How much do you trust FIRECalc?

Non-investing personal finance issues including insurance, credit, real estate, taxes, employment and legal issues such as trusts and wills.
Topic Author
glamdring269
Posts: 40
Joined: Wed Feb 10, 2016 10:08 am

How much do you trust FIRECalc?

Post by glamdring269 »

Hi all. I've been running Firecalc scenarios like crazy over the past couple of months and every time I do I just find the results a little unbelievable. Some high level background (not looking for investment advice just a sanity check on the calculator):

Calc link for reference: https://www.firecalc.com/

Goal: I hope to retire at 55 (~11 years from now). Spouse will as well, or may lag 1 additional year.

Income: ~$230k
Current holdings: $800k (mixture of 401k/457/403/Roth/Taxable)

Debt: House will be paid off by retirement. Currently putting ~$2k/mo towards the house to accomplish this.

Yearly Savings: ~$105k (after 401k, 403b, 457, Roth, employer match, taxable contributions)

Yearly Expenses: ~$80k (including the house, so ~$56k excluding house)

Pension Estimate: ~35k/ yr

So here's the simple question. I've plugged all of this into firecalc, and not even accounting for social security it is leading me to believe that as long as we continue this savings rate (at min) up to retirement then we could actually increase our spending into the lower $100k range per year. I suppose this is plausible as we would no longer be saving $100k/yr and would no longer be paying ~$24k/yr in house payments. Plus would be pulling in half of our current spend in pension. It gets even crazier ($150+) when you use the model that assumes lowered spending once you reach a certain age.

This is probably one of those posts where I don't even need to post it but it just feels so unbelievable to me that I'd really appreciate some feedback. Thanks.
User avatar
Wiggums
Posts: 7050
Joined: Thu Jan 31, 2019 7:02 am

Re: How much do you trust FIRECalc?

Post by Wiggums »

It really comes down to your retirement expenses and how much you saved when you arrive at retirement decision time. I’m in favor of saving a little extra and keeping a record of your current expenses. This will help you tremendously whenever you make the decision to retire. Firecalc is a good tool.
Last edited by Wiggums on Fri May 07, 2021 10:08 am, edited 1 time in total.
"I started with nothing and I still have most of it left."
Wanderingwheelz
Posts: 3145
Joined: Mon Mar 04, 2019 8:52 am

Re: How much do you trust FIRECalc?

Post by Wanderingwheelz »

I trust it knowing I’ve got enough wiggle room to make adjustments come retirement time if need be.

No retirement calculator can see the future. It’s using the past to model the most likely future based on what’s happened already.
Being wrong compounds forever.
User avatar
David Jay
Posts: 14586
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: How much do you trust FIRECalc?

Post by David Jay »

If it projects that you have, say, $3.2M at age 55, it would be reasonable to withdraw 3.5% or about $110,000 a year. Do keep in mind that the spending power of $110,000 will be considerably less in 2032 than in 2021, but still likely well more than your $56K minimum living expense.

...and you still have SS to back up your portfolio in the out years.
It's not an engineering problem - Hersh Shefrin | To get the "risk premium", you really do have to take the risk - nisiprius
Grt2bOutdoors
Posts: 25625
Joined: Thu Apr 05, 2007 8:20 pm
Location: New York

Re: How much do you trust FIRECalc?

Post by Grt2bOutdoors »

What is the current allocation of your retirement account and will you continue to hold that allocation up until planned retirement and what is your allocation in retirement (same, higher or lower equity allocation split - domestic/international)?

I get $3Mn with your current balance and $105k contribution for 11 more years. That plus a $35k pension at a 3 percent draw rate on account of early retirement age is $120-125k. But wait, if returns are 4%, then it’s $118,000. At 3%, it’s $110,000.

But that assumes a continuous return and we know that nothing is guaranteed in the markets. Keep saving.
"One should invest based on their need, ability and willingness to take risk - Larry Swedroe" Asking Portfolio Questions
User avatar
RickBoglehead
Posts: 7877
Joined: Wed Feb 14, 2018 8:10 am
Location: In a house

Re: How much do you trust FIRECalc?

Post by RickBoglehead »

I don't know what "running it like crazy" is worth.

You want to retire at 55. Figure you'll live to 95. That's 40 years.
You will save $100k+ for 11 more years. That's $1.1 million

$110k per year for 40 years is $4.4 million. Subtract pension of $1.4 million (assumes zero growth over 40 years). $3M needed.

The $800k you have now will likely double in 11 years. That's $1.6M.
The $1.1 million you'll save over the next 11 years will increase by 50%. That's $550,000 more.

So at retirement, you'll have $3.25 million, and you need $3 million over 40 years. Seems a no brainer with normal returns.
Last edited by RickBoglehead on Fri May 07, 2021 10:43 am, edited 1 time in total.
Avid user of forums on variety of interests-financial, home brewing, F-150, EV, home repair, etc. Enjoy learning & passing on knowledge. It's PRINCIPAL, not PRINCIPLE. I ADVISE you to seek ADVICE.
Exchme
Posts: 1335
Joined: Sun Sep 06, 2020 3:00 pm

Re: How much do you trust FIRECalc?

Post by Exchme »

FireCalc is a good blend of simplicity of input while providing a stress test to see if your portfolio would have handled prior downturns.

The weakest point of FireCalc in my view is taxes. You have to estimate them and include them as part of your "expenses". Those can vary over time and in particular, they will go up once you start withdrawing (or taking RMDs) from your tax deferred accounts and taxes will obviously be higher if your portfolio grows. But the program doesn't handle any of those changes.

A small step up in complexity might be i-orp.com, scroll down and find the "Extended" input. It is more thorough and makes better estimates of taxes (still not everything, but a reasonable estimate of how they will change). You can input rates of return or use the 3-PEAT feature to do historical simulations and it also has a Monte Carlo feature. Free, though I never quite figured out how to save my inputs once I close the browser. Different approach than the others in that it is designed to tell you the maximum you can withdraw each year and spend your last dime as you expire.

