Clearly, diversification can protect a portfolio. T. Rowe Price analyzed the performance of different asset allocations from Dec. 31, 1954, to Dec. 31, 2014, and found that while an all-equity portfolio delivered the highest return for its best year, and the highest average annual nominal return, at 10.4 percent, it also delivered the worst return for its worst year.
Between Dec. 31, 1985, and Dec. 31, 2014, T. Rowe Price found that a diversified portfolio invested 60 percent in equities, 30 percent in bonds and 10 percent in cash would have delivered 91 percent of the returns generated by 100 percent stock exposure, with about 83 percent of the volatility.
Optimal asset mixes also change as people age. Stuart Ritter, a senior financial planner at T. Rowe Price, frames that concept around inflation.
Read MoreWhy catch-up retirement contributions don't work for many savers
"The reason you hold stocks or don't hold stocks is that you are balancing short-term market risk against inflation risk," he said. "If I am using my money in two years, inflation isn't going to make that much difference, but market volatility would be a huge risk," so less equity exposure would make sense. "If I'm not using that money for half a century, market volatility doesn't feel good but I need to manage that inflation risk, and that means holding equities."
For that reason, Ritter is especially perturbed by the 9.9 percent of 20-somethings in the EBRI data who reported no equity exposure in their 401(k)s. "You've got an age group that has literally half a century before they use this money. The biggest threat is the increase in asset prices due to inflation," he said. (Tweet This)
Ritter has analyzed the performance of the S&P 500 index relative to inflation, looking at 15-year periods beginning each month from January 1926-December 1940 to January 2000-December 2014, and he found that the index beat inflation 95 percent of the time, by an average of 8 percentage points.
"When your goal is to keep up with inflation, historically, stocks have been the best asset class to do that," he said.
Read MoreAsset allocation key in retirement plans
To be sure, the EBRI data on 401(k)s only illustrates what people do with one portion of their savings. And while there are similar analyses like that conducted by the Federal Reserve, which examine asset allocation within other savings vehicles such as IRA accounts—many of which also show significant levels of extreme asset allocation—without more comprehensive data on those households it is hard to know just how many people actually have potentially problematic extreme asset allocations.
"It is hard to evaluate the overall balance sheet" of a household, said Sarah Holden, senior director of retirement and investor research at the Investment Company Institute, which has also examined retirement asset allocation. She added that when the institute conducts household surveys, people indicate that they do intend to rebalance their portfolios as they approach retirement.
Jack VanDerhei, research director at the EBRI, agreed that for many people a look at 401(k) allocations alone provides an incomplete picture. But 20-somethings are less likely to have other investment assets, he said, so the 9.9 percent of that cohort who reported having no equity allocation may be a concern.
Some of those people may have seen parents or a family member have an "unfortunate situation" in 2008 or 2009 when the stock market crashed. "There are always going to be some people who, unless you auto-enroll them, are going to be too scared to pull the trigger and go into equities on their own," he added.
Read MoreTime to rebalance your retirement portfolio?