Final edition

_assets/scripts/avg_of_avg.jl

@@ -1,20 +1,20 @@
 using Plots
-pa = [ 73.0, 53.0, 48.0, 70.0, 71.75 ]
-ga = [ 36.0, 42.0, 18.0, 3.0, 2.97 ]
-nc = [ 30.0, 24.0, 29.0, 16.0, 15.39 ]
-mi = [ 73.0, 82.0, 81.0, 92.0, 92.25 ]
-az = [ 66.0, 29.0, 26.0, 8.0, 8.14 ]
-wi = [ 68.0, 60.0, 81.0, 86.0, 85.11 ]
-nv = [ 25.0, 27.0, 31.0, 55.0, 54.84]
-
+pa = [ 73.0, 53.0, 48.0, 70.0, 71.75, 70.78 ]
+ga = [ 36.0, 42.0, 18.0, 3.0,   2.97,  2.81 ]
+nc = [ 30.0, 24.0, 29.0, 16.0, 15.39, 10.00 ]
+mi = [ 73.0, 82.0, 81.0, 92.0, 92.25, 94.87 ]
+az = [ 66.0, 29.0, 26.0,  8.0,  8.14,  6.23 ]
+wi = [ 68.0, 60.0, 81.0, 86.0, 85.11, 81.35 ]
+nv = [ 25.0, 27.0, 31.0, 55.0, 54.84, 55.51 ]
+x  = 1:6
 p = plot(x, [pa ga nc mi az wi nv], 
      title="Modeled Likelihood of Harris Win",
      xlabel="Time Period",
      ylabel="Likelihood (%)",
      label=["PA" "GA" "NC" "MI" "AZ" "WI" "NV"],
-     xticks=(1:5, ["early August", "late August", "early September", "late September", "mid October"]),
+     xticks=(x, ["early August", "late August", "early September", "late September", "mid October", "late October"]),
      xrotation=45,
-     legend=:topleft,
+     legend=:false,
      linewidth=2,
      marker=:circle)
 
@@ -26,3 +26,4 @@ annotate!([(2.5, 52, ("Win", 10, :black, :center)),
            (2.5, 48, ("Loss", 10, :black, :center))])
 
 savefig("../img/avg_of_avg.png")
+

_assets/scripts/clust.jl

@@ -0,0 +1,79 @@
+using Clustering
+using CSV
+using DataFrames
+using Plots
+using Statistics
+
+df = CSV.read("objs/margins.csv",DataFrame)
+
+function cluster_data(df::DataFrame, n_clusters::Int=3)
+    # Input validation
+    if nrow(df) < n_clusters
+        error("Number of rows ($(nrow(df))) must be greater than number of clusters ($n_clusters)")
+    end
+
+    # Create position array based on order in dataframe
+    positions = collect(1:nrow(df))
+
+    # Extract features into a matrix - using position and margin
+    features = hcat(positions, df.margin)'
+
+    # Normalize features to give equal weight
+    features_normalized = (features .- mean(features, dims=2)) ./ std(features, dims=2)
+
+    # Perform k-means clustering
+    result = kmeans(features_normalized, n_clusters)
+
+    # Add cluster assignments to the original dataframe
+    df_clustered = copy(df)
+    df_clustered.cluster = result.assignments
+
+    # Calculate cluster statistics
+    cluster_stats = DataFrame()
+    for i in 1:n_clusters
+        cluster_indices = findall(x -> x == i, df_clustered.cluster)
+        cluster_data = df_clustered[cluster_indices, :]
+        stats = (
+            cluster = i,
+            size = nrow(cluster_data),
+            mean_position = mean(positions[cluster_indices]),
+            mean_margin = mean(cluster_data.margin),
+            std_position = std(positions[cluster_indices]),
+            std_margin = std(cluster_data.margin)
+        )
+        push!(cluster_stats, stats)
+    end
+
+    # Define custom colors for clusters
+    cluster_colors = [:green, :orange, :yellow]
+
+    # Create visualization with all specified formatting
+    plt = Plots.scatter(
+        positions,
+        100 .* df_clustered.margin,  # Multiply by 100 to convert to percentage
+        group=df_clustered.cluster,
+        xlabel="Order of AP Call",
+        ylabel="Democratic Margin (%)",
+        title="2020 Election",
+        legend=:topright,
+        yformatter=y->string(round(Int, y)) * "%",  # Format as percentage with no decimals
+        markersize=8,
+        framestyle=:box,
+        grid=true,
+        gridlinewidth=0.5,
+        gridstyle=:dash,
+        gridcolor=:gray,
+        palette=cluster_colors,
+        legend_title="Clusters"
+    )
+
+    # Add emphasized zero line
+    hline!(plt, [0], color=:black, linewidth=1.5, label=nothing)
+
+    return df_clustered, cluster_stats, plt
+end
+
+# Example usage:
+df_with_clusters, stats, cluster_plot = cluster_data(df)
+display(cluster_plot)
+println(stats)

