Sunday, June 22, 2014


Earlier today, on Fox News Sunday, George Will listed six reasons we need a special prosecutor for the burgeoning IRS scandal:

CHRIS WALLACE: Well, usually, mild mannered Congressman Paul Ryan ripping into the IRS chief this week over claims the agency lost thousands of e-mails connected with the IRS targeting scandal. And we're back now with the panel to continue the conversation. The IRS commissioner says back in 2011, when Lois Lerner's hard drive crashed, that she immediately called the I.T. people, asked them to retrieve any of the files that they could. George, how credible?

GEORGE WILL, SYNDICATED COLUMNIST: Incredible. And here's -- let me tell you why we need a special prosecutor in this. One of my pet peeves with current English usage is the treatment of disinterested and uninterested as synonyms. We wish the Justice Department were interested in this. It were a disinterested investigator of this. Instead, it's uninterested in the investigation.

We can no more expect Mr. Holder to investigate this White House than we could have expected John Mitchell to investigate the Nixon White House.

Here's -- we know six things, Chris. We know first the targeting occurred.

Therefore, second, we know that this is worse than article two of the Nixon impeachment count, which said Nixon endeavored to use the IRS. The IRS back then resisted.

Third, we know that this became public in an act of deceit when Lois Lerner planted a question with a friend in an audience to try and get this out on her own terms.

Fourth, we know that she has taken the Fifth Amendment because she has a right to do this when she has a reasonable suspicion that there might be criminal activity involved.

Fifth, we know that from the timeline you put up today, that there has been 13 months of stonewalling on this.

And sixth, now we know that not only her hard drive, but six other people intimately involved in this suddenly crashed in an amazing miraculous coincidence. Religions have been founded on less, ten days after the investigation started.

That's why we need a special prosecutor.

It's gotten so bad even Woodward and Bernstein are ripping old media and the corrupt, hard left Democrat Party.

There's a Pulitzer and a lifetime of riches for a journalist willing to go to the mat on this, but it's not altogether clear whether such a breed still exists in America.

Hat tip: BadBlue News.


Anonymous said...

I run a data center. Disk drives that are left running continuously last between two and three years. Three years is about 36 months.

The odds of a disk failing in any given month are roughly one in 36. The odds of two different drives failing in the same month are roughly one in 36 squared, or 1 in about 1,300. The odds of three drives failing in the same month is 36 cubed or 1 in 46,656. The odds of seven different drives failing in the same month is 37 to the 7th power = 1 in 78,664,164,096.

Of course this is very simplified because disk failure modes are more at end-of-service-life rather than linearly spread over median life. So what if I am off by a factor of 4X? This crude calculation gets us into the same astronomical ballpark. You could insure against this event happening by buying lottery tickets.


Anonymous said...

Great, George.

Just where have YOU been?

Donald Sensing said...

I am not a mathematician, but ISTM that Anonymous, the data-center operator, has made a statistical error in treating the HDs failures, or not, as related events when they are independent events.

If the expected life span of an HD is 36 months, and for simplicity ignoring that failures occur nearer the end than the beginning, then each HD has a 1/36 chance of failing in any given month - regardless of what the HD one office away does. So the 1/36 odds never change.

The incredulity therefore is not over failing HDs per se, but that the exact same people for whom the committee wants to read their emails are the ones whose HDs failed.

As Yogi berra said in a different context, "It's too coincidental to be a coincidence."

To calculate the odds of that we have to place those HDs into the universe of possibles, which would be the total number of like workstations in the entire IRS.

Googling tells me that the IRS has 89,500 employees. Not all have email, of course, but let's be very generous and say only 50,000 do. That 50K HDs, each with a 1/36 chance of failure in any given month.

That means that in any given month, 1/36 of the 50K drives will fail, or 1,388.89 drives each month.

But - and please do check my math - that means that the chances of specifically Lerner's HD failing in that particular month is 1/1,388. And the same odds for each of the other six drives.

That's where you start multiplying the odds together. Excel tells me that 1/1388 is 0.00072, or 0.072 percent chance. Now we calculate the odds of all seven specific HDs failing, which is .072 pc X .072 ... seven times.

And that makes the final odds 0.0000000000000000000000000000000000000000000000000000000000000001 percent.

Expressed in notation, it is 1.01E-66. (I'm letting Excel calculate all this.)

Top what may we compare this? Well, how about the number of stars in the entire universe? According to,

Kornreich used a very rough estimate of 10 trillion galaxies in the universe. Multiplying that by the Milky Way's estimated 100 billion stars results in a large number indeed: 100 octillion stars, or 100,000,000,000,000,000,000,000,000,000 stars, or a "1" with 29 zeros after it. Kornreich emphasized that number is likely a gross underestimation, as more detailed looks at the universe will show even more galaxies.

