Data Scientist and Visualization Engineer with a background in spatial analytics and mapping.
Public
Edited
Feb 23, 2024
3 forks
22 stars
Topographic MappingBubble mapChoroplethAccess to Family planningMD Counties Total Cases MapTendance de la production des déchets en Union EuropéenneAndy's Walgreens COVID-19 Tracker TrackerA better U.S. house election results map?1983 Mayoral Election, Dot density mapsMastodon 🐘Cheat sheet bertinBertin.js: regular squaresWaterlinesNeumorphism Contour Density MapCartographic DoodlesStars and constellationsPlot: Grid choroplethHello Polygon MorphingMapping with pie chartsU.S. Geographic DataHow big are countries... like really!AttitudeB&W ChoroplethWeb Mercator Tile VisibilityMARTINI: Real-Time RTIN Terrain Mesh"Magnifying-Glass" projectionsTissot's indicatrixAntipodal mapMapping gridded data with a Voronoi diagramA Map of Every BuildingUrbano Monti’s Planisphere (1587)Bivariate choroplethDIY HillshadeMapbox Map MakerWorld tourHillshaderSimplified Earth with curved shapesHexbin mapInner glowNicolosi vs. StereographicData-driven projections: Darwin's worldSatellite ground track visualizerDirection to shoreHello, OpenLayers!U.S. airports VoronoiHello, NYC Geosearch API!Mapbox Fly-ToSpilhaus shoreline mapWalmart’s growthHow well does population density predict U.S. voting outcomes?Drawing maps from geodata with D3 & ObservableHexgrid maps with d3-hexgridTissot's indicatrixWorld airports VoronoiSwiss Elevation Line GraphsVector tilesVersor draggingOrthographicSolar TerminatorStreaming ShapefilesFake GlobesPeirce Quincuncial🍃 LeafletU.S.G.S. World Earthquake MapUsing Mapbox GL JSUsing Google Maps
Election Maps for Incomplete Results
Also listed in…
Partial Election Results
data
curData = data.filter(d => d.State == state & d.Scenario == scenario & d.StatePctRecorded == statePctRecorded)
curStateData = ({
'DemVotes': d3.sum(curData, d=> +d.DemVotes),
'RepVotes': d3.sum(curData, d=> +d.RepVotes),
'TotalVoters': d3.sum(curData, d=> (+d.DemVotes) + (+d.RepVotes) + (+d.RemainingVotes))
})
curStateShape = stateShapes.features.filter(d => d.id == states.get(state).FIPS)
curCountyShapes = countyShapes.features.filter(d => d.id.slice(0,2) == states.get(state).FIPS)
curCountyShapes
shapesWithDataCounty = curCountyShapes // Counties for current state selected
.map(s => {
let curCountyResults = curData.filter(d=> d.FIPS == s.id)[0]
let curColor = curCountyResults && +curCountyResults.DemVotes == +curCountyResults.RepVotes ? '#ccc' :
curCountyResults && +curCountyResults.DemVotes > +curCountyResults.RepVotes ? '#2e74c0' :
'#cb454a';
let colRamp = d3.scaleLinear().domain([0, 1]).range(["#2e74c0", "#cb454a"])
let curColorRamped = colRamp(+curCountyResults.RepVotes / (+curCountyResults.DemVotes + +curCountyResults.RepVotes))
return ({
...s,
...curCountyResults,
color: curColor,
colorRamped: curColorRamped,
centroid: turf.centroid(s),
TotalVoters: (+curCountyResults.DemVotes) + (+curCountyResults.RepVotes) + (+curCountyResults.RemainingVotes)
})
})
colRamp = d3.scaleLinear().domain([0, 1]).range(["#2e74c0", "#cb454a"])
colRamp(0.75)
curStateCities = stateCitiesGeo.get(state)
packedDotRadius = ( 0.037 * width * mapScale )* (1.5 / 30 ) // radius for inner dot. Should probably change it to function of max donut size
maxPackedDots = 208 // number of dots to put in the biggest County.
packedDotPop = // population associated with 1 dot it biggest County.
d3.max(shapesWithDataCounty, d=> d.TotalVoters) / maxPackedDots
pack = (data,outerRadius) => d3.pack()
.size([outerRadius*2, outerRadius*2])
.padding(1)
.radius(d=> d3.min([packedDotRadius, outerRadius]))
(d3.hierarchy(data));
circlePackData =
circles.map( d=> {
let packedDotCount = d3.max([1, Math.round(d.TotalVoters / packedDotPop)]) // d3.max to ensure dots > 0
let countyPack = pack( {children:d3.range(packedDotCount)}, d.r)
return {
r: countyPack.children.length > 1 ? countyPack.r : d.r,
x: d.x,
y: d.y,
children: countyPack.children.map(c => ({x: c.x, y: c.y, r: c.r, col: circlePackDotColor(d.DemVotes, d.RepVotes, d.TotalVoters)}))
} ;
})
circlePackDotColor = (demVoters, repVoters, totalVoters, colorScale = ({dem: "#2e74c0", rep: "#cb454a", remaining: '#BBB' })) => {
let rand = Math.random()
if (rand < +demVoters / +totalVoters ) {
return colorScale.dem
} else if (rand < (+demVoters + +repVoters)/ +totalVoters) {
return colorScale.rep
} else {
return colorScale.remaining
}
}
curProps = countyPartyProps.get(state).get(scenario).get(statePctRecorded)
dotPop = // population associated with 1 dot.
