I'd also like to point out that you probably know more about what some good investments are, I've done little (I've made like 4 predictions over my life, all turned out to be true, but they were mostly fairly obvious and I had no money to invest with anyway. Specifically that AIG would be bailed out, that Gold and Silver would go up long term, and two specific instances of what Silver would rise and fall to). I'd just like to share a little piece of math from learning to count cards -
What I'm referring to as EV incoporates the possibility of complete failure. So say that you have a Stock worth 10000 dollars. You estimate that it has...
1 in 5 million: Lose all
1 in 10000: Lose half
1 in 100: Lose 25%
1 in 10: Lose 10%
1 in 5: Gain 20%
1 in 2: Gain 50%
1 in 5.266: Gain 400%.
This would give you an EV of 103.7049% and would be, from what I could tell pretty volatile, because of the possibility of large losses. I'm not going to re-open a stats back to find out how to caculate variance here, but the idea is, to achieve a 2/3 probability of doubling your bankroll before you lose half, you want to bet the advantage in EV (in this case 3/7049%) divided by volatility, which I haven't caculated based on this hypothetical.
So what I'm trying to explain, is that sometimes a lower EV can be a better bet to an invester, if it's less volatile. An easy example, that can have pure math would be the comparison of a baised roulette wheel and a biased blackjack deck.
The formula is: EV/Variance (Variance is an approximation, there is a more precise measure used by the Kelly Criterion)
For the wheel, let's say that their are two numbers that have a combined odds of 17.5 to 1 of coming out. This gives the player an edge of 2.857%, and a variance around 18. The optimal bet is (0.0015873*bankroll.) That's about 15 dollar per 10000 dollar in bank. You will win 42.855 cents per 10000 dollars, on average, over time if you continue to take this bet.
For a bias blackjack deck, let's say that their are enough low cards removed to present an advantage of 1% and a Continuous shuffling machine to keep keep all those cards in the deck. The Volatility in a game of blackjack is about 1.3, so let's just overestimate the variance to 1.3333, so that it's easy to calculate. You should bet .75 percent of you bankroll per hand, which works out to 75 dollars per 10000 in your bankroll. You will win .75 cents per 10000 dollars in the bank, on average every time you take this bet.
As you can see, the second situation is preferable, even with a lower EV. Like you said, you do have to take inflation to account during investments, and maybe that makes it preferable to ignore low volatility investments. I was just trying to explain what I meant when I said that there probably is a lower EV than you think, due to the small chances of you being catastrophically wrong, and that that would add to variance, as well.
If you're not a Kelly Investor, this is of less significant to you, but I'd advise at least some sort of investing that's proportional to you bankroll. A lot of people do question the validity of Kelly betting or investing, especially since investing is based off of people's approximations and sometimes, emotions.