Some shoe brand (Zeezys) boasts having technology to make the wearer faster, and your friend is planning to test their claims. She buys 18 pairs of the Zeezys and 18 pairs of Nikes, and recruits her...

Some shoe brand (Zeezys) boasts having technology to make the wearer faster, and your friend is planning to test their claims. She buys 18 pairs of the Zeezys and 18 pairs of Nikes, and recruits her 36 friends who are all identical twin pairs. She assumes that the twins are comparable in activeness because she doesn’t want anyone to get tired running more than once. Each twin wears one of the shoe brands and are asked to run 4km. She records their speed and compares whether Zeezys really do make the runner faster. What statistical test should she use?

Paired samples t-test OR Independent samples t-test?