_assets/scripts/cons.jl

@@ -1,5 +1,3 @@
-# const STATES   = ["NV", "WI", "AZ", "GA", "MI", "PA", "NC"]
-# const states   = ["NV", "WI", "AZ", "GA", "MI", "PA", "NC"]
 const FLAGRED  = "rgb(178,  34,  52)"
 const FLAGBLUE = "rgb( 60,  59, 110)"
 const PURPLE   = "rgb(119,  47,  81)"

_assets/scripts/data/2020age.csv

@@ -0,0 +1,39 @@
+st,cohort,totpop,cpop,reg,vot
+AZ,18 to 24,653,629,371,327
+AZ,25 to 34,941,850,619,579
+AZ,35 to 44,929,756,565,509
+AZ,45 to 64,1833,1627,1301,1242
+AZ,65+,1282,1213,1023,992
+GA,18 to 24,990,925,556,489
+GA,25 to 34,1496,1354,913,794
+GA,35 to 44,1287,1083,723,690
+GA,45 to 64,2634,2454,1854,1783
+GA,65+,1625,1584,1187,1132
+MI,18 to 24,932,887,546,469
+MI,25 to 34,1344,1257,910,777
+MI,35 to 44,1130,1063,764,666
+MI,45 to 64,2663,2547,1969,1798
+MI,65+,1721,1713,1324,1284
+NV,18 to 24,284,262,110,96
+NV,25 to 34,463,415,275,236
+NV,35 to 44,352,310,193,176
+NV,45 to 64,804,734,505,481
+NV,65+,500,477,373,363
+NC,18 to 24,872,820,465,404
+NC,25 to 34,1412,1259,809,706
+NC,35 to 44,1315,1068,705,637
+NC,45 to 64,2524,2303,1699,1638
+NC,65+,1992,1941,1483,1396
+PA,18 to 24,1049,1017,601,519
+PA,25 to 34,1789,1747,1354,1207
+PA,35 to 44,1360,1281,948,866
+PA,45 to 64,3212,3122,2491,2315
+PA,65+,2492,2453,1943,1850
+PA,18 to 24,581,577,333,323
+PA,25 to 34,762,714,582,557
+PA,35 to 44,650,617,437,426
+PA,45 to 64,1604,1572,1241,1171
+PA,65+,941,941,797,776
+"Source: U.S. Census Bureau, Current Population Survey, November 2022",,,,,
+"url: ""https://www.census.gov/data/tables/time-series/demo/voting-and-registration/p20-585.html""",,,,,
+"Table 4c. Reported Voting and Registration of the Total Voting-Age Population, by Age, for States: November 2022",,,,,

_assets/scripts/data/2020states.csv

@@ -0,0 +1,8 @@
+st,totpop,eligible,registered,pctreg`,voted,pctvot
+AZ,5638,5075,3878,76.4,3649,71.9
+GA,8032,7400,5233,70.7,4888,66.1
+MI,7790,7467,5513,73.8,4994,66.9
+NV,2402,2198,1455,66.2,1351,61.5
+NC,8113,7391,5161,69.8,4780,64.7
+PA,9902,9621,7337,76.3,6756,70.2
+WI,4538,4421,3391,76.7,3253,73.6