More complex models include the free Retiree Portfolio Model spreadsheet available at this site, which is based on you providing the rate of return and a sister sheet that does Monte Carlo. Requires a lot more input but handle taxes more thoroughly.

The most thorough I've seen is and Pralana Gold ($99, requires Excel). It has features you didn't even know you needed until you use them. Search some other threads here for those terms and you can see some feature comparisons.

As long as you keep working and saving and the markets don't crash, it sounds very believable that you can retire at 55.
User avatar
BolderBoy
Posts: 6753
Joined: Wed Apr 07, 2010 12:16 pm
Location: Colorado

Re: How much do you trust FIRECalc?

Post by BolderBoy »

I trusted it enough to pull the plug and retire. Ran the numbers through cFIREsim.com as well.

In my case I used outrageous, worst-case scenarios such as no social security, 10% inflation, 1% market returns, 2x annual spending + taxes, etc.
"Never underestimate one's capacity to overestimate one's abilities" - The Dunning-Kruger Effect
User avatar
vitaflo
Posts: 1905
Joined: Sat Sep 03, 2011 3:02 pm

Re: How much do you trust FIRECalc?

Post by vitaflo »

I use it as a "back of the napkin" calculator. IE, I mostly use it to see if I'm on the right track. I'm not looking to see if my success rate is 95% or 96%. I'm looking to see if it's 95% or 50%. I can then adjust my expectations accordingly. But I definitely use it as a course measurement tool than a granular one.
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: How much do you trust FIRECalc?

Post by dbr »

The FireCalc methodology is simply to tabulate what would have happened to you if you had implemented your plan starting in each of a hundred or so historical years and observed the actual result. In that sense the output is not subject to uncertainty or doubt because it is a tabulation of established facts. I am not aware of someone finding any errors or faulty assumptions or incorrect algorithms or bad data in any of this.

So what is implied by this is that this historical tabulation represents a sample from which a future distribution of possible wealth over time can be estimated and from which the factual fraction of success can be used as an estimate of future probability of success. As such the result is uncertain in the way any sample is in error regarding the universe of underlying data. The result is also uncertain in that the universe of hypothetical underlying data is not stable and varies over time, including that the statistics of the future will be different from the statistics of the past to a great enough degree for a long enough time to produce a misleading estimate.

I am personally willing to accept that the result is good enough for a person to estimate whether or not their proposed retirement will probably work. This is provisional on understanding that all retirements, like all of life, are uncertain and that a person is going to have to adjust all the time as a matter of course. A person arriving at retirement should have had enough practice at adjusting to be good at it.

A solution to doubt is to allow contingency. A cost of taking too pessimistic an outlook is to choose to work longer, spend less, and save more together with plan for a less expensive retirement.
h82goslw
Posts: 491
Joined: Fri Jun 17, 2016 5:44 am

Re: How much do you trust FIRECalc?

Post by h82goslw »

Glad to see someone else is questioning firecalc. I go in several times a year and input data based on current amounts and future savings rate and always stare at amazement at what it says my spending level can be while still maintaining a 95-100% success rate.
Mike Scott
Posts: 3579
Joined: Fri Jul 19, 2013 2:45 pm

Re: How much do you trust FIRECalc?

Post by Mike Scott »

If you are saving half or more of your gross income, why would you be surprised that you will probably have a lot of money during retirement?
Barsoom
Posts: 727
Joined: Thu Dec 06, 2018 8:40 am

Re: How much do you trust FIRECalc?

Post by Barsoom »

Just remember that FIRECalc does not account for taxes. The number that it reports is TOTAL spending. You will have to deduct your expected taxes from that amount, and the remainder is what is available for living expenses.

If you want a tool like FIRECalc that goes to the next level, try I-ORP. https://www.i-orp.com/Plans/index.html

Use the Extended ORP feature.

-B
makeitcount
Posts: 288
Joined: Wed Aug 01, 2018 8:07 pm

Re: How much do you trust FIRECalc?

Post by makeitcount »

dbr wrote: Fri May 07, 2021 11:34 am The FireCalc methodology is simply to tabulate what would have happened to you if you had implemented your plan starting in each of a hundred or so historical years and observed the actual result. In that sense the output is not subject to uncertainty or doubt because it is a tabulation of established facts. I am not aware of someone finding any errors or faulty assumptions or incorrect algorithms or bad data in any of this.

So what is implied by this is that this historical tabulation represents a sample from which a future distribution of possible wealth over time can be estimated and from which the factual fraction of success can be used as an estimate of future probability of success. As such the result is uncertain in the way any sample is in error regarding the universe of underlying data. The result is also uncertain in that the universe of hypothetical underlying data is not stable and varies over time, including that the statistics of the future will be different from the statistics of the past to a great enough degree for a long enough time to produce a misleading estimate.

I am personally willing to accept that the result is good enough for a person to estimate whether or not their proposed retirement will probably work. This is provisional on understanding that all retirements, like all of life, are uncertain and that a person is going to have to adjust all the time as a matter of course. A person arriving at retirement should have had enough practice at adjusting to be good at it.

A solution to doubt is to allow contingency. A cost of taking too pessimistic an outlook is to choose to work longer, spend less, and save more together with plan for a less expensive retirement.
as per usual, DBR knows what's up
"Yeah, well, you know, that's just like, uh, your opinion, man." - J. Lebowski
Normchad
Posts: 5648
Joined: Thu Mar 03, 2011 6:20 am

Re: How much do you trust FIRECalc?

Post by Normchad »

makeitcount wrote: Fri May 07, 2021 1:03 pm
dbr wrote: Fri May 07, 2021 11:34 am The FireCalc methodology is simply to tabulate what would have happened to you if you had implemented your plan starting in each of a hundred or so historical years and observed the actual result. In that sense the output is not subject to uncertainty or doubt because it is a tabulation of established facts. I am not aware of someone finding any errors or faulty assumptions or incorrect algorithms or bad data in any of this.

