This column is called Critical Thinking because people just don’t do enough of it these days. Eyeball-catching headlines and controversial sound bites get posted and tweeted a thousand times before anyone stops to consider the validity of the content.

Case in point: Left-leaning think tank Institute for Policy Studies released a report Wednesday concluding that 38% of the most “lavishly compensated CEOs” of the past 20 years were poor performers that ended up getting fired, paying massive settlements or fines for fraud, or led firms that crashed and had to be bailed out by the feds.

Never mind that the headline reads “nearly 40%” but it’s really 38%. What’s a few percent when you’re bashing greedy executives, right?  

Not surprisingly, the news was picked up by Reuters and echoed by everyone from The New York Times and CNBC to Yahoo and the Huffington Post. And none of these fine media outlets spent a few minutes to look into the report and see if its conclusions held water. They don’t. They’re so full of holes it isn’t funny.

Of all the misleading reports I’ve seen over the years – and I’ve seen a lot – this one takes the cake. It’s one of the more blatant examples of manipulating data to fit a politically expedient conclusion I can remember. Here are just a few examples of what a little digging uncovered:

Perhaps the most misleading aspect of the report is the “bailout” category, which includes the majority of the “poor performing” CEOs. Roughly half of the companies listed – including Wells Fargo (WFC), Goldman Sachs (GS), Morgan Stanley (GS), JP Morgan Chase (JPM), State Street (STT), and Bank of New York Mellon (BK) – were and still are healthy banks forced to take bailouts by the federal government.

In reality, the U.S. Treasury Department received the equivalent in stock. In other words, the net to taxpayers is more or less zero. And nearly all the TARP money has been paid back, with dividends and interest.

Just as troubling is the report’s methodology that, to me, completely destroys its credibility. Check out this example:

Since Wells Fargo was a TARP recipient, and its CEOs made the top 25 highest paid CEOs list six times, the report counts all six as incidents of “poor performing” CEOs. In other words, since Wells Fargo was forced to take TARP money in 2008, and Carl Reichardt, its CEO in 1994, made the highest paid list that year, he’s included as a “poor performer.”

Yes, I know that doesn’t make any sense. What can I tell you; that’s how they did it.

Granted, there are a few CEOs on that list – most notably Dick Fuld of Lehman Brothers – that deserve to be there. But the vast majority of the 112 slots that make up 22% of the poor performing CEOs cited by the report were simply high-paid CEOs over the past 20 years whose companies received and paid back TARP funds.

A number of CEOs listed in the “booted” category are pretty suspect, as well. For example, Mark Hurd, who executed a miraculous turnaround of Hewlett-Packard (HPQ), was forced out over a sex scandal that never happened. So the only decent CEO H-P has had in decades somehow made the “poor performer” list. Go figure.  

As for the “busted” category, it’s almost as misleading as the “bailout” section. It includes former Intel (INTC) CEO Craig Barrett – I can’t even begin to imagine what he’s doing on the list. And what former AOL (AOL) chief Steve Case is doing there is anyone’s guess. Granted, he helped engineer one of the worst mergers of all time, but the report says “accounting fraud,” which makes no sense to me.

Also, Oracle’s (ORCL) founding CEO Larry Ellison is listed in the “busted” category over an insider trading charge that he was actually found innocent of. Yes, he did settle a derivatives lawsuit related to the same charge by agreeing to give $100 million to charity on behalf of Oracle, but that was probably just to make the annoyance go away. And since he made the highest paid CEO list 10 times, that’s right, it counts 10 times.

Speaking of Ellison, the report chides him as having “led America’s CEO paypalooza, with more than $1.8 billion in pay over the past 20 years.” What’s funny is that Ellison’s compensation is entirely the result of Oracle’s stellar stock performance. After all, he’s the founding CEO and far and away the company’s largest shareholder. You’d be hard-pressed to find an Oracle investor who doesn’t think Ellison deserves every penny of his fortune.

The study further characterizes some of the CEO pay as taxpayer-subsidized because, under federal tax code, incentive compensation such as stock options is tax deductable by the employer.

The problem with that argument is this. America has one of the highest corporate tax rates in the world with all sorts of deductions. Actually, most CEOs would agree to a much lower tax rate without all the loopholes, as Apple (AAPL) CEO Tim Cook argued before Congress some months ago.

It isn’t corporate America’s fault that the IRS has a 70,000 page tax code full of dumb loopholes.

Look, I happen to think that, in the case of many big-company CEOs, executive pay is out of control and lucrative severance packages destroy accountability. As Peter Drucker argued in a 1984 essay, “This is morally and socially unforgivable,” he said of the executive pay problem, “and we will pay a heavy price for it.”  

Indeed. But the most misleading thing about CEO pay is that it’s a relatively small group of big-company CEOs that grab headlines. If you look past the S&P 500 to the thousands of CEOs of smaller companies, you won’t find the kind of disparity that sensationalist and misleading reports like this one go to town on.

Steve Tobak is a Silicon Valley-based strategy consultant and former senior executive of the technology industry.

Contact Tobak; follow him on Facebook, Twitter, or LinkedIn.