TidyTuesday: Trading P/L Calendar Heatmap

TidyTuesday

Data Viz

Finance

Trading

A traditional calendar visualization showing daily realized profit/loss from stock trading using ggplot2

Author

chokotto

Published

January 21, 2026

Overview

This week’s TidyTuesday visualizes my personal stock trading realized profit/loss in a traditional calendar format. Using ggplot2’s geom_tile(), each day displays: - Date number - Profit/Loss amount (in JPY) - Color coding: Green for profit, Red for loss, Gray for no trading

This visualization makes it easy to identify profitable and losing days across the year.

Dataset

Source: Personal trading records from Rakuten Securities (JP/US) and SBI Securities (US)

Show code

# Increase R stack size to prevent overflow with large plots
options(expressions = 50000)

library(tidyverse)
library(scales)
library(arrow)

# Load TidyTuesday helpers
# Project root is 5 levels up from this qmd file
project_root <- normalizePath(file.path(getwd(), "../../../../.."))
helpers_path <- file.path(project_root, "config", "R", "tidytuesday_helpers.R")
if (file.exists(helpers_path)) {
  source(helpers_path, local = FALSE)
} else {
  message("Helpers not found at: ", helpers_path)
}

Show code

# Load aggregated daily P/L data (pre-processed)
daily_pl_path <- file.path(getwd(), "data", "daily_pl.parquet")
daily_pl <- read_parquet(daily_pl_path) %>%
  mutate(date = as.Date(date))

cat("Trading days:", nrow(daily_pl), "\n")

Trading days: 231

Show code

cat("Date range:", as.character(min(daily_pl$date)), "to", as.character(max(daily_pl$date)), "\n")

Date range: 2021-02-22 to 2026-01-16

Show code

cat("Profitable days:", sum(daily_pl$profit_jpy > 0), "\n")

Profitable days: 162

Show code

cat("Losing days:", sum(daily_pl$profit_jpy < 0), "\n")

Losing days: 69

Show code

daily_pl %>% head()

# A tibble: 6 × 3
  date       profit_jpy trades
  <date>          <dbl>  <int>
1 2021-02-22      59825      1
2 2021-03-02     431465      1
3 2021-09-21     -74794      1
4 2021-09-24     -62818      1
5 2021-11-25     -34035      1
6 2021-12-02      52740      4

Calendar Visualization Function

Show code

create_calendar_heatmap <- function(data, year, title_suffix = "") {
  # Filter for specified year
  year_data <- data %>%
    filter(year(date) == year) %>%
    mutate(
      month = month(date),
      day = day(date),
      wday = wday(date, week_start = 1),  # Monday = 1
      week_of_month = ceiling((day + wday(floor_date(date, "month"), week_start = 1) - 1) / 7)
    )

  # Create full calendar grid
  all_dates <- tibble(
    date = seq(ymd(paste0(year, "-01-01")), ymd(paste0(year, "-12-31")), by = "day")
  ) %>%
    mutate(
      month = month(date),
      day = day(date),
      wday = wday(date, week_start = 1),
      week_of_month = ceiling((day + wday(floor_date(date, "month"), week_start = 1) - 1) / 7),
      month_name = factor(month.name[month], levels = month.name)
    )

  # Join with trading data
  calendar_data <- all_dates %>%
    left_join(year_data %>% select(date, profit_jpy, trades), by = "date") %>%
    mutate(
      status = case_when(
        is.na(profit_jpy) ~ "no_trade",
        profit_jpy > 0 ~ "profit",
        profit_jpy < 0 ~ "loss",
        TRUE ~ "no_trade"
      ),
      # Format label
      label = case_when(
        is.na(profit_jpy) ~ "",
        abs(profit_jpy) >= 1000000 ~ sprintf("%+.1fM", profit_jpy / 1000000),
        abs(profit_jpy) >= 1000 ~ sprintf("%+.0fK", profit_jpy / 1000),
        TRUE ~ sprintf("%+.0f", profit_jpy)
      )
    )

