![]() Select(candidate, nickname, starts_with("sim_")) %>% ![]() `pvote_norm_Black Voters`:`pvote_norm_White Moderates`, Mutate(total = 0.35 * `pvote_norm_Black Voters` 0.3 * `pvote_norm_Wealthy Progressives` 0.25 * `pvote_norm_White Moderates` 0.1 * `pvote_norm_Hispanic Voters`) %>% Values_from = c(pvote, pvote_norm, votes),Ĭandidate="ROBERT A BRADY" ~ "BOB BRADY",Ĭandidate="JOSEPH R BIDEN" ~ "JOE BIDEN", ![]() Id_cols = c(year, office_pretty, candidate, winnum), Group_by(year, office_pretty, candidate, cat, winnum) %>% Left_join(div_cats %>% select(-year)) %>% Mutate(winnum = (winner_pvote second_pvote)/2) %>% Proportion of the vote received in bloc \(i\).Ĭonsider, for example, just the Black Voter and Wealthy Progressiveįilter(candidate != "Write In", party = "DEMOCRATIC") %>% The winner will be the candidate who achieves \[Ġ.35 p_ \ge 0.25, What kinds of coalitions could put a candidate over the top? Let’sĪssume a candidate (a) needs 25% of the vote to win, and (b) theīreakdown of votes is 35% Black Voter divisions, 30% Wealthy Progressiveĭivisions, 25% White Moderate divisions, and 10% Hispanic Voter Title = "Voting Bloc proportions of the vote" Scale_color_manual(NULL, values=cat_colors) Mutate(total_votes = sum(votes), pvote = votes / total_votes) %>%Īes(x = asnum(year), y = 100*pvote, color=cat) Left_join(div_cats %>% select(-year), by = "warddiv") %>% Let’s say Black Voter divisions will cast more than 35% of votes, Wealthy Progressives about 30% of the vote, White ModeratesĢ5%, and Hispanic North Philly 10%. Low proportion of the vote since November 2020. Philadelphia’s Black Wards have had a relatively # theme(legend.position="bottom", legend.direction="horizontal") Scale_fill_manual(NULL, values=cat_colors) Mutate(warddiv = pretty_div(DIVISION_N))source("././data/prep_data/div_svd_time_util.R")ĭiv_cat_fn % get_row_cats(2017) %>% rename(warddiv = row_id) I’ve analysed these patterns before, creating my voting Or above 10 come May, especially with such high-profile names, that winĭivision across the city turn out or stay home in patterns. We haven’t seen 10 candidates in a recent election. The win percentage at ten candidates looks like it will be 25%, or lower. Y = "Win Percent\navg(first place, second place)" Subtitle = "Philadelphia Democratic Primaries", Title = "With 10 candidates, the win number could be < 25%", Ggrepel::geom_text_repel(aes(label=paste(year, office_pretty))) Geom_point(size=3, color = strong_purple) # filter(election_type = "primary", party = "DEMOCRATIC") %>%Īes(x=ncand, y=100*(winner_pvote second_pvote) / 2) Office = "UNITED STATES SENATOR" ~ "Senate", Office = "PRESIDENT OF THE UNITED STATES" ~ "President", Group_by(year, election_type, office) %>% ![]() Group_by(year, election_type, office, candidate) %>% Ifelse(election_type = "primary", party = "DEMOCRATIC"), View code comp_elections % inner_join(comp_elections) %>% So it looks like it took ~30% to win (halfway between first and second). In 2017, Larry Krasner beat Joe Khanģ8% to 20%. The two most competitive recent, many-candidate races wereĢ007 Mayor and 2017 D.A., both with seven candidates. With so many candidates, the winner won’t need a very high Given that turnout hasĭramatically jumped post-2016 and this race is shaping up to be hyperĬompetitive, I’d expect turnout around that 310,000 mark or higher. In the last two competitive mayoral races, 20, Mayoral primaries usually see the second highest turnout, after Title="Votes cast in Philadelphia elections" Group_by(year, election_type, office, district, warddiv) %>%Ĭycle_colors % mutate(election_type = "general"),įacet_grid(~format_name(election_type)) Group_by(office, candidate, party, warddiv, year, election_type, district, ward, is_topline_office) %>% ![]() Setwd("C:/Users/Jonathan Tannen/Dropbox/sixty_six/posts/council_ballot_position_23/") How many voters should we expect? View code library(tidyverse) ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |