4 ms·

One car isn't enough to form a true conclusion, but what about 76? [1] This poll is biased in the sense that "0" isn't an option; still, it's an alarming amount

One car isn't enough to form a true conclusion, but what about 76? [1] This poll is biased in the sense that "0" isn't an option; still, it's an alarming amount of failed drive units. If this is a real problem, it won't scale---they can't keep replacing these things under warranty. [1] http://www.teslamotorsclub.com/showthread.php/29834-Drive-Unit-Replacement-Poll http://www.teslamotorsclub.com/showthread.php/29834-Drive-Un...

robterrell 12y ago
Don't forget about the Toyota RAV4 EV, which uses the same Tesla drive unit. There are lots of reports of similar problems on the RAV4 EV forums. Tesla is a public company, so this cost must make its way into an annual report, right?
puzzlingcaptcha 12y ago
There is also a sampling bias, people who have experienced problems with the car are more likely to hang around in online forums trying to share their experience.
- jarek 12y ago
  So the rate the problem is being reported at should be comparable to other premium cars? How does it compare with BMW, an enthusiast brand with high online engagement, many unofficial forums and so on? Audi?
awalton 12y ago
1: The people who are most likely to bitch about it are also the people that congregate online. 2: Troll factor - people who just want to see the company fail. 3: Tesla's own professional paranoia. It's cheaper and easier to replace a faulty unit now to inspect it and determine how to make a future unit more resilient than it is to deal with bad press in the future. Then you have some people who baby their cars, and then you have the people who drive cars professionally who are looking for ways they can break things. 76 is not a signal when the company is making 700 cars a week, it's well below the noise threshold. So there's enough points to counter. The simple truth is, anyone who's trying to come to a conclusion on this car this early is stupid. The fact that this test was called "Long-Term" should tell you something about the industry.
Systemic 12y ago
The survey isn't just "biased" in that 0 isn't an option, it represents a fundamentally different question. The results of this survey can be phrased as: "Out of 76 cars which had their drive units replaced, how many drive units did they have replaced?" From this poll you can't even say you have a sample size of 76 cars as you only have a sample of 76 cars whose drive units have broken. A different survey on the same website includes zero [1], and is more appropriate to answering the question of "How likely is it that a Model S will have its drive unit replaced?" [1] http://www.teslamotorsclub.com/showthread.php/32608-Drive-Unit-Problem-Fixes http://www.teslamotorsclub.com/showthread.php/32608-Drive-Un...

robterrell 12y ago

Don't forget about the Toyota RAV4 EV, which uses the same Tesla drive unit. There are lots of reports of similar problems on the RAV4 EV forums. Tesla is a public company, so this cost must make its way into an annual report, right?

puzzlingcaptcha 12y ago

There is also a sampling bias, people who have experienced problems with the car are more likely to hang around in online forums trying to share their experience.

jarek 12y ago
So the rate the problem is being reported at should be comparable to other premium cars? How does it compare with BMW, an enthusiast brand with high online engagement, many unofficial forums and so on? Audi?

awalton 12y ago

1: The people who are most likely to bitch about it are also the people that congregate online. 2: Troll factor - people who just want to see the company fail. 3: Tesla's own professional paranoia. It's cheaper and easier to replace a faulty unit now to inspect it and determine how to make a future unit more resilient than it is to deal with bad press in the future. Then you have some people who baby their cars, and then you have the people who drive cars professionally who are looking for ways they can break things. 76 is not a signal when the company is making 700 cars a week, it's well below the noise threshold. So there's enough points to counter. The simple truth is, anyone who's trying to come to a conclusion on this car this early is stupid. The fact that this test was called "Long-Term" should tell you something about the industry.

Systemic 12y ago

The survey isn't just "biased" in that 0 isn't an option, it represents a fundamentally different question. The results of this survey can be phrased as: "Out of 76 cars which had their drive units replaced, how many drive units did they have replaced?" From this poll you can't even say you have a sample size of 76 cars as you only have a sample of 76 cars whose drive units have broken. A different survey on the same website includes zero [1], and is more appropriate to answering the question of "How likely is it that a Model S will have its drive unit replaced?" [1] http://www.teslamotorsclub.com/showthread.php/32608-Drive-Unit-Problem-Fixes http://www.teslamotorsclub.com/showthread.php/32608-Drive-Un...