I have never seen an estimate of 10 trillion galaxies before, the top number I have ever seen in "only" 500 billion. But let's leave it at 10 T:

Number stars in 10 trillion galaxies: 1.00E+29

Odds of those seven particular IRS HDs failing the same month:

Please note that according to, there are about 1.0E+80 atoms in the entire universe.

So the odds against those seven identified HDs failing at the same time is sensibly comparable to the inverse of the number of atoms in the entire universe.

Again, I would welcome math checking!

Benson II said...

Because most so called Journalist in the MSM are leftist and draw praise for being that and paid well besides there is no incentive or need for further praise or money to investigate. The thought of doing the right thing doesn't occur to them because the lawlessness and lies of Obama and Democrats is something they agree with and to them is the right thing. The end justifies the means might as well be tattooed on all of their foreheads.

Andrew_M_Garland said...

To Donald Sensing,

Alas, your math is bad. 1389 IRS hard drives fail each month out of the proposed 50,000 drives. Lerner has an IRS HD, so its chance of failure is 1389 / 50,000, which is 1/36, back to the start. Consider that her HD chance of failure is 1/36 each month and nothing changes that.

-- --
To theBuckWheat,

Your math is good and I would state it a bit differently. Say I choose 7 particular HD's. The odds that they all fail in the next month is 1 in 36^7, as you say about 1 in 78 million, or about once in 6.5 million years.

It is even less probable than that. Sometime within the prior 3 years most of those drives will have failed and been replaced at various times. Some will always be fresh, dramatically cutting the odds that they will all fail in the same month.

The best evidence against the IRS is that they live with this reality of HD failures. They could not function without reliable backup going back years, not merely 6 months. Their claimed backup system is ludicrous and unworkable. If it existed as claimed, then it had to be an intentional desire to lose data when needed and claim plausible deniability.

I doubt that such poor backups apply to the typical IRS worker. We will find that they apply only to the management of politically active groups such as Lerner's.

The standard for criminal conviction is commission of a crime beyond a reasonable doubt.
The data disappeared. I say that less than one chance in 6.5 million years of this being an accident is beyond a reasonable doubt. Jail them all.

Anonymous said...

Man, those numbers are so mind boggling its hard to believe. And well, ya, don’t believe them because the methodology is all wrong. Here is why:

First, this argument is specious because we don’t know anything about the condition of the hard drives/computers at time of failure. Hard drives follow, like most things, a bathtub failure rate (see the link below). So without statistical information on how often the IRS replaces their hardware, we can’t even bother guessing what an average failure rate would be. If they replace them every year, their failure rate would be EXTREMELY low. However, they have admitted that they don’t back up more than 6 months of email on tape and each employee is only allowed 500mb storage on the server (due to prohibitive cost), so with 89,500 employees, my guess is they use some pretty antiquated junk.

But whatever, that doesn’t really matter since the methodology here is completely wrong as well. What we are looking at is not a simple Bernoulli random variable problem (in which we find the probability of 7 out of 7 failures each of which is independent and identically distributed, which is what the author of this did) because it involves finding the probability of a number of failures from a specific population or sample size. Instead, it can be looked at as either a binomial probability mass function or a hypergeometric experiment.

A binomial probability mass function finds the probability of a given number of successes (in this case failure of a hard drive) given a population size, and probability of success. The hypergeometic experiment looks at a sample drawn from a population of known size, i.e. total number of personal computers in IRS custody (maybe around 50,000), sample size, 83, number of successes in the population (50,000*1/36) and the number of successes, 6. How I got these numbers is explained below, however, as the population size gets big, the function asymptotically approaches that of the binomial probability so we can simply use the binomial.

The commission targeted 83 individuals, not including Lerner, so the population size is 83. We are looking for the probability of getting 6 or more failures from those 83 and, for the sake of argument, we use 1/36 as the failure rate (which as indicated above is probably wrong). I won’t show the equation, you can look that up on google. Plugging into Excel “=1 – BINOM.DIST( 5, 83, 1/36, TRUE)” you get 2.82%. This is the probability, given the afore mentioned conditions, that out of 83 computers, at least 6 would have suffered a hard drive crash in the same month.

It have also had trouble finding in any of the reporting that the IRS has claimed that all 7 failed in the same month. Given that the time frame in question is somewhere between 2011-2013, there is a good 2 years in which these computers could have crashed, losing the questionable data. In that case, the odds go way up. Maybe I just can’t find the correct article, a link would be appreciated.

I’m not saying that it isn’t suspicious, but that 2.82% probability is a whole lot different than 1 in 78,664,164,096.