d3.sum(shapesWithDataCounty, d=> d.TotalVoters) / curDots.length
dotColor = function(propArray, randNum, colorScale = ({dem: "#2e74c0", rep: "#cb454a", remaining: '#fbf418' })) {
if(+randNum < +propArray[0]){
return colorScale.dem
} else if(+randNum < +propArray[1]) {
return colorScale.rep
} else {
return colorScale.remaining
}
}
curDots = stateDots.get(state)
.map( d => {
return {coordsProjected: projection(d.coords),
color: dotColor(curProps.get(d.FIPS), d.rand)
}
})
proportionToHex = x => (Math.round(x * 255)).toString(16);
proportionToHex(0.75)
curStateCentroidCoords = turf.centroid(curStateShape[0]).geometry.coordinates
projection =
d3.geoAlbers()
.rotate([(state == 'Arizona' ? 111.5 : -curStateCentroidCoords[0]), 0])
.center([0, curStateCentroidCoords[1]])
.fitSize([width*mapScale, height*mapScale], curStateShape[0])
mapScale = (sidebarPlotPixels > 0.12 * width ? 1 : 0.75)
height = width * 7/9
bubbleRadiusScale = d3.scaleSqrt()
.domain([0, d3.max(shapesWithDataCounty.map(d=> d.TotalVoters))])
.range([0, 0.035 * width * mapScale])
donutRadiusScale = d3.scaleSqrt()
.domain([0, d3.max(shapesWithDataCounty.map(d=> d.TotalVoters))])
.range([0, 0.037 * width * mapScale])
addStateOutlineToMap = (g) => {
g.append( "path" )
.attr( "d", d3.geoPath().projection(projection)(curStateShape[0] ))
.attr( "stroke", "#888")
.attr( "fill", "none")
.attr( "stroke-width", 1.8)
.attr( "opacity", 0.8 )
}
curStateShape[0]
d3.geoPath().projection(projection)(curStateShape[0] )
curStateShape[0]
addCountiesToMap = (g, mapSpecifier, opacity = 0.8) =>
g.selectAll("path")
.data(shapesWithDataCounty)
.join( "path" )
.attr( "fill",d => (mapSpecifier == "choropleth" ? d.color : "#FFFFFF00"))
//.attr( "fill",d => (mapSpecifier == "choropleth" ? d.color : "white"))
.attr( "opacity", opacity )
.attr( "stroke", "#999")
.attr( "d", d3.geoPath().projection(projection) )
.on('click', function(event, d) {console.log(d)} )
.on('mouseenter', function(event,d) {
d3.select('#county-results-' + mapSpecifier).style('visibility','visible')
.html(makeResultsTable(d.properties.name + ' County', d, mapSpecifier))
})
.on('mouseleave', function(event,d){
d3.select('#county-results-' + mapSpecifier).style('visibility','hidden')
})
addCityLabels = g => {
g.style('pointer-events', 'none')
g.selectAll("circle")
.data(curStateCities.features)
.join("circle")
.attr("fill", "#333")
.attr("r", 2)
.attr("transform", d => {
let p = projection(d.geometry.coordinates)
return `translate(${p})`
})
g.selectAll("text")
.data(curStateCities.features)
.join("text")
.attr("fill", "#222222")
.text(d=> d.id)
.style('font-weight', 'bold')
.style('font-size', '19px')
.attr("x", 4)
.attr("y", 10)
.attr("transform", d => {
let p = projection(d.geometry.coordinates)
return `translate(${p})`
})
}
addBubblesToMap = (g, remainingOpacity = false, colorRamp = false) => {
let opacityMultiplier = (d) => remainingOpacity == false ? 1 : (1.5 * (+d.DemVotes + +d.RepVotes)/+d.TotalVoters)
g.style('pointer-events', 'none')
g.selectAll( "circle" )
.data(shapesWithDataCounty)
.join( "circle" )
.attr( "fill", d=> colorRamp ? d.colorRamped : d.color )
.attr( "fill-opacity", d=> 0.4 * opacityMultiplier(d) )
.attr( "stroke", d=> colorRamp ? d.colorRamped : d.color)
.attr( "stroke-opacity", d=> 0.4 + ( 0.4* opacityMultiplier(d)))
.attr('r',d=> bubbleRadiusScale(d.TotalVoters))
.attr("transform", d => {
let p = projection(d.centroid.geometry.coordinates)
return `translate(${p})`
})
}
addDonutsToMap = g => {
g.style('pointer-events', 'none')
g.selectAll("g")
.data( pieData )
.join('g')
.attr("class", 'circles')
.attr("id", d => ('circles' + d.id))
.attr('opacity', 0.75)
.selectAll('path')
.data((d,i)=> pie( pieData[i].data ))
.join("path")
.attr("fill", d => colors[d.data.name])
.attr("d", d3.arc()
.innerRadius( d=> 0.4 * d.data.r)
.outerRadius(d=> d.data.r))
.attr("transform", d=> "translate(" + d.data.x + "," + d.data.y + ")")
}
addSidebar = (container, mapSpecifier) => {
// add sidebar with counts for current state and placeholder for counties (to be triggered on mouseenter of counties).