So what is implied by this is that this historical tabulation represents a sample from which a future distribution of possible wealth over time can be estimated and from which the factual fraction of success can be used as an estimate of future probability of success. As such the result is uncertain in the way any sample is in error regarding the universe of underlying data. The result is also uncertain in that the universe of hypothetical underlying data is not stable and varies over time, including that the statistics of the future will be different from the statistics of the past to a great enough degree for a long enough time to produce a misleading estimate.

I am personally willing to accept that the result is good enough for a person to estimate whether or not their proposed retirement will probably work. This is provisional on understanding that all retirements, like all of life, are uncertain and that a person is going to have to adjust all the time as a matter of course. A person arriving at retirement should have had enough practice at adjusting to be good at it.

A solution to doubt is to allow contingency. A cost of taking too pessimistic an outlook is to choose to work longer, spend less, and save more together with plan for a less expensive retirement.
as per usual, DBR knows what's up
Yep. This is perhaps the best, most concise explanation I’ve seen for it.

I completely trust it 100%. Only because I understand what it’s doing and what it can’t do. The obvious concern is somebody making a terrible decision because they read too much into what it is saying.....
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: How much do you trust FIRECalc?

Post by dbr »

One should also keep in mind that what one actually holds may not be exactly represented in the data set FireCalc uses. There are all kinds of tilts and combinations an investor today might choose and even assets that didn't exist over most of the historic periods. If you are heavily invested in foreign stocks, hold a lot of TIPS, have only CDs and not any bonds, and so on and so on, then the results could look different. I don't think that creates an issue of not trusting the output, but it is something to be aware of in terms of how well a historic tabulation would represent your own possible future experience.

I think a bigger issue is for the user to be acutely aware of what it means to talk about the future evolution of a portfolio, contributions, withdrawals, or whatever. That issue is that what is being projected is a set of points (portfolio trajectories from each starting year) that are to be interpreted as a representation of a distribution of possible outcomes one single one of which will be the actual experience of the investor. The challenge to the investor is to comprehend what it means to him that the future is that uncertain. The derived result, the fraction of successful outcomes, treated as a probability of success, also is plagued by the same problem of interpretation. What does it mean that there is x% chance of failure. Shouldn't the number of failures be zero. If the program has no failures should one presume failure is impossible? If there is a 5% chance of failure should that mean the investor is doomed? Keep in mind that one is trying to find the occurrence of extreme results and that is always problematic in any exercise of statistical inference. A person might do well to read Bernstein's series on The Retirement Calculator from Hell for perspective. It is an article in five parts the first of which is here: http://efficientfrontier.com/ef/998/hell.htm The second is here: http://efficientfrontier.com/ef/101/hell101.htm Part 3: http://www.efficientfrontier.com/ef/901/hell3.htm Part 4: http://efficientfrontier.com/ef/103/hell4.htm Part 5: http://www.efficientfrontier.com/ef/403/hell5.htm
Barsoom
Posts: 727
Joined: Thu Dec 06, 2018 8:40 am

Re: How much do you trust FIRECalc?

Post by Barsoom »

dbr wrote: Fri May 07, 2021 3:04 pmA person might do well to read Bernstein's series on The Retirement Calculator from Hell for perspective. It is an article in five parts the first of which is here: http://efficientfrontier.com/ef/998/hell.htm The second is here: http://efficientfrontier.com/ef/101/hell101.htm Part 3: http://www.efficientfrontier.com/ef/901/hell3.htm Part 4: http://efficientfrontier.com/ef/103/hell4.htm Part 5: http://www.efficientfrontier.com/ef/403/hell5.htm
From the article, part 2:
Remember too, in each period we have to withdraw 5% of the initial capital. We generate 50 different random normal numbers to calculate a run of 50 years. The run is categorized as a "success" if we have any money left after 50 years, and a "failure" if we don’t. Repeating the process thousands of times gives a fairly solid statistical estimation of the true success rate.

Monte Carlo simulation requires a large amount of computation. For example, 100,000 runs of 50-year simulations using monthly returns involves 60,000,000 separate monthly calculations. As a practical matter, this is beyond the memory capacity of most desktop spreadsheet systems. (The outer limits of microprocessor/spreadsheet analysis is about 30 annual periods x 10,000 runs.)
Note that the copyright on the article is 2001, which sets the state of the referenced PC power to 20 years ago.

Running FIRECalc and I-ORP drove me to want to write my own Monte Carlo simulation based on slightly different assumptions than what was mentioned in this article. Instead of taking a normal distribution of returns, I took the historical returns of various indexes from the Simba Backtest Spreadsheet and correlated them. I then used these correlations to generate 1,000 possible 40-year (annual) sequences for inflation, bond, stock, and money market returns. This allows me to avoid the "one rate for all years" problem, and it also generates many more diverse possible futures. Plus, it runs in under 30 seconds in Excel.

The base for my Monte Carlo analyzer was the BH Retiree Portfolio Model, which has a 40-year planning horizon for retirees, breaks out the portfolio into after-tax, two before-tax IRAs, two ROTH IRAs, and two inherited IRAs. Separate returns can be entered for each type of account or for each asset class (regardless of which accounts they are in). My Monte Carlo analyzer supports the latter, the asset class rates. The Retiree Portfolio Model also calculates taxes for you, which made it a solid basis for Monte Carlo analysis.

When interpreting Monte Carlo results, it is important to know what the tails are saying, but to also not discard the approach because of extreme high-side tails. On the low side, the tail shows the failure rate, that is, the probability of running out of money. I think the cumulate distribution curve should cross the zero line no higher than at 5%, but it should also cross very close to the bend in the curve (the point where the angle of the tail begins to move away from the flat portion). The bend in the curve is where the extreme cases emerge. The low end of the tail will hopefully be contained, because the point of running out of money will likely be very near the end of the planning horizion.

