One of my favorite graphs of all time is this one from Kathy Sierra:

I used that in my RStudio::conf(2019) presentation in Austin last week. I’m working on a related prez and I decided to create my own version of Kathy’s chart using R.

I played around with some simulations and found that the cumulative *t* and *normal* distributions gave good curves if I fiddled with their parameters. I wanted my lines a little bit jagged and not overly smooth so I opted to simulate 1000 points:

```
n <- 1000
df <- data.frame(curve = 'Learn Fast',
x = rt(n, 100))
df <- rbind(df, data.frame(curve = 'Struggle Long',
x = rnorm(n, 3, 2)))
```

Then it’s an exercise in building up all the `ggplot`

options that I want:

```
ggplot(df, aes(x, color = curve)) +
stat_ecdf(geom = "step", size = 2) +
theme_classic(base_size = 18) +
scale_colour_wsj("colors6") ->
p
p
```

```
p +
theme(
axis.text.x = element_blank(),
axis.ticks.x = element_blank(),
axis.text.y = element_blank(),
axis.ticks.y = element_blank()
) +
theme(legend.position = c(0.75, 0.15)) +
theme(legend.title = element_blank()) +
labs(x = "Time", y = "Skill") ->
p
p
```

Now I want to add the horizontal lines and their labels:

```
p +
geom_hline(yintercept = .25,
color = "orangered",
linetype = 2) +
annotate(
"text",
x = -2.4,
y = .29,
label = "Suck Threshold",
size = 5
) +
geom_hline(yintercept = .75,
color = "forestgreen",
linetype = 2) +
annotate(
"text",
x = -2. ,
y = .79,
label = "Kick Ass Threshold",
size = 5
)
```

That looks pretty good. I think I like it!