I'm still wondering why you ignored a 19.4% drop in the market.
Regarding the "above Portfolio Visualizer success rate analysis thru 2017 assumes ..." It makes no assumptions at all. That's what I was trying to convey in saying that backtesting works with real data. It takes the market as it performed; there are no odds, assumed or otherwise. Run it, run again, run it again, and you'll get the identical result over and over, because the past performance of the market doesn't change.
These results of past investments are as real and meaningful as your statement that the last two (bear) market drops were 57% and 49%.
I stated that I have significant issues with existing tools for simulations of future results. So I'm in agreement that these tools have major limitations, not just under current market conditions, but anytime. That said, I disagree with some aspects of your concerns.
Most simulation tools use mean and standard deviation to project future results. Think bell curve centered at some point above zero (since the average return is positive). So the way they work, they figure that more than half the time the market moves up - there are more points to the right of zero than to the left. They don't assume equal odds.

A problem is that returns aren't symmetric like a bell curve. There are slightly more up days than down days, even though the up days aren't (on average) as big as down days. Here's Forbes' data on the DJIA from October 1, 1928 through January 28, 2016:
https://www.forbes.com/sites/mikepatton/2016/01/29/fast-facts-on-the-dow-jones-stock-index/#145fcc6a6972
"An investors quality of life can be affected if annual income is reduced by a significant amount due to decreases in portfolio value (with static 4% withdrawal program)."
I didn't use a "static 4% withdrawal program." I used as static, inflation adjusted dollar amount independent of portfolio performance.
Using your $1M pot, I started with a $3,333/mo ($40K/year) draw, and adjusted that draw only for inflation. I did not increase or decrease the draw based on the value of the portfolio. I mirrored what Bengen did in his original 4% rule analysis. Only later did he suggest modifying the amount withdrawn based on portfolio value.
In your question about what to do with a $1M inheritance, you didn't say how this person with no assets and no income was living. For the sake of argument, I'll assume his cash flow needs are $40K/year (4%), inflation adjusted.
There are many ways of mitigating sequence risk. I've posted this link before:
https://www.forbes.com/sites/wadepfau/2017/04/12/4-approaches-to-managing-sequence-of-returns-risk-in-retirement/#2b188f486fcf
I like Buffett's approach (which is one of the ideas in the article above - buffer assets, avoid selling at a loss). Buffett suggested keeping 10% in short term treasuries (effectively cash) and the rest in the S&P 500. Personally I'd up the cash figure to 12% or so, and diversify the equity portfolio more broadly, but that's the general idea.

"For an investor in retirement down 57% and continuing to withdraw for income he won't care whether its called a correction or bear market."
For an investor in retirement down 19.4% and continuing to withdraw for income he won't care whether it is called a correction (-19%) or a bear market (-20%).
Yet you ignored a drop of this magnitude even though names don't matter. Was it too small a drop? How large does a market decline have to be for you to care?
We've got at least one fan here of Monte Carlo simulation. I've given many reasons why I find existing tools (not the concept) inadequate. But they apply to probabilistic future projections. Backtesting shows what would have actually happened - no probabilities involved since real market data are used over real periods.
So just for fun I ran portfolioVisualizer from Oct 2007 to Oct 2017 with a representative 50/50 retirement portfolio (SPY/AGG), rebalanced annually. I started monthly withdrawals at 0.33% of the start amount (i.e. 4% annual rate) and increased the withdrawal amount just for inflation.
http://traderhq.com/illustrated-history-every-s-p-500-bear-market/
https://www.portfoliovisualizer.com/backtest-portfolio
Compound aggregate growth rate (CAGR) was positive - around 1.8%/year. It was even positive after adjusting for inflation (about 0.1%/year real return).
Ran the same simulation starting March 2000, except I had to pick another bond representative (AGG started Sept 2003). So I used VBMFX. Sure enough, with two crashes within a decade, that one didn't come out as well.
CAGR was still positive, at 0.23%/year, but after adjusting for inflation, the portfolio lost 1.9%/year in real value. About 18 years after starting, the retiree's portfolio in nominal dollars was up only 4.2% from where it started.
Still, not the dire straits it seems you were expecting.

For an investor in retirement down 57% and continuing to withdraw for income he won't care whether its called a correction or bear market. The pain is the same. Good luck NOT timing the market. Re: 1% cash...IMHO how much you don't lose is more important than how much you make.