On the high side, the tail can appear to run away. This should happen no lower than the 92%-95% point in the curve, and shouldn't have too much drag on the expected value (the probability-weighted average of all the results). We're really interested in the 90% of the curve between the 5% and 95% points. The key numbers to see are the mode (the "likely" result, or the result that occurs most often) and the mean (the expected value or probability-weighted average). My model breaks the probability distribution into a histogram of 21 "buckets" and identifies the "likely" result as the mid-point of the bucket with the most results in it). The portfolio should be expected to fall somewhere between these two values.

The other thing that's nice is coupling Monte Carlo analysis with Excel's Solver. This allows me to set a desired risk level and let solver find the maximum annual withdrawal that won't exceed the risk level. It can also look for asset allocations that maximize withdrawal rates or final portfolio balances (for inheritance).

Here are the links to the spreadsheets that I talked about.

BH Retiree Portfolio Model

BH Retiree Portfolio Model with Monte Carlo Variant

-B
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: How much do you trust FIRECalc?

Post by dbr »

These sorts of things should be helpful for insight into what one is doing with investments.
MrCheapo
Posts: 1468
Joined: Tue Dec 22, 2020 2:43 pm

Re: How much do you trust FIRECalc?

Post by MrCheapo »

Not to hijack this thread but doesn't Firecalc assume the "expenses" you state will not increase with inflation? I read that somewhere on this board. If so, then OP entering $56K (his expenses after paying off his home) is assuming he'll have expenses of $56K (in 2021 dollars) in 2060!
docL
Posts: 25
Joined: Sat Jul 10, 2021 8:25 am

Re: How much do you trust FIRECalc?

Post by docL »

MrCheapo wrote: Tue Jul 13, 2021 8:00 pm Not to hijack this thread but doesn't Firecalc assume the "expenses" you state will not increase with inflation? I read that somewhere on this board. If so, then OP entering $56K (his expenses after paying off his home) is assuming he'll have expenses of $56K (in 2021 dollars) in 2060!
FireCalc's default is to adjust retirement spending for inflation every year.
MrCheapo
Posts: 1468
Joined: Tue Dec 22, 2020 2:43 pm

Re: How much do you trust FIRECalc?

Post by MrCheapo »

docL wrote: Tue Jul 13, 2021 8:08 pm
MrCheapo wrote: Tue Jul 13, 2021 8:00 pm Not to hijack this thread but doesn't Firecalc assume the "expenses" you state will not increase with inflation? I read that somewhere on this board. If so, then OP entering $56K (his expenses after paying off his home) is assuming he'll have expenses of $56K (in 2021 dollars) in 2060!
FireCalc's default is to adjust retirement spending for inflation every year.
Then I'm completely baffled by it. It says for a spend of 84K a year my $2.7M of investments over 40 years I'll end up with $10M on average?

Edit. I see that it assumes solely investing in the US equities market by default. Perhaps it assumes the last 20 years are a good barometer to the next? That would be great but unlikely!
User avatar
Lee_WSP
Posts: 10401
Joined: Fri Apr 19, 2019 5:15 pm
Location: Arizona

Re: How much do you trust FIRECalc?

Post by Lee_WSP »

It doesn't do a Monte Carlo, it just looks at historical periods.
FIRECalc looked at the 121 possible 30 year periods in the available data
I mean, if it comes back with a zero possibility of failure, I think it's safe to trust the result. Any more than that and it'll depend on your risk tolerance for fudge factor.
Silk McCue
Posts: 8951
Joined: Thu Feb 25, 2016 6:11 pm

Re: How much do you trust FIRECalc?

Post by Silk McCue »

MrCheapo wrote: Tue Jul 13, 2021 8:16 pm Then I'm completely baffled by it. It says for a spend of 84K a year my $2.7M of investments over 40 years I'll end up with $10M on average?

Edit. I see that it assumes solely investing in the US equities market by default. Perhaps it assumes the last 20 years are a good barometer to the next? That would be great but unlikely!
Maybe you should read the main page and see what it says about the tool and understand it’s inputs and options.

Cheers
hi_there
Posts: 1182
Joined: Sat Aug 29, 2020 7:00 pm

Re: How much do you trust FIRECalc?

Post by hi_there »

MrCheapo wrote: Tue Jul 13, 2021 8:16 pm
docL wrote: Tue Jul 13, 2021 8:08 pm
MrCheapo wrote: Tue Jul 13, 2021 8:00 pm Not to hijack this thread but doesn't Firecalc assume the "expenses" you state will not increase with inflation? I read that somewhere on this board. If so, then OP entering $56K (his expenses after paying off his home) is assuming he'll have expenses of $56K (in 2021 dollars) in 2060!
FireCalc's default is to adjust retirement spending for inflation every year.
Then I'm completely baffled by it. It says for a spend of 84K a year my $2.7M of investments over 40 years I'll end up with $10M on average?

Edit. I see that it assumes solely investing in the US equities market by default. Perhaps it assumes the last 20 years are a good barometer to the next? That would be great but unlikely!
Are these numbers in real or nominal dollars? If it is not inflation adjusted, then that does not seem far fetched at all with your 3.1% withdrawal rate. Even if it is in real dollars, it looks aggressive but not implausible. 40 years is a long, long time for investments to compound.
Exchme
Posts: 1335
Joined: Sun Sep 06, 2020 3:00 pm

Re: How much do you trust FIRECalc?