  # Weekday labels
  wday_labels <- c("Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun")

  # Create plot
  p <- ggplot(calendar_data, aes(x = wday, y = week_of_month)) +
    # Tiles
    geom_tile(
      aes(fill = profit_jpy),
      color = "white",
      linewidth = 0.5
    ) +
    # Day numbers
    geom_text(
      aes(label = day),
      vjust = -0.3,
      size = 3,
      fontface = "bold",
      color = "#334155"
    ) +
    # P/L labels
    geom_text(
      aes(
        label = label,
        color = status
      ),
      vjust = 1.2,
      size = 2.2,
      fontface = "plain"
    ) +
    # Color scales
    scale_fill_gradient2(
      low = "#ef4444",
      mid = "#f1f5f9",
      high = "#22c55e",
      midpoint = 0,
      na.value = "#f1f5f9",
      limits = c(-max(abs(calendar_data$profit_jpy), na.rm = TRUE),
                  max(abs(calendar_data$profit_jpy), na.rm = TRUE)),
      labels = label_number(scale = 1e-6, suffix = "M"),
      name = "P/L (JPY)"
    ) +
    scale_color_manual(
      values = c("profit" = "#166534", "loss" = "#991b1b", "no_trade" = "#94a3b8"),
      guide = "none"
    ) +
    scale_x_continuous(
      breaks = 1:7,
      labels = wday_labels,
      expand = c(0, 0)
    ) +
    scale_y_reverse(
      expand = c(0.02, 0)
    ) +
    # Facet by month
    facet_wrap(~ month_name, ncol = 4, scales = "free_y") +
    # Labels
    labs(
      title = paste0(year, " Trading P/L Calendar", title_suffix),
      subtitle = "Daily realized profit/loss from stock trading",
      caption = tt_caption(source = "Personal trading records"),
      x = NULL,
      y = NULL
    ) +
    # Theme
    theme_minimal(base_size = 11) +
    theme(
      plot.title = element_text(face = "bold", size = 18, hjust = 0.5),
      plot.subtitle = element_text(color = "#64748b", size = 12, hjust = 0.5),
      plot.caption = element_text(color = "#94a3b8", size = 9),
      strip.text = element_text(face = "bold", size = 12),
      panel.grid = element_blank(),
      axis.text.y = element_blank(),
      axis.text.x = element_text(size = 8, color = "#64748b"),
      legend.position = "bottom",
      legend.key.width = unit(2, "cm"),
      plot.margin = margin(20, 20, 20, 20)
    )

  return(p)
}

2025 Calendar

Show code

p_2025 <- create_calendar_heatmap(daily_pl, 2025)
p_2025

2026 Calendar (Year to Date)

Show code

p_2026 <- create_calendar_heatmap(daily_pl, 2026, " (YTD)")
p_2026

Monthly Summary Statistics

Show code

# Calculate monthly statistics for 2025
monthly_stats <- daily_pl %>%
  filter(year(date) == 2025) %>%
  mutate(month = month(date, label = TRUE, abbr = FALSE)) %>%
  group_by(month) %>%
  summarise(
    `Total P/L` = sum(profit_jpy),
    `Trading Days` = n(),
    `Avg P/L` = mean(profit_jpy),
    `Profitable Days` = sum(profit_jpy > 0),
    `Win Rate` = mean(profit_jpy > 0) * 100
  ) %>%
  mutate(
    `Total P/L` = scales::number(`Total P/L`, prefix = "¥", big.mark = ","),
    `Avg P/L` = scales::number(`Avg P/L`, prefix = "¥", big.mark = ","),
    `Win Rate` = paste0(round(`Win Rate`, 1), "%")
  )

monthly_stats %>%
  knitr::kable()

month	Total P/L	Trading Days	Avg P/L	Profitable Days	Win Rate
January	¥14,432	4	¥3,608	3	75%
February	-¥46,363	8	-¥5,795	6	75%
March	-¥3,282,961	10	-¥328,296	4	40%
May	¥1,811,121	11	¥164,647	9	81.8%
June	¥1,823,545	12	¥151,962	12	100%
July	¥533,590	5	¥106,718	4	80%
August	¥3,253,378	13	¥250,260	10	76.9%
September	¥2,643,828	12	¥220,319	11	91.7%
October	¥5,055,854	16	¥315,991	14	87.5%
November	¥2,780,676	8	¥347,584	8	100%
December	¥1,112,490	4	¥278,122	4	100%