As hard rules, they're inconsistent. Most people aren't going to max out their work retirement plans and IRAs with "just" 20% savings.
To max out $18K (401k) + $5500 (IRA) while saving 20% would require an income of at least $117,500. Even more if you're over 50, or have a 457(b) plan that lets you save another $18K on top of your 403(b).
On the other hand, if you are earning that much, there are tax advantages to investing (in limited amounts) in your company stock (an individual security). I've done that through ESPP and ESOP plans.
As davidrmoran wrote, flexibility is the key. I also agree with Crash, that the last item (the one that expands the 3x5 card to a 4x6 card) makes it worth splurging on the larger index card.

These rules may be dumb and hard for financial literates like the ones on this board, but provide very good, simple directions for 70% of us who are not financially literate by at least one measure (Google three question quiz). Those who do not save at all, those who don't know that minimum payment on credit cards should be ignored, those who do not contribute to retirement accounts, etc.

There are parts I agree with, and parts where I see things differently.
I agree that the intent was likely, at least in part, to prevent gaming the system. I make multiple/excess conversions in a year. Then after I see how each investment has done and how much I need to roll back to stay within a tax bracket, I select which investments and how much of each to undo. I'm sure there are others here who do something similar.
All perfectly legal, and all gaming the system. Heads I win, tails the IRS loses.
A simplified version of this is: if the market goes up, keep the conversion. If the market goes down, back it all out. Ed Slott doesn't see this as gaming. I disagree. Still, heads I win, tails the IRS loses.
Ed Slott thinks that if this provision passes, you won't be able to undo your 2017 conversions after December 31, 2017. I read the proposal differently. He's certainly being more conservative, and that's probably good, especially for an advisor.
What the proposed legislation says is: "EFFECTIVE DATE.—The amendments made by
this section [removing the ability to recharacterize] shall apply to taxable years beginning after December 31, 2017."
I take that as meaning that you will not be able to recharacterize anything for tax year 2018. Recharacterizing your 2017 conversions, whenever it is done, applies to tax year 2017.
As supporting evidence, I point to the wording in the current code (which would be deleted):

if, on or before the due date for any taxable year, a taxpayer transfers in a trustee-to-trustee transfer any contribution to an individual retirement plan made during such taxable year from such plan to any other individual retirement plan, such contribution shall be treated as having been made to the transferee plan

A bit of a mouthful, but basically saying that if you undo your tax year 2017 Roth conversion by Oct 15, 2018, it's treated as if you'd just moved the money from your traditional IRA back into a traditional IRA in tax year 2017.
Of course this isn't advice, Ed could be right, or the law might not be changed at all.

FYI: (Click On Article Title At Top Of Google Search)
Tax professionals are ringing alarm bells that a House proposal unveiled last week deserves financial advisers' attention. Should the measure become law, taxpayers who decided to convert a Roth IRA to a traditional, or pre-tax, individual retirement account, would no longer be allowed to elect to change it back to a Roth within a certain time frame. And that means advisers should be checking in on clients about Roths before the start of 2018, said Ed Slott, founder of Ed Slott's Elite IRA Advisor Group in a conversation with InvestmentNews reporter Greg Iacurci.
Regards,
Ted
https://www.google.com/search?source=hp&ei=AH8BWtDnA8y3jwTe5qeAAQ&q=Ed+Slott+on+Roth+IRA+conversions+becoming+permanent&oq=Ed+Slott+on+Roth+IRA+conversions+becoming+permanent&gs_l=psy-ab.3..33i160k1.4590.4590.0.7188.3.2.0.0.0.0.93.93.1.2.0....0...1..64.psy-ab..1.2.178.6..35i39k1.85.2pswNHS_oXI

@Bitzer, AQR LSE and EMN are both closed to new investors anyhow. But fyi, in case they reopen, they were available at Fidelity (and maybe other brokerages, dunno) in IRAs for $2,500 minimum and even less in group retirement accounts. See the Fees & Distributions page.