Post by Exchme »

MrCheapo wrote: Tue Jul 13, 2021 8:16 pm
docL wrote: Tue Jul 13, 2021 8:08 pm
MrCheapo wrote: Tue Jul 13, 2021 8:00 pm Not to hijack this thread but doesn't Firecalc assume the "expenses" you state will not increase with inflation? I read that somewhere on this board. If so, then OP entering $56K (his expenses after paying off his home) is assuming he'll have expenses of $56K (in 2021 dollars) in 2060!
FireCalc's default is to adjust retirement spending for inflation every year.
Then I'm completely baffled by it. It says for a spend of 84K a year my $2.7M of investments over 40 years I'll end up with $10M on average?

Edit. I see that it assumes solely investing in the US equities market by default. Perhaps it assumes the last 20 years are a good barometer to the next? That would be great but unlikely!
Doesn't seem surprising, remember the 4% SWR guideline for 30 years retirement came from looking at the worst case so far. The average case is far better. Before you spend all that future dough though, realize that FireCalc is blind to taxes, you have to enter them yourself, but if your assets skyrocket, your taxes will too. That's not really a defect of the program, it does a good job of giving people a feel for the variability of outcomes and give them insight into answering the "Can I retire?" question from simple inputs.
User avatar
tooluser
Posts: 1557
Joined: Sat Oct 01, 2011 7:04 pm

Re: How much do you trust FIRECalc?

Post by tooluser »

My experiences, normalized to my personal retirement income number (myspreadsheet). Bottom line: they are all in the ballpark, most are probably okay to trust if you have vetted your inputs and understand any assumptions they may have made.

Code: Select all

myspreadsheet		101%
firecalc		101%
c-firesim		Always very close to firecalc
Vanguard		99%
Fidelity		No data yet
TD Ameritrade		99% success rate of withdrawing 99% for 39 years
Merrill Edge		No data yet
VPW Spreadsheet		102%
ABW Spreadsheet		102%
Essential iORP		94%
Extended iORP		No data yet
mycompanyplanner	95%
ERN SWR Spreadsheet	3.46% perpetual SWR for a 70/30 portfolio similar to my own
Portfolio Visualizer	3.5% perpetual SWR for a 70/30 portfolio similar to my own
myspreadsheet currently predicts 101% of my desired income at my desired retirement age, which is coming up soon, not predicting out many years. It assumes a 3.5% withdrawal rate.
All the calculators work differently and it takes quite a while to work through some of them.
Most produce similar results once you work through getting the inputs correct.

The Fidelity, Merrill, iORP, and my company's calculators all produce a number exclusive of taxes. I hate that. It just adds extra work to add the taxes back in, and I may not have formatted the inputs correctly. I've run them all but don't yet trust most of the numbers they are putting out. They all predict somewhat lower numbers, I'm not sure why. They are hard to compare so I tend to not use them.
Charon
Posts: 631
Joined: Thu May 03, 2018 12:08 pm

Re: How much do you trust FIRECalc?

Post by Charon »

David Jay wrote: Fri May 07, 2021 10:09 am If it projects that you have, say, $3.2M at age 55, it would be reasonable to withdraw 3.5% or about $110,000 a year. Do keep in mind that the spending power of $110,000 will be considerably less in 2032 than in 2021
FIRECalc works in real dollars. It wouldn't make sense to compare nominal dollars from a variety of historical periods with a wide variety of inflation rates.
flyingaway
Posts: 3908
Joined: Fri Jan 17, 2014 9:19 am

Re: How much do you trust FIRECalc?

Post by flyingaway »

IMHO, there is nothing special about taxes. You have income, you pay taxes. Your expenses should include income taxes, property taxes, etc. FIRECalc is just the dynamic demonstrator of the 4% rule. It tells you some useful information. But the future cannot be predicted precisely.
sycamore
Posts: 6359
Joined: Tue May 08, 2018 12:06 pm

Re: How much do you trust FIRECalc?

Post by sycamore »

tooluser wrote: Tue Jul 13, 2021 9:22 pm My experiences, normalized to my personal retirement income number (myspreadsheet). Bottom line: they are all in the ballpark, most are probably okay to trust if you have vetted your inputs and understand any assumptions they may have made.

Code: Select all

...
ERN SWR Spreadsheet	3.46% perpetual SWR for a 70/30 portfolio similar to my own
Portfolio Visualizer	3.5% perpetual SWR for a 70/30 portfolio similar to my own
Thanks for that list of calculators, I've already tried some and now will try some of the others.

But I'm wondering what "perpetual SWR" is? Perpetual I take to mean the portfolio is never depleted. And SWR means Safe Withdrawal Rateas defined in the wiki. What's a perpetual Safe Withdrawal Rate?
dcabler
Posts: 4539
Joined: Wed Feb 19, 2014 10:30 am
Location: TX

Re: How much do you trust FIRECalc?

Post by dcabler »

dbr wrote: Fri May 07, 2021 11:34 am The FireCalc methodology is simply to tabulate what would have happened to you if you had implemented your plan starting in each of a hundred or so historical years and observed the actual result. In that sense the output is not subject to uncertainty or doubt because it is a tabulation of established facts. I am not aware of someone finding any errors or faulty assumptions or incorrect algorithms or bad data in any of this.

So what is implied by this is that this historical tabulation represents a sample from which a future distribution of possible wealth over time can be estimated and from which the factual fraction of success can be used as an estimate of future probability of success. As such the result is uncertain in the way any sample is in error regarding the universe of underlying data. The result is also uncertain in that the universe of hypothetical underlying data is not stable and varies over time, including that the statistics of the future will be different from the statistics of the past to a great enough degree for a long enough time to produce a misleading estimate.

I am personally willing to accept that the result is good enough for a person to estimate whether or not their proposed retirement will probably work. This is provisional on understanding that all retirements, like all of life, are uncertain and that a person is going to have to adjust all the time as a matter of course. A person arriving at retirement should have had enough practice at adjusting to be good at it.

