← Back to Paradoxes

The Waiting Time Paradox

Why does the bus always seem late?
You're more likely to arrive during a long gap than a short one!

Bus Stop Simulation

Regular Buses

Every 10 minutes exactly

Irregular Buses

Average 10 min (Poisson)

Bus arrivals
You arrive
Your wait time
Your Arrivals
0
Average Wait
0 min
Expected (Theory)
5 min
Bus Interval Avg
10 min

The Paradox Revealed

Regular Buses

Interval: 10 min

Expected wait: 5 min

vs

Irregular Buses

Average interval: 10 min

Expected wait: 10 min!

Same average frequency, but DOUBLE the wait time with irregular arrivals!

Why Does This Happen?

Length-Biased Sampling

When you arrive at a random time, you're more likely to land in a long gap than a short one—simply because long gaps take up more time!

Imagine buses sometimes come 5 minutes apart and sometimes 15 minutes apart (averaging 10 min). The 15-minute gaps are three times longer, so you're three times more likely to arrive during one of them!

The Mathematics

For regular buses with interval T, your expected wait is simply:

E[wait] = T/2 = 10/2 = 5 minutes

For irregular buses (Poisson process) with mean interval T:

E[wait] = T = 10 minutes

The irregular case has double the expected wait! This is because the variance in arrival times creates longer gaps that "trap" random arrivals.

Related Paradoxes

The inspection paradox appears everywhere in length-biased sampling:

  • Class Size: Average class has 30 students, but average student experiences 50+ classmates
  • Friendship Paradox: Your friends have more friends than you (on average)
  • Hospital Stays: Average stay is 3 days, but a random check finds patients staying 7+ days
  • Traffic: Asking drivers "how heavy is traffic?" oversamples congested times

Real-World Implications

This paradox explains why public transit often feels worse than statistics suggest. If buses are supposed to come every 10 minutes on average, passengers experience much longer waits during peak variance times.

Solution: Regularity matters more than frequency! A bus every 12 minutes like clockwork beats one that averages 10 minutes but varies from 5 to 20.

The General Principle

Whenever you sample something proportional to its size/duration, you oversample the large ones. This is called the inspection paradox or size-biased sampling.

E[observed interval] = E[interval] + Var[interval]/E[interval]

The higher the variance, the worse your experience compared to the average!