In general, if I were 30-40 years from retirement I’d prefer to use a pre-tax program like the Traditional IRA. At 10 years from retirement I’d begin transitioning (through Roth contributions and/or conversions). At 60+ I very much like the Roth concept. So these considerations affect my view of the anticipated change (not favorably).
Reason: I have no confidence in being able to anticipate the rules of the road as Congress, the executive branch and federal courts may define them 65-75 years from now (anticipating 35-40 years of making contributions and 30-35 years of withdrawing money. The “promise” of tax exempt contributions exists now. Use it. The promise of tax exempt withdrawals in 75 years is anybody’s guess. I do, however, have a somewhat higher predictive confidence for 20-30 years out. So for a shorter time frame like that, I’ll take my chances transitioning to a Roth-type product.
Memory is a funny thing, especially with politicians. Who knows what the federal budget, tax collection needs and political mood of the nation will be so far out? For some perspective - 75 years ago (1942): Pearl Harbor had just been attacked, FDR was President, the interstate highway system hadn’t yet been built, and the setting for one of my favorite films was taking shape along Nantucket Beach (BTW - not the actual filming location).

Short summary - I haven't been able to find any Obama Roth IRA taxation proposal that would amount to double taxation. Further, the most obvious way of taxing Roths - making the earnings (but not contributions) taxable - would neither amount to double taxation nor be feasible to implement in any case.
----------
Obama floated a lot of ideas regarding IRAs, but I'm having a real problem finding any proposal that wold amount to double taxation of Roth IRAs (i.e. taxation of money in (as is done now) and taxation of the same money (not its earnings) on the way out.
Taxing the earnings would simply make the Roth look like a nondeductible IRA, which no one says is double taxation. A scheme like that would be, IMHO, virtually impossible to implement. That's because taxpayers typically discard records of Roth contributions (and conversions after the five year waiting period is up). So they can't identify what part of a Roth distribution constitutes earnings. In contrast, when you make a nondeductible IRA contribution, you file a Form 8606, and each year after that carry over the information into the current year's tax filing.
Getting back to whether Obama ever proposed double taxing Roths. The best I've been able to find is this statement suggesting that Obama's proposed cap of 28% on the tax benefit (exclusion from income or tax deduction) would result in double taxation. The page notes that this was a new proposal for for FY2017 (2016), i.e. for Obama's last budget.
http://www.pension-specialists.com/hottopics/0216Obama.pdf (item #1)
Under this proposal as described, if you contributed to a traditional IRA, and you were in the 33% tax bracket, you'd pay 5% (33% minus 28%) on your "deductible" contribution, and as the "streamlined" summary appears to read, also pay income tax when you withdrew that contribution. But that's for a traditional IRA, not a Roth.
Further, the government proposal anticipated this problem and addressed it. Here's the full set of FY2017 (2016) proposals:
https://www.treasury.gov/resource-center/tax-policy/Documents/General-Explanations-FY2017.pdf
On pp. 153-154 (pdf pp. 164-165) you'll find the proposal in question. It says that "If a deduction or exclusion for contributions to retirement plans or IRAs is limited by this proposal, then the taxpayer’s basis will be adjusted to reflect the additional tax imposed."
In other words, instead of treating the money coming out as having a zero basis, the cost would be increased (on paper) so that the taxpayer doesn't pay that extra 5% a second time. Part of the contribution would be treated as a nondeductible contribution (which is what it really would be).
I don't see this proposal affecting Roths at all, but given the special handling of traditional IRAs to ensure there's no double taxation, it seems highly unlikely that the same issue wouldn't be addressed, assuming that Roths are even covered by the proposal.

