Originally Posted by
terrymorse
With a sample size of just 3 titanium frames, the margin of error is kind of big.
If the actual Ti failure rate is 10%, the chance of getting at least 1 failure is:
1 - (0.9 * 0.9 * 0.9) = 27%
Statistics rambling aside, point taken about the way the material is used being the most important factor.
Yep. The other way to state that: there is a 73% chance you will not see any failures with 3 titanium bikes, if the known failure rate for titanium bikes is 10%. In fact, it is still likely (51% chance) that you will not still not see any failures with 3 titanium bikes, even if the known failure rate for titanium bikes is 20%.