Recall the bet first: It’s between Berkshire Hathaway (BRK.B) CEO Warren Buffett and a money management firm, Protégé Partners. Buffett is betting (with his own money, not Berkshire’s) that the stock market performance of a big index fund can beat the performance—averaged—of five funds of hedge funds carefully chosen by Protégé.

And that positive news about the hedge fund side? The five funds of funds won in 2015, beating the index fund for only the second year out of eight.

Even so, the winning margin was slight. The five funds, on the average, gained 1.7% during 2015 vs. the 1.36% return rung up by Buffett’s index fund.

And then there’s what really counts: This bet is only two Decembers from being over, and Buffett has a commanding lead. For the first eight years of the bet, the index fund he’s backing is up 65.67% against 21.87%, on the average, for the five funds of funds.

That makes Buffett appear unbeatable. And yet…

Hold that thought while we recap a few more details about the bet. It was initiated by Protégé after the firm learned that Buffett had thrown out a challenge to the hedge fund world in a speech. Protégé also set one important rule of the game: It chose not to disclose the names of the five funds of funds participating in the bet. (One of the five, however, has always been assumed to be a fund of funds run by Protégé itself).

Buffett’s contender in the race has a less complicated pedigree: It is an S&P 500 fund run by Vanguard: the Vanguard 500 Index Fund Admiral Shares.

The money at stake will go to charity and is slated to be at least $1 million (the “at least” being an evolving story we’ll get to in a minute). Add a detail also about Fortune: In a 2008 article, it disclosed the existence of the bet and has since exclusively reported its annual results.

Now, about Buffett’s big lead in the bet: Whatever the probabilities, his advantage does not guarantee him a win. A year ago at this time, the man who engineered the competition for Protégé, Ted Seides, sketched out the only scenario that he believed could change the complexion of the bet. “The odds now,” he said, “are that we’ll need to see a severe market contraction for our side of the ledger to state an epic comeback.” Seides also undercut that thought about a contraction almost before he stated it, saying: “No one wins when that occurs.”

Still, “a severe market contraction”—or at least the beginning of one—sounds an awfully lot like what we’ve been seeing in the first six trading weeks of 2016. Through last Friday, notwithstanding a strong stock market rally on Friday itself, the Admiral shares were down for the year by 8.52%.

We do not know, of course, how Protégé’s five funds have fared in those six weeks. But we absolutely know that in a bear market, most hedge funds can—and do—turn to options that are not open to a straightforward index fund, such as Admiral. Most especially, the hedge funds short stocks or flee them to buy other types of securities.

In the first year of the bet, the torture chamber of 2008, Protégé’s five hedge funds of funds indeed profited from some of these strategies and walloped Buffett’s Admiral shares. Even so, the funds, on the average, got the kind of no-win victory that Seides was talking about: They simply had a lesser loss for the year—23.9%–than did the Admiral fund, which plunged by 37%.

After that, for six straight years until 2015 came along, Admiral whipped the funds of funds.

Whatever the results from here on in—even, indeed, in the face of a continuing contraction—one of two charities is going to get at least $1 million. If Buffett wins the bet, the money goes to Girls Inc. of Omaha. If the Protégé camp instead makes an epic comeback, Absolute Return for Kids gets the haul.

That “at least” qualifier reflects a 2012 change in the bet. Originally, the stake was simply $1 million, an amount that Protégé and Buffett guaranteed by each putting up $320,000 in cash that was invested in zero-coupon bonds structured to ascend gradually over the bet’s 10 years to a value of $1 million. But falling interest rates quickly changed the plot, precipitously vaulting the bonds’ value to nearly $1 million way ahead of schedule.

At that point, in late 2012, Seides and Buffett reworked the bet, agreeing that they would sell those fully-priced bonds and put the proceeds into Berkshire Hathaway B stock. Because the stock could obviously fall, Buffett guaranteed the winning charity a minimum of $1 million. But if the bet’s 11,200 shares of Berkshire B are worth more than that at the end of 2017, the winning charity walks off with their full value. As of last Friday, the shares were valued at better than $1.4 million.

To that change in the bet, add a couple of developments concerning its Protégé instigator, Ted Seides. He resigned as a Protégé partner in 2015; wrote a book, So You Want To Start A Hedge Fund, to be published this month; and is now thinking about how he will next connect to the money management business. Seides’ new book says little about the bet, except that it reprints portions of his 2007 opening-shot letter to Buffett. Seides, now 45, wrote enthusiastically in the letter about the funds of funds he proposed to line up for the bet, describing each as trained “in the discipline of value investing with a long time horizon.”

In the bet’s start-up days—as Fortune’s 2008 article reported—Protégé calculated its chances of winning as about 85%. Buffett was less confident: He gave himself a 60% chance, a slimmer edge than he would normally like to have.

So now we know once again that the stock market is a tough baby to predict.

Carol Loomis, the author of this article, is a retired senior editor-at-large of Fortune. She is also a long-time friend of Warren Buffett and a shareholder of Berkshire Hathaway.