Hello,
After some thought, Old_Skeet has decided it best to keep posting the barometer updates under the Buy-Sell-and-Ponder thread rather than reopening the Markets as More thread. The big reason in me doing this is that I wanted to move away from hosting an ongoing thread and to reopen the Markets and More thread would put me back to doing just that. Thank you @Puddenhead for taking the thread over. It is much appreciated and from my perspective you have brought some new and good insights to the thread. But, what really makes the thread is the post of many and not just a few.
This week Old_Skeet's market barometer closed the week with a reading of 138 and with that moved back into overbought territory. Last week the barometer closed the week with a reading of 143. Generally a higher barometer reading indicates there is more investment value in the S&P 500 Index over lower reading. The three best performing sectors for the Index last week were energy (XLE), real estate (XLRE) and technology (XLK). The only sector found this week, from a technical perspective, as being scored undervalued is staples (XLP). In addition, the Index advanced +0.26% for the week and closed just short of 2588 for a year-to-date gain of 15.59%.
http://www.sectorspdr.com/sectorspdr/tools/sector-tracker
Within, Old_Skeet's global compass the three best performing etfs, that are followed, were VT, EFA and EEM. With this, the global and foreign etf's out performed the domestics this past week.
From an equity allocation review Old_Skeet's equity weighting matrix indicates that due to elevated stock valuations I should be position somewhere towards the low range within equities; however, due to a seasonal investment strategy I am currently overweight (what the matrix calls for) by five percent. Within equities, according the a recent Xray analysis, I am (about) ... 60% domestic and 40% foreign ... 75% large and 25% smids ... 40% value, 30% blend and 30% growth. Looking back to the first of the year I was (about) ... 70% domestic and 30% foreign. Because of my use of a good number hybrid and dynamic type funds (that adjust their asset allocation, styles and sector weightings) has made my portfolio more adaptive to the ever changing market climate without me having to throttle it as much as I use to. Currently, Morningstar Portfolio Manager list my year-to-date investment return at 12.17%. In comparison, the Lipper Balance Index is reported by the WSJ year-to-date at 11.86%. I'm thinking, this is not too bad of a return for a 50/50 portfolio that adjust within certain ranges (of course) and from time-to-time +/- 5% from its neutral position. I'm sure there are others out there that have out performed me. But, being in retirement I have now dialed my risk down a good bit within my mutual fund portfolio along with not being as active with position changes as I use to be.
And, with an overbought barometer reading, I remain in a cash build mode with all mutual fund distributions paying to cash area of the portfolio while I await the next stock market pullback. My next scheduled calendar rebalance is May, 2018 unless market conditions should warrant otherwise.
Thanks for stopping by and reading.
Wishing all ... "Good Investing."
Old_Skeet

https://www.investors.com/etfs-and-funds/retirement/house-tax-reform-bill-leaves-current-401k-contribution-cap-in-place/
Never mind tax cuts for the rich. Here's great news for people who actually work for a living in the tax-reform bill.

This is not surprising if you look at the constituencies.
- Estate tax - largest 0.2% of estates are subject to this tax. Money talks.
- Property (and presumably income) tax deduction - Enough Republican members of Congress from higher tax states to sink any proposal to eliminate this deduction
- 401k cap reduction - Only 1/3 of employees have access to defined contribution plans and participate; of those, only 10% max out. While it is true that nearly all who do are high earners (1/3 of $100K+ earners max out, while 4% or fewer of employees earning $75K- max out), these are still not the really high rollers involved with the estate tax. They have other mechanisms, like top hat plans, self-employed plans (with $55K limits), etc.
So few people are affected by reducing the cap, and those who do, while somewhat better off, aren't the ones pulling the strings.
The distribution of those maxing out comes from this 2017 Vanguard report, linked to by Ted:
https://mutualfundobserver.com/discuss/discussion/33534/vanguard-how-america-saves-2017
----
Edit: Regarding property/income tax deductions - more complete stories report that only the property tax deduction would be preserved. ISTM that's a direct slight of mid-to-upper middle class people, especially in areas like high tax areas like San Francisco and New York, where the majority of people (64% and 51% respectively) are renters, but deduct substantial local income taxes.
Often the combination of state/local income tax and municipal property taxes is enough to exceed the standard deduction, while the property tax alone isn't. (That's especially true in NYC, where the property taxes are lower but there's a city income tax in addition to the state income tax.) So those with extremely expensive homes get to keep itemizing their property taxes, while middle class people lose the deduction, and lower class people never itemized.
"'No matter how they construct this compromise, Republicans are still socking it to the middle class and the upper-middle class, but this time picking winners and losers,' Senate Minority Leader Charles E. Schumer (D-N.Y.) said Monday."
https://www.washingtonpost.com/news/business/wp/2017/10/30/house-gop-tax-plan-would-now-allow-americans-to-deduct-property-taxes/?utm_term=.0eb51c6e000b

Why do we keep on talking about this... They aren't going to change 401(k) limits and are not going to reneg on Roth IRAs tax-free growth.

And how can you be so confident in that statement? Is your crystal ball infallible?
My crystal ball says whether it's this administration or the next or the next after that, rules around taxation on retirement savings will change...
p.s, I also asked my magic 8-ball this exact question and it said "yes" 3 out of 4 times.