A solution to doubt is to allow contingency. A cost of taking too pessimistic an outlook is to choose to work longer, spend less, and save more together with plan for a less expensive retirement.
Bingo! I've used this, along with several other calculators including Fidelity's RIP as only an extremely rough estimate for a go/no-go type of evaluation. I would never use any of these to calculate my actual withdrawals in retirement. For that, I use my own spreadsheet based on what is now called ABW on the forum to determine what my withdrawals would be if I retired today and it will be what I use once I do retire. It has its own separate backtester I've used to estimate how much variability my withdrawals would have been in the past, in real terms. If I can survive the lowest historical percentage drop from my initial withdrawal, with some additional margin, I consider myself good to go. No surprise that this happens for the same beginning year that SWR methods also show is the worst historical case. Regardless of what a spreadsheet tells you, it's still important to maintain flexibility and have other contingency plans, though, both for upsides and downsides.

Plan B if I predecease my spouse or become mentally incapacitated and she's uninterested in my spreadsheet is for her to use a fixed % withdrawal.

Cheers.
Last edited by dcabler on Wed Jul 14, 2021 6:03 am, edited 2 times in total.
dcabler
Posts: 4539
Joined: Wed Feb 19, 2014 10:30 am
Location: TX

Re: How much do you trust FIRECalc?

Post by dcabler »

sycamore wrote: Wed Jul 14, 2021 5:11 am
tooluser wrote: Tue Jul 13, 2021 9:22 pm My experiences, normalized to my personal retirement income number (myspreadsheet). Bottom line: they are all in the ballpark, most are probably okay to trust if you have vetted your inputs and understand any assumptions they may have made.

Code: Select all

...
ERN SWR Spreadsheet	3.46% perpetual SWR for a 70/30 portfolio similar to my own
Portfolio Visualizer	3.5% perpetual SWR for a 70/30 portfolio similar to my own
Thanks for that list of calculators, I've already tried some and now will try some of the others.

But I'm wondering what "perpetual SWR" is? Perpetual I take to mean the portfolio is never depleted. And SWR means Safe Withdrawal Rateas defined in the wiki. What's a perpetual Safe Withdrawal Rate?
Perpetual withdrawal rate is commonly referred to as the withdrawal rate where, at the end of your retirement, you still have as much in your accounts as you started with, in inflation adjusted terms. Again, usual disclaimers about this being based on past history, future is unknown, yada yada...

Explained here on portfoliocharts: https://portfoliocharts.com/2016/12/09/ ... etirement/
Dusn
Posts: 217
Joined: Thu Feb 15, 2018 8:59 am

Re: How much do you trust FIRECalc?

Post by Dusn »

One thing that I find weird with firecalc is that it will often give me a higher percentage of success over a 60 year period than a 50 year one.

I assume that this is because there are just fewer 60 year periods to run the simulator with but it does make it seem like the simulator could be improved.
sureshoe
Posts: 2165
Joined: Tue Jan 15, 2019 3:26 pm

Re: How much do you trust FIRECalc?

Post by sureshoe »

MrCheapo wrote: Tue Jul 13, 2021 8:16 pm
docL wrote: Tue Jul 13, 2021 8:08 pm
MrCheapo wrote: Tue Jul 13, 2021 8:00 pm Not to hijack this thread but doesn't Firecalc assume the "expenses" you state will not increase with inflation? I read that somewhere on this board. If so, then OP entering $56K (his expenses after paying off his home) is assuming he'll have expenses of $56K (in 2021 dollars) in 2060!
FireCalc's default is to adjust retirement spending for inflation every year.
Then I'm completely baffled by it. It says for a spend of 84K a year my $2.7M of investments over 40 years I'll end up with $10M on average?

Edit. I see that it assumes solely investing in the US equities market by default. Perhaps it assumes the last 20 years are a good barometer to the next? That would be great but unlikely!
That feels a bit high, but not shocking. You can calc a simple projection in Excel. Put 2,700,000 in one cell and under it put: =SUM((G1-80000)*1.08)
Copy that 40 times. Very rough math. That assumes an 8% average return. I got $31M. Makes sense, at 8% you're going to double-ish every 9 years.

I spitball it's averaging around 5.6% maybe? This is my ending balance by year... but just playing and don't know the calculator that well.

5.60%
2,700,000
2,759,328
2,821,978
2,888,137
2,958,001
3,031,777
3,109,684
3,191,955
3,278,832
3,370,575
3,467,455
3,569,760
3,677,795
3,791,880
3,912,353
4,039,573
4,173,917
4,315,784
4,465,596
4,623,797
4,790,858
4,967,274
5,153,569
5,350,297
5,558,042
5,777,420
6,009,084
6,253,720
6,512,057
6,784,860
7,072,940
7,377,153
7,698,401
8,037,640
8,395,875
8,774,172
9,173,654
9,595,507
10,040,983
10,511,406
User avatar
RickBoglehead
Posts: 7877
Joined: Wed Feb 14, 2018 8:10 am
Location: In a house

Re: How much do you trust FIRECalc?

Post by RickBoglehead »

All these models are fine, but nothing beats making your own spreadsheet. I made a spreadsheet with a row for each year. It starts with the amount in our taxable account on 1/1. I then forecast dividends, interest, and capital gains, add in Social Security (when it will kick in) and pension. Then selling of assets (i.e. house). Then forecast expenses, using our Quicken data, and then making some changes to it. Medicare, and then total it all to net earnings for the year, minus standard deduction, and taxes. That gives me a projected ending balance without market growth, and I then used 2% growth for my taxable investments. I have one more column which grows our retirement balances by 6%. At the end of 30ish years, that untapped retirement balance is huge, yes. Sell house at age 75, move into assisted living at $200,000 per person.

You can build anything you want, and make any assumptions you want. The point is that doing it yourself lets you see how comfortable you feel with what others do with smoke and mirrors.
Avid user of forums on variety of interests-financial, home brewing, F-150, EV, home repair, etc. Enjoy learning & passing on knowledge. It's PRINCIPAL, not PRINCIPLE. I ADVISE you to seek ADVICE.
MrCheapo
Posts: 1468
Joined: Tue Dec 22, 2020 2:43 pm

Re: How much do you trust FIRECalc?