_assets/scripts/data/2020voter.csv

@@ -0,0 +1,145 @@
+demograph,gender,cohort,registered,voted
+Male over 18,m,121870,65.1,59.5
+18 to 24 years,m,14382,54.7,45.8
+25 to 34 years,m,22406,58.1,51.0
+35 to 44 years,m,20466,59.6,53.3
+45 to 54 years,m,19442,65.8,60.7
+55 to 64 years,m,20318,71.2,67.1
+65 to 74 years,m,15401,75.8,73.2
+75 years and over,m,9456,77.1,72.5
+18 years,m,1998,42.9,37.2
+19 years,m,1936,52.7,43.2
+20 years,m,2361,56.8,47.2
+21 years,m,2134,56.0,44.1
+22 years,m,1909,58.5,47.2
+23 years,m,2025,58.7,51.6
+24 years,m,2018,57.0,50.0
+25 years,m,2045,53.3,45.6
+26 years,m,2060,50.6,43.3
+27 years,m,2347,61.0,53.5
+28 years,m,2331,58.5,49.6
+29 years,m,2456,58.3,51.2
+30 years,m,2233,57.3,50.8
+31 years,m,2396,59.5,52.3
+32 years,m,2113,62.4,56.3
+33 years,m,2080,58.9,53.1
+34 years,m,2344,60.6,53.0
+35 years,m,2210,58.7,51.7
+36 years,m,2325,58.0,51.6
+37 years,m,2002,60.1,55.5
+38 years,m,2083,60.5,52.6
+39 years,m,1954,62.4,54.4
+40 years,m,2049,56.6,51.2
+41 years,m,1924,56.6,51.3
+42 years,m,1972,62.0,55.6
+43 years,m,1960,61.9,56.0
+44 years,m,1986,59.8,53.8
+45 years,m,1921,62.4,57.6
+46 years,m,1918,64.5,59.1
+47 years,m,1834,66.4,59.1
+48 years,m,1937,64.2,58.7
+49 years,m,1925,64.0,60.5
+50 years,m,2005,63.4,59.7
+51 years,m,2082,68.4,61.7
+52 years,m,1958,68.5,65.1
+53 years,m,1999,68.0,61.8
+54 years,m,1863,68.1,63.3
+55 years,m,2186,66.6,61.5
+56 years,m,1951,72.0,67.6
+57 years,m,1989,69.8,65.4
+58 years,m,2091,70.2,65.3
+59 years,m,1961,72.3,67.7
+60 years,m,2196,69.0,66.0
+61 years,m,2119,68.0,64.3
+62 years,m,1902,73.1,68.3
+63 years,m,1962,77.8,74.4
+64 years,m,1960,74.8,71.2
+65 years,m,1909,73.7,70.4
+66 years,m,1765,74.5,72.0
+67 years,m,1705,75.9,73.1
+68 years,m,1610,75.4,72.6
+69 years,m,1519,75.9,73.3
+70 years,m,1441,73.3,71.5
+71 years,m,1347,74.7,73.0
+72 years,m,1536,76.9,75.2
+73 years,m,1318,82.7,78.8
+74 years,m,1251,76.9,73.3
+75 years,m,1073,76.6,75.2
+76 years,m,901,82.0,77.7
+77 years,m,806,79.9,76.3
+78 years,m,859,81.5,77.4
+79 years,m,749,79.7,76.1
+80-84 years,m,2838,77.7,73.0
+85 years and over,m,2229,71.0,64.1
+Female  18 years and over,f,130404,68.2,63.0
+18 to 24 years,f,14277,56.8,50.2
+25 to 34 years,f,22466,63.8,56.4
+35 to 44 years,f,21092,64.7,59.3
+45 to 54 years,f,20316,68.5,64.0
+55 to 64 years,f,21837,73.4,69.7
+65 to 74 years,f,17553,76.0,72.8
+75 years and over,f,12864,74.4,68.5
+18 years,f,1928,48.4,43.1
+19 years,f,1872,55.4,50.5
+20 years,f,2213,58.0,50.8
+21 years,f,2045,53.7,48.0
+22 years,f,2108,62.7,53.8
+23 years,f,1935,62.4,53.8
+24 years,f,2176,56.7,50.7
+25 years,f,2010,62.1,54.2
+26 years,f,2201,61.6,54.4
+27 years,f,2389,63.9,55.2
+28 years,f,2187,62.7,54.4
+29 years,f,2413,64.5,58.6
+30 years,f,2504,64.2,55.4
+31 years,f,2225,64.0,56.4
+32 years,f,2212,67.4,61.5
+33 years,f,2304,62.9,56.5
+34 years,f,2022,64.0,57.3
+35 years,f,2295,64.4,57.5
+36 years,f,2264,65.4,59.1
+37 years,f,1992,61.2,55.8
+38 years,f,2325,70.5,64.5
+39 years,f,1970,66.1,61.3
+40 years,f,2250,64.3,59.2
+41 years,f,2028,65.6,60.1
+42 years,f,2052,61.9,58.4
+43 years,f,1944,63.2,59.4
+44 years,f,1974,63.6,57.2
+45 years,f,2101,62.9,59.0
+46 years,f,1896,67.4,61.9
+47 years,f,1956,65.2,60.8
+48 years,f,1984,68.5,63.0
+49 years,f,2038,68.6,64.1
+50 years,f,2192,71.9,68.4
+51 years,f,2180,67.7,62.6
+52 years,f,2042,70.4,65.4
+53 years,f,1975,72.8,68.9
+54 years,f,1951,69.9,65.3
+55 years,f,2299,70.4,67.5
+56 years,f,2126,68.1,63.9
+57 years,f,2006,77.7,71.9
+58 years,f,2117,70.5,66.9
+59 years,f,2293,74.0,69.6
+60 years,f,2326,74.6,71.4
+61 years,f,2241,71.5,68.3
+62 years,f,2125,76.5,73.0
+63 years,f,2116,77.2,74.3
+64 years,f,2187,74.1,70.8
+65 years,f,2084,73.8,71.7
+66 years,f,1979,77.5,73.8
+67 years,f,1895,74.9,71.6
+68 years,f,1826,74.3,71.8
+69 years,f,1766,78.0,74.6
+70 years,f,1740,72.6,69.7
+71 years,f,1642,76.5,73.0
+72 years,f,1640,79.5,75.5
+73 years,f,1633,78.4,74.3
+74 years,f,1348,75.4,72.5
+75 years,f,1176,81.6,76.2
+76 years,f,1228,77.1,73.6
+77 years,f,1138,77.1,74.3
+78 years,f,1077,81.1,78.1
+79 years,f,873,75.0,70.0
+80-84 years,f,3715,74.0,67.9
+85 years and over,f,3656,68.7,59.8