Someone gonna paid for the massive cut on their tax burden. This is a slap on the face for those who are responsible for planning their retirement. There is also discussion the Roth IRAs may be subject to taxation upon withdraw at later days.

FYI: Are we living in an upside-down bizarro world where lawmakers want us to save less — instead of more — for retirement?
Regards,
Ted
http://www.philly.com/philly/business/vanguard-greatly-concerned-over-changes-like-congress-proposed-cap-on-401k-plans-20171030.html

Ted - thank you. Also see discussion here regarding expense ratios and SWR's:
Pfau & Dokken: Why 4% Could Fail - Rethinking Retirement
https://mutualfundobserver.com/discuss/discussion/23490
Wade Pfau’s new book: How Much Can I Spend in Retirement? (@ Bogleheads.org)
https://www.bogleheads.org/forum/viewtopic.php?t=229814

msf - thank you for listing of benefits.
So - if I could summarize (at the risk of putting words in your mouth...) it seems as if the benefits of a brokerage (as opposed to a non-brokerage) account, largely benefit traders, as opposed to investors (i.e., "buy and holders").
Note: While there is some "reduced tax preparation paperwork" benefit for brokerage accounts, that evaporates when the accounts - either brokerage or non-brokerage - are held in tax deferred retirement accounts.
Paradoxically then, for Vanguard to promote brokerage, no? I assume that this must come down to costs - since what doesn't at Vanguard? Brokerage must be cheaper for Vanguard and the funds it administers (and which own Vanguard).
So what I think we have is that Vanguard is promoting a platform for which the apparent initial benefits would seem to benefit traders, rather than investors (an unusual stance for them) except that in the longer run, the impact of lower (assumed) costs in brokerage flow through to the funds in lower expense ratios - benefiting (eventually) both traders and investors.
However, as an individual - who rarely trades - it would seem that the optimal position would be to maintain non-brokerage accounts to benefit from the flexibility of directed dividends and automatic investment, while at the same time benefiting from the lower expense ratios as everyone else is herded into brokerage.
... Having said that (and 'figured' it out) I should say no more on the topic.
PS: Thank you Anna for this link:
https://www.thetaxadviser.com/content/dam/tta/issues/2014/jan/stateirachart.pdf

FYI: While individual securities (such as shares of stock in a publicly traded company or a bond issued by a company or government) do not have an annual expense ratio, mutual funds and ETFs always have an expense ratio. The annual expense ratio of a stock or bond mutual fund directly reduces the return of the investor, which reduces the amount of money that can be safely withdrawn during retirement.
Regards,
Ted
http://www.aaii.com/journal/article/the-impact-of-expense-ratios-on-retirement-income

On Oct 27 msf asked:

"How does directing dividends from an IRA tremendously simplify handling of IRA RMD's?"

Here is one answer. Suppose that ...

1. You own a number of Vanguard mutual funds, in a retirement account which is held in the form of a VG regular [non-brokerage] account.
2. Many of the funds pay regular or semi-regular distributions.
3. You take your RMD monthly.
4. Interest rates are unattractively low.

(Sound familiar?)
Then you could - for example - direct all of your distributions (from all of your other funds) to a single -for want of a better term- 'distribution fund'. This might be the Vanguard Wellesley fund or the new Global Wellesley fund.
The 'distribution fund' might have been pre-seeded or funded with some money in addition to the distributions that it (automatically) collects.
You could then take all of your monthly RMD distributions from the single 'distribution fund' (e.g., the Wellesley Fund).
When you check your RMD at the beginning of each year, you could look at the balance of the 'distribution fund', and fudging a little for expected distributions and appreciation or depreciation - make sure that you have enough in the 'distribution fund' to cover the upcoming year's RMD requirements.
Can't do this with a brokerage account.
While one may hold ETFs in addition to VG Mutual Funds prior to retirement and time to take RMD's, once one starts taking RMDs, can liquidate all ETFs in retirement accounts, transfer to more-or-less equivalent Vanguard funds (held in regular non-brokerage retirement accounts) and take advantage of the wonders of directed dividends.
One can also - from time to time - "buy low" or reinforce an existing position, by directing all of the dividends from all of the other mutual funds to the fund that is "low" or which you want to gradually build a position.
etc...