Post by MrCheapo »

Silk McCue wrote: Tue Jul 13, 2021 8:23 pm
MrCheapo wrote: Tue Jul 13, 2021 8:16 pm Then I'm completely baffled by it. It says for a spend of 84K a year my $2.7M of investments over 40 years I'll end up with $10M on average?

Edit. I see that it assumes solely investing in the US equities market by default. Perhaps it assumes the last 20 years are a good barometer to the next? That would be great but unlikely!
Maybe you should read the main page and see what it says about the tool and understand it’s inputs and options.

Cheers
I did read through it and am still baffled.

I entered my full information very conservatively based on today's numbers (I'll retire in 5 years).

2.7M investments, (assumes I don't make any further contributions)
84K a year in pensions (assumes no pay increases for 5 years)
30K a year in SS collected at 67 (this assumes SSN will pay of just 60% on the amount due)

and then I played with how much I can spend for 45 years (by which time I'll be 100).

Turns out I can spend unto $175K a year (with inflation adjustments) each and every year for next 45 years and NEVER get a scenario where I'm out of money.

I can see my future self at 99 sipping cocktails and sucking caviar (because I have no teeth and can't eat steak) ...

I wonder if the scenarios it uses are heavily skewed to the recent 20 years which have been the greatest secular bull market of all time.
Silk McCue
Posts: 8951
Joined: Thu Feb 25, 2016 6:11 pm

Re: How much do you trust FIRECalc?

Post by Silk McCue »

MrCheapo wrote: Wed Jul 14, 2021 8:37 am
I wonder if the scenarios it uses are heavily skewed to the recent 20 years which have been the greatest secular bull market of all time.
It doesn't skew to the recent 20 years. Even the worst possible historical scenarios it modeled gave you success. The results are simply a factor of your guaranteed income which is healthy, your substantial investable assets and whatever other parameters you modeled.

Cheers
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: How much do you trust FIRECalc?

Post by dbr »

MrCheapo wrote: Wed Jul 14, 2021 8:37 am
Silk McCue wrote: Tue Jul 13, 2021 8:23 pm
MrCheapo wrote: Tue Jul 13, 2021 8:16 pm Then I'm completely baffled by it. It says for a spend of 84K a year my $2.7M of investments over 40 years I'll end up with $10M on average?

Edit. I see that it assumes solely investing in the US equities market by default. Perhaps it assumes the last 20 years are a good barometer to the next? That would be great but unlikely!
Maybe you should read the main page and see what it says about the tool and understand it’s inputs and options.

Cheers
I did read through it and am still baffled.

I entered my full information very conservatively based on today's numbers (I'll retire in 5 years).

2.7M investments, (assumes I don't make any further contributions)
84K a year in pensions (assumes no pay increases for 5 years)
30K a year in SS collected at 67 (this assumes SSN will pay of just 60% on the amount due)

and then I played with how much I can spend for 45 years (by which time I'll be 100).

Turns out I can spend unto $175K a year (with inflation adjustments) each and every year for next 45 years and NEVER get a scenario where I'm out of money.

I can see my future self at 99 sipping cocktails and sucking caviar (because I have no teeth and can't eat steak) ...

I wonder if the scenarios it uses are heavily skewed to the recent 20 years which have been the greatest secular bull market of all time.
The scenarios are skewed if anything to horrible years for investment returns and inflation such as the 1960's, the last decades of the 19th century, and so on. The way the model works for a 45 year retirement the last scenario year that is used is a retirement starting in 1976. The effect of the last ten years bull market is pretty much not present in the FireCalc data yet. The 1982-2000 bull market is in there, of course.

I think 45 years is a pretty long retirement to run in FireCalc because almost all the recent data is not used and that would be the bias.

Note the methodology of FireCalc using a set of historical scenarios is the same as that used in the Trinity Study which first observed the 4% phenomenon, and FireCalc results are compatible with that.
Statistical
Posts: 571
Joined: Tue Jul 06, 2021 1:08 pm

Re: How much do you trust FIRECalc?

Post by Statistical »

Mike Scott wrote: Fri May 07, 2021 12:13 pm If you are saving half or more of your gross income, why would you be surprised that you will probably have a lot of money during retirement?
This. Saving half your income plus decades of compounded gains is winning the game. Then the OP also has two pensions on top of that and SS. You could win the game with just the savings the rest is just racking up the scoreboard.

I am surprised the OP is surprised. Also OP understand that tools like FIRECalc are looking to avoid running out of money in the worst case scenario. In a projection which avoids that 95%+ of the time, the median (likely) scenario is 20 years into retirement your NW will have grown to multiples of what it was when you started.
Last edited by Statistical on Wed Jul 14, 2021 9:59 am, edited 2 times in total.
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: How much do you trust FIRECalc?

Post by dbr »

Statistical wrote: Wed Jul 14, 2021 9:09 am
Mike Scott wrote: Fri May 07, 2021 12:13 pm If you are saving half or more of your gross income, why would you be surprised that you will probably have a lot of money during retirement?
This. Saving half your income plus decades of compounded gains is winning the game. Then the OP also has two pensions on top of that and SS. You could win the game with just the savings the rest is just racking up the scoreboard.

I am surprised the OP is surprised. Also OP understand that tools like FIRECalc are looking at the worst case scenario. The median scenario is 20 years into retirement you have 3.5x your net wealth on the day you retired.
I agree people may not fully recognize what planning tools like this are doing when they estimate "successful" conditions. Possibly the most useful output of FireCalc to examine is that chart of what the course of all the sample portfolios actually did. One may note that along with a couple of portfolios that run out of money 25 years into a 30 year run, there are tens that end up with a person dying with fabulous wealth. I am not aware of any approach to retirement that resolves this dilemma in a completely satisfactory manner for everyone.

