Hi Bob,

As I've remarked on a couple of threads over on Bogleheads I actually think it's surprising that they of all people are so quick to accept Norstad et al.'s critique of time diversification.  Best I can tell it's a /really/ complicated question of both math and model specification, but I myself actually agree with Siegel, Ibbotson and the pro-time diversification crowd on this one (see the first section and Fig. 1 of this paper):

http://www.ifebp.org/pdf/webexclusive/07mar.pdf

I do try to keep an open mind about it, of course, and most recently I very directly asked David Grabiner if he'd weigh in on the case for modeling cumulative one-year returns rather than multi-year returns, but as yet neither he nor others have done so . . . . Note that if Norstad et al. are right and Siegel/Ibbotson are wrong, then (at least given certain model assumptions) the entire idea of Lifecycle/Target Retirement funds with decreasing allocations to stocks as one ages is bunk, and to me that's a pretty big claim  . . .

All best,

Pete