcAl.Comm.Non-offSVS
HospitalsYESYESYESYESYES100
K-12 EducationYESYESYESYESYES100
Public Admin – LocalYESYESYESYESYES100
Ports / MaritimeYESYESYESNOYES75
TruckingYESYESYESNOYES75
WarehousingYESYESYESNOYES75
UtilitiesYESYESTBDNOYES~55 (pending)
Public Admin – StateYESYESTBDBRDLNYESflagged
Higher EducationNONONOYESYES25
Mfg – DurableYESNONONOYES25
Social AssistanceNONONOYESYES25
Public Admin – FederalNONONONOYES10
Mfg – Non-durableNONONONOYES10
ConstructionNONONONOYES10
RetailNONONONOYES10
HospitalityNONONONOYES10
Healthcare – AmbulatoryNONONONOYES10
Other ServicesNONONONOYES10
Financial ServicesNONONONONO0
Tech / InformationNONONONONO0
⚠️ Utilities and Public Admin – State are flagged for review. EMS is currently folded into Healthcare – Ambulatory (SVS=10) but should score 100. Both are Phase 2 corrections.
Plain Language

A checklist: Can workers in this sector stop something critical? Is their work woven into community life? Can it be shipped overseas? The answers determine the sector's score. Hospitals and schools answer YES across the board. Financial services and tech answer NO — they can move, they can automate, and nobody's neighborhood depends on them showing up.

Organizer's Take

Nurses' strike shuts the city down and your neighbor's kid is in that school. That's the combination. Retail scores 10 because Amazon opens a new store. Know which sector you're entering before you spend a year of someone's life on a campaign — the terrain either works for you or against you from day one.

🌱
Organizing Potential Score
How large is the unorganized workforce here?
45% of OOS
Social Science

Measures the size of the available unorganized labor pool — total county employment minus estimated union members, scaled by the county's share of national employment. Operationalizes McCarthy & Zald's (1977) classical resource mobilization concept of "available population": the raw material for movement formation, analytically independent of structural leverage quality. Combined with SVS, it produces the OOS.

Unorganized workforce = Total employment − Estimated union members → scaled 0–100

Known ceiling: ~2,192 LM-2 union locals (~20% of filers) are unresolvable due to PO boxes, international addresses, or territories. Union member counts in every county are floors, not ceilings.

Plain Language

How many workers in this county don't have a union yet? This is about quantity of opportunity — the raw material. Sector quality tells you the terrain; this tells you how many people you could bring into the fight. Both have to be strong for a county to score high overall.

Organizer's Take

A county with 50,000 unorganized hospital workers is a jackpot. A county with 500 unorganized retail workers is a long slog. This score tells you how full the room is before you walk in — before you've knocked a single door or made a single ask.

🗳️
Presidential Sub-score
Electoral College leverage — how much does this county matter for the presidency?
Goal: Presidential Battlegrounds
Social Science

Electoral college leverage score combining state-level swing weight (derived from historical presidential margin tightness and Electoral College value) with county-level 2024 presidential margin. Implements spatial targeting logic from electoral studies literature: organizing resources have highest marginal electoral value in counties within competitive states that are themselves competitive (Hersh, 2015; Enos & Fowler, 2018). State weights: PA/WI=100, AZ/NV=90, GA/MI=85, NC/TX=70, FL=60, others scaled. 28 counties — mostly Alaska boroughs — have no election data.

Plain Language

How much does organizing in this county matter for winning the presidency? Counties in close states that are themselves close score highest — those are the places where mobilizing a few thousand more union households could actually flip the outcome.

Organizer's Take

If you're in Pennsylvania and your county went R by 3 points, you're gold — that's the margin labor can move. Don't burn your members' shoe leather in a county going D by 30. Take that capacity somewhere it counts. Presidential math is about marginal returns; this score does that math for you.

🏛️
Statewide Sub-score
Senate and governor — how competitive is this state for statewide power?
Goal: Long-Term State Transformation
Social Science

State-level competitive index for statewide office (Senate, governor) derived from historical margin tightness and Cook Political Report ratings. Based on the theoretical premise that labor mobilization has asymmetric electoral returns — high in competitive states, near-zero in noncompetitive states (Key, 1949; Ansolabehere et al., 2006). Critical validity limitation: gubernatorial, Senate, and presidential races each require distinct intervention logic. This sub-score does not distinguish between them — race-specific targeting is a Phase 2 refinement.

Plain Language

How competitive is this state for Senate and governor races? Counties in states with close statewide races score higher because organized labor's electoral work there has real marginal value. But be careful — a competitive state for a Senate race may be irrelevant for the governor's race, and vice versa.

Organizer's Take

The statehouse controls labor law. A competitive Senate seat is worth years of infrastructure because winning it can change the rules for everyone. But know which race you're targeting — the Senate district map and the governor's map are different games, and your precinct strategy has to match the race.

