When did employees join Lumon?

This analysis was made possible by the mdr R package. It estimates when the employees joined Lumon based on their employee codes printed on their badges. In episode 1 of the first season (around forty minutes and thirty nine seconds into the episode), Mark mentions that he has been at Lumon for around 2 years. A Linkedin ad campaign mentioned that Irving joined Lumon 9 years ago. Indexing off of Helly’s start date (Day 0), we predicted days since joining Lumon using the employee code by fitting a polynomial regression model to these data where their employee code was raised to the 30th power to account for the potential non-linear relationship. While this approach likely does not generalize well beyond the observed data, nevertheless we have made a plot of the predicted values.

Please enjoy all predictions equally.

library(mdr)
library(tidyverse)
library(ggiraph)

preds <- tibble(id_numeric = 8039:8988)

preds$joined_predicted <- lm(joined ~ I(id_numeric^30), badges) |>
  predict(newdata = preds)

p <- ggplot(badges, 
            aes(x = id_numeric,
                y = joined_predicted,
                tooltip = glue::glue("test"))) +
  geom_point_interactive(aes(tooltip = glue::glue("{name}<br>employee code: {id}<br>predicted start: {round(joined_predicted / 365, 1)} years ago")),
                         color = "#6AF307",
                         size = 3) +
  geom_line(data = preds, aes(y = joined_predicted), color = "white",
            linetype = "dashed") + 
  scale_y_continuous(
    breaks = badges$joined_predicted,
    labels = badges$name[order(badges$joined_predicted, decreasing = TRUE)],
    sec.axis = sec_axis(~ ., breaks = badges$joined_predicted,
                        labels = round(badges$joined_predicted / 365, 1),
                        name = "Predicted Start (Years Ago)")
  ) +
  scale_x_continuous(
    breaks = badges$id_numeric,
    labels = function(x) glue::glue("08-{substr(x, 2, 4)}")
  ) +
  labs(x = "Employee Code", y = "") +
  annotate("label", x = 8907, y = 365*2,
           size = 3,
           label = "~2 years\nsource: S1E1 (40:39)",
           hjust = 1,
           fill = "#CFE0E1",
           family = "mono") +
  annotate("label", x = 8430, y = 365*9,
           size = 3,
           label = "~9 years\nsource: Linkedin",
           hjust = 1,
           fill = "#CFE0E1",
           family = "mono") +
  theme_mdr() +
  theme(
    panel.grid.minor = element_blank(),
    axis.text.x = element_text(angle = 45, hjust = 1)
  )

ggiraph::girafe(ggobj = p)