Key Insights

Show code

# Calculate key metrics
total_2025 <- daily_pl %>%
  filter(year(date) == 2025) %>%
  summarise(total = sum(profit_jpy)) %>%
  pull(total)

best_day <- daily_pl %>%
  filter(profit_jpy == max(profit_jpy))

worst_day <- daily_pl %>%
  filter(profit_jpy == min(profit_jpy))

avg_profit <- daily_pl %>%
  filter(profit_jpy > 0) %>%
  summarise(avg = mean(profit_jpy)) %>%
  pull(avg)

avg_loss <- daily_pl %>%
  filter(profit_jpy < 0) %>%
  summarise(avg = mean(profit_jpy)) %>%
  pull(avg)

cat("=== 2025 Trading Summary ===\n")

=== 2025 Trading Summary ===

Show code

cat("Total P/L:", scales::number(total_2025, prefix = "¥", big.mark = ","), "\n")

Total P/L: ¥15,699,590

Show code

cat("Best Day:", as.character(best_day$date), "(", scales::number(best_day$profit_jpy, prefix = "¥", big.mark = ","), ")\n")

Best Day: 2024-01-05 ( ¥2,087,731 )

Show code

cat("Worst Day:", as.character(worst_day$date), "(", scales::number(worst_day$profit_jpy, prefix = "¥", big.mark = ","), ")\n")

Worst Day: 2024-08-07 ( -¥3,478,929 )

Show code

cat("Average Profit (winning days):", scales::number(avg_profit, prefix = "¥", big.mark = ","), "\n")

Average Profit (winning days): ¥301,577

Show code

cat("Average Loss (losing days):", scales::number(avg_loss, prefix = "¥", big.mark = ","), "\n")

Average Loss (losing days): -¥373,637

Design Approach

ggplot2 + geom_tile(): Creates a clean grid-based calendar layout with proper spacing.
Diverging Color Scale: scale_fill_gradient2() centers at zero, making profits (green) and losses (red) immediately distinguishable.
Faceted Layout: facet_wrap() with 4 columns displays all 12 months in a compact 3x4 grid.
Dual Text Labels: Day numbers at top, P/L values below, with conditional coloring for readability.
Clean Theme: Minimal gridlines and muted colors keep focus on the data.

This post is part of the TidyTuesday weekly data visualization project.

Disclaimer

This analysis is for educational and practice purposes only. The profit/loss figures shown are from personal trading records and should not be considered as investment advice or performance guarantees.

