At full Kelly, anything else is overbetting 
Yes it’s possible to create returns out of negative expected return assets but … you could make life easier and just invest in positive yield assets.
And at a portfolio level, assets with yields below inflation act as a performance drain.
In the last 15 years, we didn’t experience a propper market crash. Most people over-estimate their risk capacity and tolerance. This will change once Shares crash…
Unless you are about to retire - not sure how money in your second pillar will keep you from panick-selling your shares when they crash; or better said how it will give you the cash you need to pay your bills as you lost your job… Money you can’t access today should not be considered in your asset allocation.
So what are the fixed income / low volatility investments you’d use to stabilize your portfolio?
If you just don’t care being down 30-40% know that its not the case for most people.
Or performance stabilizer 
Bonds proved not to be much of a stabilizer at all in recent years but the picture is changing
In the classic model in a recession share prices go down and interest rates get cut, which means bond prices go up.
In recent years this model was broken because rates were zero which inflated the price of all assets including both shares and bonds
If I was based in UK or US I would be looking at bonds now. The interest rate is far above the long term inflation target and has room to come down.
We arguably did experience one in the last four years.
As pointed toward by @nabalzbhf and @Barto, when VIAC was using cash (only) instead of bonds was when safe CHF bonds were emitted with negative built in returns and we actually had free lunch at the banks’ expenses with non-negative to slightly positive bank accounts (which was the case for VIAC).
Now that that anomally is behind and bonds have some (nominal) yield again, they are getting again into VIAC’s portfolios.
For retail swiss investors, negative returns bonds were a hard sell when other fixed income vehicles were available with higher returns while being arguably as safe (insured by esisuisse). The competition was (and is) between bonds and other fixed income vehicles of a similar duration.
I think that we soon will experience a decent market crash 
My money is on this coming market crash having been so widely predicted that it is not more likely to happen anytime now than at any other point in time. Unexpected events could trigger it, but unexpected events can happen at any time too.
All very interesting. This raises two questions:
Which bonds to invest in? With stocks it’s easy → VT. But there isn’t a bond equivalent or is there?
And, if you think a market crash is likely, what are you doing? Increasing bond share in your portfolio? If so, the first question gets more interesting still…
I don’t hold any actual bonds atm. I used to consider my pension fund and cash reserve as bond equivalents, only that they obviously don’t appreciate in a stock market crash (to be fair, neither do bonds reliably these days, but who knows, maybe in the next crash).
Tho personally I’m not sure that’s the duration I’d want.
There are two important lessons:
This is what happened in 2007/2008. Everyone expected a Crash, Prices came down, … we had a modest contraction and then Boom.
I am not concerned about a doomsdayer that predicts a market crash out of the blue - I didn’t see any risk for a crash last year. This year however, everyone expects a small contraction from the Chinese Bubble and Geopolitical Fuzz. Trends are going down, which is still no issue. But if we stay in this „gentle contraction“ for lets say 9+ Months, … it could EVOLVE into a big crash.
Way too early to call any alarm bells and always. DO NOT ACT on it, other than you prepare yourself for potential pain. Stay the course but be mentally prepared.
I won’t be the one to argue against that. A crash certainly could happen and we should be ready for it.
I’d be wary not to limit my crash awareness to periods with bad outlook only, though. Crashes can happen in good times too and they don’t always give a polite warning before happening. We should always be ready.
Since we can’t simply always stay in cash just because a crash might happen (well, Swiss people actually can but that would leave quite a bit of potential money on the table), best is to invest according to our need, ability and willingness to take risk. Neither more aggressive nor more conservative than our best assessment of what our risk tolerance and appetite is. For many people (not all), that means holding some amount of stocks and some amount of more conservative assets, which could be bonds.
To be fair, I’m one to try and time the market, mostly out of misplaced pride and for entertainment value. I consider it gambling. Gambling carries the risk of loosing and I strongly advise against gambling money one would regret loosing. Edit: To expand on that, money taken out of the market in anticipation of a crash with the plan to get back in when things look better is not safe money despite how it might look: it is liable to shrink at the re-entry step and that shrinkage may not be small.
I think people avoid bonds due to: lack of knowledge, recency bias, under-estimation of risk, over-estimation of risk tolerance.
I’m currently in 40% bonds. IMO, the risk/reward for stocks looks quite poor at the moment.
But the important thing to remember is by taking advantage of differences in correlation between different asset classes, you can build a portfolio that gives a higher return than an single asset class alone.
e.g. although bonds might have a lower expected return than stocks, a portfolio of 95% stocks and 5% bonds might have a better return than 100% stocks.
At least a simple simulation on portfoliovisualizer.com disagrees with this statement. Adding 5% ITTs or LTTs underperforms 100% US Stock anywhere but in valleys of drawdown. Rolling Returns also show the same picture. Although the difference is small and different periods have Treasuries win sometimes, it is in favor of Stocks only.
That of course changes with higher leverage factors. But you better know what you are doing, when using significant leverage.
It does not disagree with the statement:
It certainly might.
Your simulation looks at a 51-year span of time. That is longer than most investors‘ (let alone early retirees‘!) accumulation phase.
The valleys of drawdown you mentioned may well be a very relevant periods of time for these investors (though a mere 5% allocation to bonds may not make a sizeable difference).
My point was that there are combination of assets that have higher returns than any on their own. Stocks and bonds at 1x leverage (aka. no leverage) is no such combination.
Indeed they might have higher returns, but so does going to the casino. Both have lower expected returns, but stock + bonds portfolios are a lot less volatile / have less drawdown for only a partial reduction in returns.
But one should mind that the backtesting period since the Volcker era has been particularly favorable to Treasuries and the 60/40 portfolio.
Fixed it for you. Over very long time periods, assets with higher expected returns would still be expected to return more than a mix of them “watered down” with assets that have lower expected returns.
The risk is higher too, though, so the actual returns may be higher, or lower, or deeply lower in the down-the-drain level territory. Risk adjusted returns take into account the risk you have to miss your target when using volatile assets. Mixing uncorrelated asset classes with positive expected returns helps to reach higher expected risk adjusted returns.
Correct, in the long run, the expected return of shares far outweights the expected return of a 60/40. The important metric to have a look at is if the lows are higher with Stocks vs 60/40 and there, the interesting observation is that it takes a long time until the lows (e.g. 2009) of stocks only portfolios were materially higher than 60/40. Hence, my Personal conclusion is that the „safe“ remaining investment duration after which one could consider a stocks only Portfolio is more in the range of 15-20 years and not 7-10 years as often stated.
But even if we invest for 15-20+ years, I would not go for Shares only. For such durations, they have the higher expected return than 60/40 and even if the 15-20 years end directly in a crash, Shares only should be better than 60/40… but still the argument against Stock only is that you should never put all your eggs in one basket. A stocks portfolio only takes loads of tail risk that only materializes once in a blue moon - lets say maybe once every 200-300 years or so. But if this tail risk materialzes and your shares are wiped out; then its sad to see your lifetime savings go. Hence, i still advocate for something like 30-40% of assets not in stocks. This may ideally be physical real estate, bonds, money market, gold, … whatever. It just serves as insurance against a write-off scenario on global stocks. Highly likely a scenario we won‘t face in our lifetimes but still something worth insuring against… lets say after your first 10-15 years of savings as you then have limitted working time to recover from such loss.
So until 35, you can go 100% stocks (unless you plan for fire), after this you should shift to 60/40, maybe 70/30. if your target fire/retirement date is closer than 15 years, you should already immediately and regardless of your age down-shift.