American Emperor's Children's shoes, do you know how to scientific taxi?

A few months ago, What's The Fare launched a mobile client that could provide specific pricing for taxis and various call-carts, helping users make decisions. By analyzing some of the data from mid-September to early October in San Francisco, New York and Las Vegas, What's Fare also published an interesting research report on the blog, which we consider as a guide to a scientific taxi in the US.

According to the data summarized by what's Fare, Uber, Lyft and Sidecar are cheaper than taxis in most cases, but using these call services during the rush hour may run into a dynamic pricing mechanism that you need to think about when it comes to carpool or taxi.

Is it really cheaper to call a car than a taxi?

(1) Carpool

When the traffic volume is large, the dynamic pricing, the carpool service is more affordable than the taxi to discuss the score. In most cases, even with dynamic pricing, San Francisco and Los Angeles have a carpool service that is cheaper than taxis for almost all of the time. In New York, at least one carpool service is cheaper than a taxi in 85% of the time.

(2) Personal car

It becomes more interesting when comparing personal services to taxis. Although collective carpool is almost always cheaper than taxis, there are many variables that can be used for personal rides. The figure below shows the proportion of time each service is less expensive than a local taxi.

In Los Angeles, any carpool service is cheaper than a taxi in 95% of the time. But in San Francisco and New York, UberX is only cheaper than taxis in three-fourths of the time, and Lyft is 72% less expensive in 64% of the time. Sidecar is not currently available in New York.

Dynamic pricing

Passenger peak dynamic pricing, this is to improve the driver on the road or deceive consumers? This is really hard to say. Let's see how the dynamic pricing affects consumption.

(1) Dynamic pricing during rush hour

It's not surprising that there is more demand in the rush hour, but how often do Uber and Lyft use the passenger and prime time prices during this period? The illustration below is the weekday dynamic pricing per hour ratio.

From the above, San Francisco's customers are more likely to experience dynamic pricing during their commute time. New York and Los Angeles are going to be less likely.

(2) Frequency of dynamic pricing

In addition to frequency, another important data is the dynamic pricing multiplier. What is the relationship between the coefficient of dynamic pricing and the ride rate? The following picture is at a glance. In short, the deeper the color, the greater the multiplication of the box, the higher the number of rides.

Although each vehicle software says, dynamic pricing does not often occur, but from the graph, the frequency is quite high. Weekend San Francisco is not easy to see dynamic pricing, Lyft dynamic pricing will be higher than UberX. (Note: Uber and Lyft use different criteria to specify dynamic prices.) ）

(3) Trend similarity of dynamic pricing

When users are experiencing dynamic pricing, they generally look at other applications called cars. is the dynamic pricing coefficient of UberX and Lyft similar on the same route at the same time? In the chart below, the horizontal axis is a multiple of the UberX, while the vertical axis is the proportion of the Lyft increase, and the size of the bubble represents the relative number that appears.

We can see from the diagram that the two are not so consistent. In fact the correlation coefficient in San Francisco is only 0.44, New York and Los Angeles are lower, respectively in 0.31 and 0.38. This may indicate that dynamic pricing is now more useful for a single ride than for all the customers needed. Because dynamic pricing doesn't have a strong correlation, if you want to save money, you know, look at multiple software to save money.

What's the strong call for a car?

So the question is, which service should I use?

In fact, this is expected to listen to Marx's grandfather, truth-seeking, concrete analysis of specific problems.

Of course, if you have some specific needs, the following content is still somewhat referential.

(1) Low price

What ' s The Fare also paints the proportion of the cheapest time for each service, and the result is that who is the cheapest is really hard to say.

In San Francisco, the proportion is also more balanced, sidecar and UberX are common cheapest, the ratio can reach 1/3, followed by Lyft, about 19%. In New York and Los Angeles, Lyft is more likely to be cheaper.

(2) Loyal users

Because there is no clear winner in any market, they also calculate the "price of loyalty," If only one service?

In San Francisco and Los Angeles, if you are loyal to a single service, Sidecar is definitely your best choice. In San Francisco or Los Angeles, loyal users of UberX or Lyft may pay 14% to 43% more. In these two cities, taxis are not the best choice, even if they are frequently priced dynamically.

In New York and Manhattan, UberX and Lyft are pretty close, and taxis are cheap only in a few dynamic pricing scenarios. However, the result is that if you do not bother to change, or taxi is the most convenient, always take a taxi seems to spend only 6% more, not to carpool.

(3) Waiting time

Waiting time is as important as price. Overall, San Francisco had the shortest waiting time, and UberX and Lyft were almost no different. 5 minutes to get on the bus the probability reached 85%.