Originally Posted by
davie355
Is it just me or has this correlation strengthened?
For years I've been accustomed to a weak relationship between a driver's star rating and their actual quality. Of course Uber removes poorly rated drivers so this weak correlation shouldn't be a surprise. Lately, however, I've noticed, riding UberX primarily in San Francisco and New York City, a near-deterministic mapping:
- 4.95 stars and up: smooth ride in a new, clean, comfortable car with a very friendly driver
- 4.90-4.94 stars: friendly driver in a car that is inferior (e.g., smaller or older) but still meets UberX standards
- 4.80-4.89 stars: friendly enough driver in a good enough car
- 4.70-4.79 stars: near certainty of a flaw that significantly affects comfort on the ride:
- aggressive driving
- loud music
- extreme air conditioning
- strong odor
- rolling down the window
Are you using a minimum number of trips for the drivers? Otherwise, seems hard to believe one can discern between a 4.96 and a 4.94.