const sidebar = container.append('div').style('width','23%').style('vertical-align','top').style('display','inline-block')
sidebar.append('div').html(makeResultsTable(state, curStateData, mapSpecifier))
sidebar.append('div').style('padding-top','40px').attr('id', 'county-results-' + mapSpecifier).style('visibility','hidden')
}
sidebarPlotPixels = 100
sideBarColOpacityHex = 'B3'
makeResultsTable = function(place, dd, mapSpecifier) {
let [demVotes, repVotes, totalVoters] = [+dd.DemVotes, +dd.RepVotes, +dd.TotalVoters]
let demoninator = demVotes + repVotes;
if(mapSpecifier != 'choropleth' & mapSpecifier != 'bubbles') {demoninator = totalVoters}
let outputHtml = `
<div id='sidebar'>
<h3>${place} Results</h3>
<table>
<thead>
<tr>
<th width="1%"></th>
<th width="30%" style="text-align: left;">Candidate</th>
<th style="vertical-align: middle;text-align: right;">Votes</th>
<th width= "40%"></th>
</tr>
</thead>
<tbody>
<tr style="max-width:20%">
<td><svg width="20" height="20" viewBox="0 0 20 20"><circle cx="50%" cy="50%" r="4" fill="${colors.rep + sideBarColOpacityHex}"></circle></svg></td>
<td>Republican</td>
<td style="text-align: right;">${d3.format(",")(repVotes)}</td>
<td><div class="bar-background" >
<span class="bar-span rep" style="width: ${100 * repVotes / demoninator}%"></span>
</div></td>
</tr>
<tr>
<td><svg width="20" height="20" viewBox="0 0 20 20"><circle cx="50%" cy="50%" r="4" fill="${colors.dem + sideBarColOpacityHex}"></circle></svg></td>
<td>Democrat</td>
<td style=";text-align: right;">${d3.format(",")(demVotes)}</td>
<td><div class="bar-background" >
<span class="bar-span dem" style="width: ${100 * demVotes / demoninator}%"></span>
</div></td>
</tr>`
if (mapSpecifier != 'choropleth' & mapSpecifier != 'bubbles' ) {
outputHtml += `<tr>
<td><svg width="20" height="20" viewBox="0 0 20 20"><circle cx="50%" cy="50%" r="4" fill=${mapSpecifier == "dots" ? "#fbf418": "#444"}></circle></svg></td>
<td>Est. Left to Count</td>
<td style="text-align: right;">~${d3.format(",.2r")(totalVoters - (demVotes + repVotes))}</td>
<td><div class="bar-background" >
<span class="bar-span" style="background:${mapSpecifier == "dots" ? "#fbf418": "#444"};width: ${100 * (totalVoters - (demVotes + repVotes)) / demoninator}%"></span>
</div></td>
</tr>
</tbody>
</table>`
} else {
outputHtml +=
` </tbody>
</table>
<div style="font-size: 12px;">Est. ${d3.format('.0%')((demVotes+ repVotes)/totalVoters)} votes in</div>`
}
outputHtml += `</div>
<style>
#sidebar td {
vertical-align: middle;
}
#sidebar .bar-background {
display: flex;
position: relative;
width: 100%;
height: 0.6rem;
}
#sidebar .bar-background::after {
content: "";
position: absolute;
width: 100%;
height: 100%;
top: 0px;
left: 0px;
background-color: #ccc;
}
#sidebar .bar-span {
z-index:1;
height: 0.6rem;
position:relative;
}
#sidebar .dem {
background:${colors.dem + sideBarColOpacityHex};
}
#sidebar .rep {
background:${colors.rep + sideBarColOpacityHex};
}
</style>`
return outputHtml
}
annote1Url = FileAttachment("BubbleAnnot1@1.png").url()
annote2Url = FileAttachment("BubbleAnnot2.png").url()
annote3Url = FileAttachment("BubbleAnnot3.png").url()
annote4Url = FileAttachment("BubbleAnnot4@2.png").url()
annoteDonut = FileAttachment("annoteDonut2.png").url()
annotePackedDot = FileAttachment("annotePackedDot.png").url()
d3 = require('d3@6')
topojson = require("topojson-client@3")
turf = require('https://bundle.run/turf@3.0.14')
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.