I had read about BigQuery but hadn't had a chance to use the external product yet. This week I came across some article about Uber and taxis etc, and it had a link to recently released data by The NYC Taxi and Limousine Commission. In addition to directly downloading the data, or looking at it through NYC OpenData, you can also play around with it using BigQuery.
So here's a sample I threw together:
SELECT trip_distance, fare_amount, tip_amount, tolls_amount, round(100.0 * tip_amount / fare_amount) as tip_ratio FROM [nyc-tlc:yellow.trips_2015] WHERE fare_amount > 50 AND trip_distance > 15 AND passenger_count > 8 ORDER BY trip_distance DESC
The screenshot shows the cached results, it took about 6 seconds to run (which is not bad, considering the dataset is about 10 Gigs !!)
(9 people went 27 miles, paid $95 and gave no tip!)