_assets/scripts/data/2022age.csv

@@ -0,0 +1,39 @@
+st,cohort,totpop,cpop,reg,vot
+AZ,18 to 24 ,751,694,322,170
+AZ,25 to 34 ,911,782,509,300
+AZ,35 to 44 ,937,800,532,428
+AZ,45 to 64 ,1800,1530,1148,975
+AZ,65  and over,1332,1287,1047,971
+GA,18 to 24 ,924,859,466,322
+GA,25 to 34 ,1620,1395,918,640
+GA,35 to 44 ,1400,1176,812,628
+GA,45 to 64 ,2610,2451,1695,1457
+GA,65  and over,1760,1720,1384,1276
+NC,18 to 24 ,921,866,372,40.3
+NC,25 to 34 ,1395,1236,652,46.7
+NC,35 to 44 ,1290,1069,675,52.3
+NC,45 to 64 ,2687,2523,1581,58.9
+NC,65  and over,1881,1838,1303,69.3
+MI,18 to 24 ,826,799,532,371
+MI,25 to 34 ,1304,1223,852,622
+MI,35 to 44 ,1228,1183,894,689
+MI,45 to 64 ,2639,2545,2025,1710
+MI,65  and over,1779,1767,1494,1365
+NV,18 to 24 ,297,287,122,63
+NV,25 to 34 ,414,355,222,153
+NV,35 to 44 ,406,358,229,170
+NV,45 to 64 ,789,694,458,378
+NV,65  and over,544,511,405,358
+PA,18 to 24 ,1145,1093,610,437
+PA,25 to 34 ,1680,1611,1041,776
+PA,35 to 44 ,1531,1397,1007,784
+PA,45 to 64 ,3139,3043,2328,1995
+PA,65  and over,2630,2597,2024,1851
+WI,18 to 24 ,588,579,341,286
+WI,25 to 34 ,724,679,474,371
+WI,35 to 44 ,693,669,447,373
+WI,45 to 64 ,1473,1440,1107,908
+WI,65  and over,1113,1095,856,777
+"Source: U.S. Census Bureau, Current Population Survey, November 2022",,,,,
+https://www.census.gov/data/tables/time-series/demo/voting-and-registration/p20-586.html,,,,,
+"Table 4c. Reported Voting and Registration of the Total Voting-Age Population, by Age, for States: November 2022",,,,,