Who This Is For

This article is for readers who can already build a solid ggplot, but are not yet confident about the finishing steps that make a figure ready to share. The goal is to teach practical add-ons such as smoothing, summary layers, annotations, and export settings.

What You Will Do

Add a trend line with geom_smooth().
Build a grouped summary plot with error bars.
Highlight points or regions with annotations.
Export figures reproducibly with ggsave().

Before You Start

You should already know the core ggplot workflow.
You need ggplot2, palmerpenguins, dplyr, and ggrepel.
You should understand that export settings are part of the final figure design, not an afterthought.

The companion script for this article is:

R draw/scripts/07-ggplot-from-zero-summaries-annotations-export.R

Step 1: Add a Trend with `geom_smooth()`

One of the fastest ways to help readers see a broad pattern is to add a smoothing layer.

ggplot(
  penguins_clean,
  aes(x = bill_length_mm, y = flipper_length_mm, color = species)
) +
  geom_point(alpha = 0.7, size = 2.1) +
  geom_smooth(se = FALSE, linewidth = 1)

geom_smooth() does not replace the original points. It adds a summary of the overall relationship.

Step 2: Summarize Groups and Add Error Bars

Some plots are more useful when you show grouped summaries instead of every raw point.

geom_col(width = 0.6) +
geom_errorbar(
  aes(ymin = mean_depth - se_depth, ymax = mean_depth + se_depth),
  width = 0.15
)

This pattern works well when you want to show a group average and a simple uncertainty interval.

Step 3: Use Annotations to Direct Attention

Annotations are especially helpful when one or two points matter more than the rest.

geom_label_repel(...) +
annotate("segment", ...) +
annotate("text", ...)

The point is not to decorate the figure. The point is to make interpretation easier.

Step 4: Export the Figure Reproducibly

Use ggsave() when you want stable export settings:

ggsave(
  filename = "R draw/figures/07-smooth-lines.png",
  plot = plot_smooth,
  width = 7.5,
  height = 5.5,
  dpi = 300
)

Important parameters:

filename controls where the file is written
width and height control physical output size
dpi controls raster sharpness

The companion script also exports a PDF version so readers can see that one plot may need more than one output format.

Show Explanation

Export Rule of Thumb

Use PNG for blog posts, slide decks, and quick sharing.
Use PDF when you want a vector format for print or editing workflows.
Decide size before export. Stretching a tiny raster image later usually looks worse than exporting correctly once.

```

How to Confirm It Worked

Your script creates:
- R draw/figures/07-smooth-lines.png
- R draw/figures/07-summary-error-bars.png
- R draw/figures/07-annotation-example.png
- R draw/figures/07-smooth-lines.pdf
You can explain why the annotation layer is there.
You can export one figure to both PNG and PDF without changing the plot logic.

Common Questions

Does `geom_smooth()` prove a real statistical relationship?

No. It helps visualize a trend, but it is still a visual summary, not a complete statistical argument.

When should I use error bars?

Use them when your audience needs a summary of spread or uncertainty and when you can explain what the interval represents.

Why not annotate everything?

Because too many annotations compete with the data. Good annotation is selective.

Review Score

Score: 94/100 Verdict: This draft is ready for human review and covers the practical finishing tools that many beginners need before they can publish a plot confidently.