--- title: "TidyTuesday: Trading P/L Calendar Heatmap" description: "A traditional calendar visualization showing daily realized profit/loss from stock trading using ggplot2" date: "2026-01-21" x-posted: true author: "chokotto" categories: ["TidyTuesday", "Data Viz", "R", "Finance", "Trading"] image: "thumbnail.svg" engine: knitr freeze: false execute: warning: false message: false code-fold: true code-tools: true code-summary: "Show code" twitter-card: card-type: summary_large_image image: "thumbnail.png" title: "TidyTuesday: Trading P/L Calendar" description: "Daily profit/loss calendar heatmap with ggplot2" --- ## Overview This week's TidyTuesday visualizes my personal stock trading realized profit/loss in a traditional calendar format. Using ggplot2's `geom_tile()`, each day displays: - **Date number** - **Profit/Loss amount** (in JPY) - **Color coding**: Green for profit, Red for loss, Gray for no trading This visualization makes it easy to identify profitable and losing days across the year. ## Dataset **Source**: Personal trading records from Rakuten Securities (JP/US) and SBI Securities (US) ```{r} #| label: load-packages #| message: false #| warning: false # Increase R stack size to prevent overflow with large plots options(expressions = 50000) library(tidyverse) library(scales) library(arrow) # Load TidyTuesday helpers # Project root is 5 levels up from this qmd file project_root <- normalizePath(file.path(getwd(), "../../../../..")) helpers_path <- file.path(project_root, "config", "R", "tidytuesday_helpers.R") if (file.exists(helpers_path)) { source(helpers_path, local = FALSE) } else { message("Helpers not found at: ", helpers_path) } ``` ```{r} #| label: load-data #| message: false # Load aggregated daily P/L data (pre-processed) daily_pl_path <- file.path(getwd(), "data", "daily_pl.parquet") daily_pl <- read_parquet(daily_pl_path) %>% mutate(date = as.Date(date)) cat("Trading days:", nrow(daily_pl), "\n") cat("Date range:", as.character(min(daily_pl$date)), "to", as.character(max(daily_pl$date)), "\n") cat("Profitable days:", sum(daily_pl$profit_jpy > 0), "\n") cat("Losing days:", sum(daily_pl$profit_jpy < 0), "\n") daily_pl %>% head() ``` ## Calendar Visualization Function ```{r} #| label: calendar-function create_calendar_heatmap <- function(data, year, title_suffix = "") { # Filter for specified year year_data <- data %>% filter(year(date) == year) %>% mutate( month = month(date), day = day(date), wday = wday(date, week_start = 1), # Monday = 1 week_of_month = ceiling((day + wday(floor_date(date, "month"), week_start = 1) - 1) / 7) ) # Create full calendar grid all_dates <- tibble( date = seq(ymd(paste0(year, "-01-01")), ymd(paste0(year, "-12-31")), by = "day") ) %>% mutate( month = month(date), day = day(date), wday = wday(date, week_start = 1), week_of_month = ceiling((day + wday(floor_date(date, "month"), week_start = 1) - 1) / 7), month_name = factor(month.name[month], levels = month.name) ) # Join with trading data calendar_data <- all_dates %>% left_join(year_data %>% select(date, profit_jpy, trades), by = "date") %>% mutate( status = case_when( is.na(profit_jpy) ~ "no_trade", profit_jpy > 0 ~ "profit", profit_jpy < 0 ~ "loss", TRUE ~ "no_trade" ), # Format label label = case_when( is.na(profit_jpy) ~ "", abs(profit_jpy) >= 1000000 ~ sprintf("%+.1fM", profit_jpy / 1000000), abs(profit_jpy) >= 1000 ~ sprintf("%+.0fK", profit_jpy / 1000), TRUE ~ sprintf("%+.0f", profit_jpy) ) ) # Weekday labels wday_labels <- c("Mon", "Tue", "Wed", "Thu", "Fri", "Sat", "Sun") # Create plot p <- ggplot(calendar_data, aes(x = wday, y = week_of_month)) + # Tiles geom_tile( aes(fill = profit_jpy), color = "white", linewidth = 0.5 ) + # Day numbers geom_text( aes(label = day), vjust = -0.3, size = 3, fontface = "bold", color = "#334155" ) + # P/L labels geom_text( aes( label = label, color = status ), vjust = 1.2, size = 2.2, fontface = "plain" ) + # Color scales scale_fill_gradient2( low = "#ef4444", mid = "#f1f5f9", high = "#22c55e", midpoint = 0, na.value = "#f1f5f9", limits = c(-max(abs(calendar_data$profit_jpy), na.rm = TRUE), max(abs(calendar_data$profit_jpy), na.rm = TRUE)), labels = label_number(scale = 1e-6, suffix = "M"), name = "P/L (JPY)" ) + scale_color_manual( values = c("profit" = "#166534", "loss" = "#991b1b", "no_trade" = "#94a3b8"), guide = "none" ) + scale_x_continuous( breaks = 1:7, labels = wday_labels, expand = c(0, 0) ) + scale_y_reverse( expand = c(0.02, 0) ) + # Facet by month facet_wrap(~ month_name, ncol = 4, scales = "free_y") + # Labels labs( title = paste0(year, " Trading P/L Calendar", title_suffix), subtitle = "Daily realized profit/loss from stock trading", caption = tt_caption(source = "Personal trading records"), x = NULL, y = NULL ) + # Theme theme_minimal(base_size = 11) + theme( plot.title = element_text(face = "bold", size = 18, hjust = 0.5), plot.subtitle = element_text(color = "#64748b", size = 12, hjust = 0.5), plot.caption = element_text(color = "#94a3b8", size = 9), strip.text = element_text(face = "bold", size = 12), panel.grid = element_blank(), axis.text.y = element_blank(), axis.text.x = element_text(size = 8, color = "#64748b"), legend.position = "bottom", legend.key.width = unit(2, "cm"), plot.margin = margin(20, 20, 20, 20) ) return(p) } ``` ## 2025 Calendar ```{r} #| label: calendar-2025 #| fig-width: 14 #| fig-height: 10 p_2025 <- create_calendar_heatmap(daily_pl, 2025) p_2025 ``` ## 2026 Calendar (Year to Date) ```{r} #| label: calendar-2026 #| fig-width: 14 #| fig-height: 10 p_2026 <- create_calendar_heatmap(daily_pl, 2026, " (YTD)") p_2026 ``` ## Monthly Summary Statistics ```{r} #| label: monthly-summary # Calculate monthly statistics for 2025 monthly_stats <- daily_pl %>% filter(year(date) == 2025) %>% mutate(month = month(date, label = TRUE, abbr = FALSE)) %>% group_by(month) %>% summarise( `Total P/L` = sum(profit_jpy), `Trading Days` = n(), `Avg P/L` = mean(profit_jpy), `Profitable Days` = sum(profit_jpy > 0), `Win Rate` = mean(profit_jpy > 0) * 100 ) %>% mutate( `Total P/L` = scales::number(`Total P/L`, prefix = "¥", big.mark = ","), `Avg P/L` = scales::number(`Avg P/L`, prefix = "¥", big.mark = ","), `Win Rate` = paste0(round(`Win Rate`, 1), "%") ) monthly_stats %>% knitr::kable() ``` ## Key Insights ```{r} #| label: insights # Calculate key metrics total_2025 <- daily_pl %>% filter(year(date) == 2025) %>% summarise(total = sum(profit_jpy)) %>% pull(total) best_day <- daily_pl %>% filter(profit_jpy == max(profit_jpy)) worst_day <- daily_pl %>% filter(profit_jpy == min(profit_jpy)) avg_profit <- daily_pl %>% filter(profit_jpy > 0) %>% summarise(avg = mean(profit_jpy)) %>% pull(avg) avg_loss <- daily_pl %>% filter(profit_jpy < 0) %>% summarise(avg = mean(profit_jpy)) %>% pull(avg) cat("=== 2025 Trading Summary ===\n") cat("Total P/L:", scales::number(total_2025, prefix = "¥", big.mark = ","), "\n") cat("Best Day:", as.character(best_day$date), "(", scales::number(best_day$profit_jpy, prefix = "¥", big.mark = ","), ")\n") cat("Worst Day:", as.character(worst_day$date), "(", scales::number(worst_day$profit_jpy, prefix = "¥", big.mark = ","), ")\n") cat("Average Profit (winning days):", scales::number(avg_profit, prefix = "¥", big.mark = ","), "\n") cat("Average Loss (losing days):", scales::number(avg_loss, prefix = "¥", big.mark = ","), "\n") ``` ## Design Approach 1. **ggplot2 + geom_tile()**: Creates a clean grid-based calendar layout with proper spacing. 2. **Diverging Color Scale**: `scale_fill_gradient2()` centers at zero, making profits (green) and losses (red) immediately distinguishable. 3. **Faceted Layout**: `facet_wrap()` with 4 columns displays all 12 months in a compact 3x4 grid. 4. **Dual Text Labels**: Day numbers at top, P/L values below, with conditional coloring for readability. 5. **Clean Theme**: Minimal gridlines and muted colors keep focus on the data. ------------------------------------------------------------------------ *This post is part of the [TidyTuesday](https://github.com/rfordatascience/tidytuesday) weekly data visualization project.* :::: {.callout-caution collapse="false" appearance="minimal" icon="false"} ## Disclaimer ::: {style="font-size: 0.85em; color: #64748b; line-height: 1.6;"} This analysis is for educational and practice purposes only. The profit/loss figures shown are from personal trading records and should not be considered as investment advice or performance guarantees. ::: ::::