🏠
Congressional Sub-score
House races — general election competition and primary challenge potential.
Goal: Flip the House
Social Science

District-level competitiveness score incorporating Cook Political Report ratings and MIT Election Lab historical margins. Includes positive weighting for safe Democratic districts with vulnerable incumbents, reflecting strategic literature on primary elections as an accountability mechanism within legislative bodies (Boatright, 2013). This is a normative model choice: primary challenges are treated as a legitimate tool for legislative accountability.

Plain Language

How much does organizing in this county matter for House races? This covers both tight general-election districts AND safe Democratic seats where a primary challenge could replace a bad incumbent with a better one.

Organizer's Take

The House is the first lever. A competitive district is obvious. But don't sleep on a Democrat who votes against card check in a safe seat — you can primary them. Safe seat, wrong incumbent? That's a target too. This score accounts for both plays.

🏗️
Infrastructure Score
Scale and culture — how much organized power already exists here?
Reference score
Social Science

Two-dimensional organizational capacity measure: Organized Scale — total union members (raw count, scaled), capturing deployable political resources; Union Culture — union density (members/total employment), capturing the normalized social acceptance of collective action, which Bronfenbrenner (2009) identifies as a significant predictor of new organizing success. Deliberately excluded from the OOS formula: density is an outcome of past strategy, not a predictor of future organizability. Including it would introduce circular measurement.

Plain Language

Two questions: how many union members are here (scale), and what fraction of the workforce are they (culture/density)? Both matter but for different reasons. Scale drives what you can deploy politically right now. Culture shapes whether new organizing will stick.

Organizer's Take

High scale means you've got members ready to knock doors this cycle. High density means workers in this county grew up with uncles in the union — the culture is already there, and a new drive has air under it from day one. Low on both? You're starting from scratch. That's not a reason not to go — but price it right.

🔀
Intervention Type (A / B / C)
What strategic mode does this terrain call for?
Categorical classifier
Social Science

Categorical classifier derived from the infrastructure–opportunity matrix, drawing on social movement theory's distinction between mobilization (activating existing resources) and organization-building (creating new resources) as distinct strategic modes with different resource requirements and timelines (McCarthy & Zald, 1977; McAlevey, 2016). Prevents the common error of applying uniform resource allocation to structurally dissimilar terrains.

Opportunity
Low Infrastructure
High Infrastructure
High
Type A ★★★
Build New Power — flagship campaign terrain
Type C ★★★
Move Together — ready to deploy now
Medium
Type A ★★
New organizing, longer timeline
Type B ★★
Wake the Giant — activate existing members
Low
— Deprioritize
Type B ★
Electoral activation only
Plain Language

Type A — Build from scratch where there's opportunity but no unions. Type B — Wake up the existing unions that aren't politically active. Type C — Coordinate unions that are already aligned and active. Different types, different resources, different timelines.

Organizer's Take

These aren't just labels — they determine your budget, your timeline, and your staff model. A Type A campaign needs an experienced lead organizer and 18 months minimum. A Type C just needs a coordinator and a voter file. Don't run a Type A campaign with Type C resources. Don't waste a Type C terrain by treating it like it needs to be built from scratch.

📖
McAlevey Framework
The whole-worker theory driving sector selection in this model.
Theoretical foundation
Social Science

McAlevey's (2016, 2020) whole-worker framework argues that durable political power requires organizing workers as complete social beings embedded in their communities — not merely as economic actors. The framework predicts that sectors with both chokepoint leverage and community embeddedness generate superior long-run political capacity compared to sectors with leverage alone. This is in direct theoretical tension with the Strategic Sector Initiative (SSI), which prioritizes manufacturing based on industrial organization theory. This model sides with McAlevey — the SVS rubric encodes her sector hierarchy, and the ×1.15 multiplier operationalizes her core claim about the non-additive effect of combining leverage with community presence.

Key citations: McAlevey (2016) No Shortcuts; McAlevey (2020) A Collective Bargain; Silver (2003) Forces of Labor (chokepoint theory alignment).

Plain Language

The model follows McAlevey's argument: organize nurses and teachers, not Amazon drivers. Hospitals have more leverage AND nurses live in the communities they serve. Both things are true, and together they're more powerful than either one alone. The SSI framework would say go to manufacturing — this model disagrees.

Organizer's Take

McAlevey's insight is simple: the best organizers don't see workers — they see whole people who go home to neighborhoods, churches, and school board meetings. If your sector is woven into community life AND can shut something down, you've found the sweet spot. That's not sentiment — it's the empirical record of which campaigns build the kind of power that lasts past a contract.

