GoFaster
Moderator at Large
It's fair to say that there have been a lot more actual failures than have been reported to NHTSA, rightly or wrongly.
About 6% of the people who have reported a HPFP failure on this forum, have subsequently had a second failure. This number is probably higher than the total failure rate due to the possibility of installation errors, etc.
Anyway, if the actual failure rate is as NHTSA quotes - call it 1 in 1000. If the probability of any single vehicle having a failure is 1 in 1000, then the probability of two specific independent vehicles both having a failure is 1 in 1,000,000. With about 139,000 sold - and most of those being one-TDI families - the probability of this ever happening, if that is really the failure rate, is very unlikely but not impossible.
How many of those 139,000 TDI cars belong to two-TDI families - no doubt there are a few - but not many.
If the failure rate is taken as 1% then it's 1 in 10,000 odds that two specific vehicles would both have a failure. If 10,000 of those 139,000 cars sold go to two-TDI families then it's likely that this would happen once (to someone). I suspect that's a high estimate of two-TDI families but not implausibly high.
I suspect based on the number of repeat failures and on the existence of a situation in which both cars in a single family had a failure that the real failure rate is probably on the order of 2% - 3%.
About 6% of the people who have reported a HPFP failure on this forum, have subsequently had a second failure. This number is probably higher than the total failure rate due to the possibility of installation errors, etc.
Anyway, if the actual failure rate is as NHTSA quotes - call it 1 in 1000. If the probability of any single vehicle having a failure is 1 in 1000, then the probability of two specific independent vehicles both having a failure is 1 in 1,000,000. With about 139,000 sold - and most of those being one-TDI families - the probability of this ever happening, if that is really the failure rate, is very unlikely but not impossible.
How many of those 139,000 TDI cars belong to two-TDI families - no doubt there are a few - but not many.
If the failure rate is taken as 1% then it's 1 in 10,000 odds that two specific vehicles would both have a failure. If 10,000 of those 139,000 cars sold go to two-TDI families then it's likely that this would happen once (to someone). I suspect that's a high estimate of two-TDI families but not implausibly high.
I suspect based on the number of repeat failures and on the existence of a situation in which both cars in a single family had a failure that the real failure rate is probably on the order of 2% - 3%.