Warren Buffett is called the Oracle of Omaha for a reason. The predictions he makes usually come true. But not always. Recently, Business Insider took a look at 12 predictions Buffett made and how they turned out. From his prognostications about Bitcoin (that it would run into serious trouble) to the Flat Earth Society (that it would flourish), he was right on the money with most of them.

But there were just a few times when his predictions failed to come true, and you can probably learn more from those times he had to admit he was wrong than the more numerous occasions when he was proved right.

Here's a look at a few of Buffett's rare wrong predictions:

1. Berkshire Hathaway's 48.2 percent net-worth growth in 1985 would never be repeated.

In fact, he wrote in that year's letter to shareholders, "It is fitting that the visit of Halley's Comet coincided with this percentage gain: neither will be seen again in my lifetime." Likewise, he said, the company's annual growth in share price of 23.2 percent averaged over the previous 21 years was "another percentage that will not be repeated."

Buffett was wrong on both counts. Berkshire Hathaway's net worth went up by 48.3 percent in 1998. But, as he cautioned investors, that number was somewhat misleading: The increase in book value resulted from the issuance of stock. As for "another percentage that will not be repeated," it's not entirely clear from Buffett's letter to shareholders whether he meant that percentage growth wouldn't ever be repeated over a 21-year period or whether it would never be repeated at all. If he meant the former then he may be right--on average, Berkshire Hathaway has grown by an annualized rate of that's closer to 20 percent than 23 percent since then. But as for an individual year with a 23.2 percent gains or higher in share price? That's happened 11 times since 1985.

The thing about Buffett is that even when he's wrong, he's right. One of his most-quoted instructions is, "Be fearful when others are greedy and greedy when others are fearful." A less pithy way of saying that is that things that are gaining value rapidly will eventually slow down or lose value, and things that are losing value will eventually regain it. So it seems likely that what Buffett was really trying to say was: "We've had a really good run, but don't count on it to continue this way." 

And, as this analysis shows, it didn't. Berkshire Hathaway stock isn't outperforming the market the way it once did, something that Buffett predicted would happen in part because of the company's sheer size.

2. Berkshire Hathaway would never sell Capital Cities/ABC or the Washington Post. 

"We expect to keep permanently our three primary holdings, Capital Cities/ABC, Inc., Geico Corporation, and The Washington Post," Buffett wrote in his 1986 shareholder letter. But "permanently" is a long time, and things can change along the way. 

For many years, Capital Cities was run by the team of Tom Murphy and Daniel Burke, executives who Buffett said were "probably the greatest two-person combination in management that the world has ever seen or maybe ever will see." He believed in this team so much that he helped them acquire ABC in what was then the biggest non-oil acquisition in history.

By 1996, though, Burke had retired and Murphy was turning 70. So when Murphy decided to sell Capital Cities to Disney--a deal that Buffett apparently suggested--Buffett sold his stake in the company to Disney as well, in a stock and cash sale worth $2.5 billion.

As for the Washington Post Company, Buffett sold Berkshire Hathaway's 28-percent stake back to the company, now named Graham Holdings, in 2014. But this was a year after the Graham family had sold the Washington Post newspaper to Jeff Bezos, thus prompting the name change. So the company Buffett originally invested in was no longer the same.

3. Gillette and Coca-Cola will always dominate their industries.

"No sensible observer--not even these companies' most vigorous competitors, assuming they are assessing the matter honestly--questions that Coke and Gillette will dominate their fields worldwide for an investment lifetime," Buffett wrote in his 1996 shareholders later.

Both companies do dominate their fields for the moment, but there are clouds on the horizon. Gillette's market share has dropped from 70 percent in 2010 to 54 percent last year, in part due to the growth of Dollar Shave Club and Harry's. Fifty-four percent of the market still means Gillette is very much the market leader. But if the downward trend continues, that dominance may not last for "an investment lifetime," depending on what Buffett meant by that phrase.

As for Coca-Cola, it's the undisputed market leader with somewhere around 40 percent of the soft drink market, well ahead of rival Pepsi. But the category itself is under siege and even though Buffett has a Coke machine in his office and drinks the stuff on camera every chance he gets, consumers are turning away from sugary (or artificially sweetened) sodas and toward healthier drinks, a trend that's put pressure on Coke's bottom line. 

What will Buffett do about these two holdings in the future? One of his strengths is that he seems to know when to double down on an existing investment and when to get out. So we'll just have to wait and see.