Dueling op ed letters

Select Board Chair Joe Michaud had an op ed letter in the Standard Times where he rebutted an earlier letter from Canadian wind turbine opponent, William Palmer. Michaud noted that even using Mr. Palmer's data, the likelihood of a failure of a wind turbine is remote,

"Using the same probability factors that Palmer uses in opposition to his country's project, we could expect our two turbines to have a catastrophic event once every 323 years."

Mr. Palmer's letter can be found at this link. Mr. Palmer feels the Select Board did not properly assess the detriments to wind power installations. In fact, the Select Board was aware of comments that Mr. Palmer made to an Ontario wind farm application ...

... through a submittal to the public hearing on the permit. You'll note that Mr. Michaud knew the data from Palmer's comment even though it was not included in his op ed letter. The Ontario authorities rejected Mr. Palmer's assessment in the Ontario installation. Mr. Palmer has been advanced as an expert by opponents of the Dartmouth wind project. While a professional engineer, Mr. Palmer's area of expertise is nuclear power, not wind power.
You can read the Ontario comments from Palmer here I will accept Mr. Palmer's calculation regarding the frequency of wind turbine failures. However, his analysis of the likelihood of failure in multiple turbines is flawed. The events are independent. That is an important distinction. Multiple installations do not increase the likelihood of an individual turbine failure. Palmer takes the low probablility of an individual failure (1 in 775 years) and divides it by the number of turbines. That is not statistically valid.
Here is an example. If the probability of a tire blowout for one tire is 1 in 100,000 miles, then what is the probability for your car? By his reckoning it is 100,000 miles divided by 4 since you have four tires or 1 every 25,000 miles. If you have two cars, by his method, the likelihood is 100,000 divided by 8 or 12,500 miles. Do you see the error? The probability remains 1 in 100,000 miles for each tire. It is not lessened by the number of tires. That is because the events in question are independent. Another example would be two fair coins, Fair meaning an equal chance of heads or tails. You flip one and get heads, what is the probability of getting heads when the second coin is flipped? It is 1 in 2. The same as the first coin. The events are independent.