⚖️
Two-Actor Negotiating Logic
Bosses and politicians respond to different kinds of pressure.
Theoretical context
Social Science

Theoretical distinction between two principal counterparties in labor strategy: private-sector employers — profit-maximizing actors amenable to cost-benefit negotiation, where workplace disruption creates measurable material losses; and political actors — responsive to electoral, ideological, and reputational incentives, capable of ideological intransigence even at substantial material cost (principal-agent theory; public choice literature). Empirically validated by cases where conventional labor leverage failed against ideologically committed political actors: Walker (WI, 2011), DeSantis (FL, 2022). The electoral sub-scores exist precisely because political actors require a different pressure mechanism — the ballot box, not the picket line alone.

Plain Language

You negotiate with bosses and politicians very differently. A boss will calculate what the strike costs them and make a deal. A politician running on an anti-union platform will let the strike drag on and call it a win. Different actor, different logic, different tools needed.

Organizer's Take

Scott Walker took the hit. DeSantis took the hit. They wanted the fight — it was their brand. Conventional pressure doesn't work on a true believer with a base. You can't bargain someone out of an ideology. You vote them out. That's why this model has electoral scores as a separate layer — because some opponents can only be beaten at the ballot box, not at the table.

⚠️
What This Model Cannot Do
The model's boundaries — what organizer judgment must supply.
Epistemological limits
Social Science

Systematic documentation of measurement validity threats (Cook & Campbell, 1979). The model cannot operationalize: associational power (Levi, 1996), organic leader network centrality, employer-specific financial vulnerability (Bronfenbrenner, 2009), local political contingency, or internal union democracy. These represent genuine construct validity gaps that structural administrative data cannot close — they require ethnographic, participatory, and field research methods. The model is a structural prioritization instrument, not a substitute for ground-level political intelligence.

Plain Language

The model sees structure, not people. It cannot see whether your county has a trusted leader who can move a shop, whether the employer is in financial trouble, whether there's a reform caucus ready to activate, or what happened at the last organizing drive. Those gaps are real and they matter.

Organizer's Take

This map shows you where to look — not what you'll find when you get there. The next step after a high-scoring county is always talking to people on the ground. The model tells you to go to Scranton; your job is to find out who in Scranton is ready to move, and whether the boss is overextended, and whether the last drive failed because of a snitch or because of a bad lead organizer. The model can't answer any of that. You have to.

Data Sources
📊 BLS QCEW
Bureau of Labor Statistics Quarterly Census of Employment and Wages — county-level employment by sector.
↗ bls.gov/cew
🏛️ DOL LM-2
Dept. of Labor OLMS union financial disclosures — local membership counts mapped to county FIPS.
↗ dol.gov/olms
🗳️ MIT Election Lab
2020 + 2024 presidential margins by county; statewide and congressional historical results.
↗ electionlab.mit.edu
📚 Cook Political Report
Congressional and statewide race competitiveness ratings.
↗ cookpolitical.com
🗺️ Census Bureau
ZCTA crosswalk, gazetteer files, County Business Patterns — FIPS codes, population, geographic linkages.
↗ census.gov
🎓 Cornell ILR Tracker
Strike data 2019–2024. Acquired for Phase 2; not yet active in the model.
↗ striketracker.ilr.cornell.edu
Known Limitations
1
Associational power not measured
DSA chapters, labor councils, interfaith coalitions — absent. On-the-ground intel fills this gap.
2
Public sector employment understated
CBP excludes government workers. DC, state capitals, county seats are systematically underscored.
3
28 counties missing election data
Primarily Alaska boroughs. Documented gaps, not errors.
4
~2,192 unresolvable union locals (~20% of LM-2)
PO boxes, territories, international addresses. Union counts in all counties are floors, not ceilings.
5
Strike history not yet scored
Cornell ILR data acquired but not incorporated.Phase 2
6
Statewide sub-score not race-specific
Governor vs. Senate vs. President require different logic. Cycle-specific targeting is Phase 2.Phase 2
7
EMS misclassified
EMS folded into Ambulatory (SVS=10). Should score SVS=100. Phase 2 correction.Phase 2
8
Organizability rubric is draft
The Harris critique — some organic leaders may be pro-management — is a live validity threat not resolved in v1.0.
AI Disclosure

Built in collaboration with Claude (Anthropic) for pipeline coding, data validation, literature review, and adversarial testing. All methodological decisions were made by the human researcher. Full documentation at github.com/kpsmas123-bit/labor-organizing-model.

About This Project

Built by Sam Kaplan-Pettus, recent UC Berkeley Political Science graduate. A personal project exploring how new AI tools can serve the labor movement — inspired by Jane McAlevey's work on labor and political prioritization. Now looking for work and projects at the intersection of labor and national political organizing.

📧 sam.kp@berkeley.edu · 📂 github.com/kpsmas123-bit/labor-organizing-model

Ready