The tool really should be used not as a go-no go nor as a fixed plan, but rather as a map of the terrain which one can use to make choices. In the end each investor takes responsibility for the consequences of their choices.
Dottie57
Posts: 12379
Joined: Thu May 19, 2016 5:43 pm
Location: Earth Northern Hemisphere

Re: How much do you trust FIRECalc?

Post by Dottie57 »

I’ve used Firecalc to reassure myself that I have enough for retirement. But it is more useful to know your actual expenses and be able to calculate your spend rate from the day of retirement until you start SS. One thing that I had to add in as expenses was income taxes and medical insurance. Start thinking about what type of accounts you want your assets in. Different types of accounts have different tax consequences.
MrCheapo
Posts: 1468
Joined: Tue Dec 22, 2020 2:43 pm

Re: How much do you trust FIRECalc?

Post by MrCheapo »

dbr wrote: Wed Jul 14, 2021 8:49 am Note the methodology of FireCalc using a set of historical scenarios is the same as that used in the Trinity Study which first observed the 4% phenomenon, and FireCalc results are compatible with that.
Thanks, and to elaborate you are saying the Trinity Study used the worst historical scenario? Does FireCalc explicitly say where/how they generate all these alternative scenarios? Or are they the scenarios in the Trinity Study.
randomguy
Posts: 11295
Joined: Wed Sep 17, 2014 9:00 am

Re: How much do you trust FIRECalc?

Post by randomguy »

MrCheapo wrote: Wed Jul 14, 2021 10:26 am
dbr wrote: Wed Jul 14, 2021 8:49 am Note the methodology of FireCalc using a set of historical scenarios is the same as that used in the Trinity Study which first observed the 4% phenomenon, and FireCalc results are compatible with that.
Thanks, and to elaborate you are saying the Trinity Study used the worst historical scenario? Does FireCalc explicitly say where/how they generate all these alternative scenarios? Or are they the scenarios in the Trinity Study.
All of the schemes look at hundreds of 30 year periods. The 4% SWR is the worst one. The average is more like 6%. And the best cases are a bit over 10%. All of the studies get slightly different results depending on if they look at periods that start at the 1st of the year, first of the month, or every day. And if they use S&P500, total stock market, intermediate treasuries, or corporate bonds. But they all come out with worst SWR right around 4% using historical data. I trust fire calc is doing the math right. How much the future resembles the past is impossible to say.

The OP in a pretty bad case is looking at like having 1.9 million (i.e. 100k/year savings, zero growth)+35k/year in a pension+ss in 10-15 years. Pretty easy to get to the low 100s in spending. In a more normal case you are looking at more like 3 million and 150k of saving. Obviously you will have a much better idea of where you are in 10 years.

On this board when talking about retirement, we pretty much only talk about the worst case. If you aren't careful that can give you a really pessimistic view. When people talk about 3% SWR and life it is important to remember we are taking a very conservative number of 4% (i.e. 95% of the time you could have taken out more) and then making it even more conservative by cutting it by 25%. F
rich126
Posts: 4475
Joined: Thu Mar 01, 2018 3:56 pm

Re: How much do you trust FIRECalc?

Post by rich126 »

With FireCalc's spending number, is that really "withdrawal" as opposed to spending since it doesn't include taxes in the calculations?

So if your spending is $60,000 you really need to estimate taxes and bump that up higher?
----------------------------- | If you think something is important and it doesn't involve the health of someone, think again. Life goes too fast, enjoy it and be nice.
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: How much do you trust FIRECalc?

Post by dbr »

rich126 wrote: Wed Jul 14, 2021 11:09 am With FireCalc's spending number, is that really "withdrawal" as opposed to spending since it doesn't include taxes in the calculations?

So if your spending is $60,000 you really need to estimate taxes and bump that up higher?
It's withdrawal and it's based on index returns. Included in spending is taxes and all investment costs.

That's why a person paying 1% AUM to an advisor and 1% in fund costs and taking a safe withdrawal of 4% can only spend 2%, and that 2% includes taxes which may also be too high if his investment managers have him in tax inefficient funds. It is not even inconceivable that investment costs reach 3%. The good news would be that our poor investor has retired in a year when 7% is possible and he can still get away with it.
User avatar
LilyFleur
Posts: 3499
Joined: Fri Mar 02, 2018 9:36 pm

Re: How much do you trust FIRECalc?

Post by LilyFleur »

One more reason I really like Schwab. Every year I have a complementary meeting with my advisor, and he runs my financial plan with any updates needed. This year we added in my new part-time job and down payments for houses and help with weddings for my children. (Schwab's planning software has the ability to input those types of "chunky" expenses in a particular year.)

The plan he gives me shows spendable income for each year up to 90.

I think it's wise to have several data points when planning retirement.

I've used FIRECalc, and I find it useful, but I wouldn't rely on it entirely.
User avatar
RickBoglehead
Posts: 7877
Joined: Wed Feb 14, 2018 8:10 am
Location: In a house

Re: How much do you trust FIRECalc?

Post by RickBoglehead »

MrCheapo wrote: Wed Jul 14, 2021 10:23 am @RickBoglehead


Your "spreadsheet" models just one average scenario. The whole point of FireCalc is to explore many different scenarios that deviate from the average.

The spreadsheet, which is not rocket science to create (hence the term), allows one to step back from the smoke and mirrors and see what simple math gets you to, which then validates that the scientific models are giving you accurate information. That can give one comfort that the models, which they may not understand in part, or at all, are at least giving perspectives around the average based on past history.
Avid user of forums on variety of interests-financial, home brewing, F-150, EV, home repair, etc. Enjoy learning & passing on knowledge. It's PRINCIPAL, not PRINCIPLE. I ADVISE you to seek ADVICE.
rich126
Posts: 4475
Joined: Thu Mar 01, 2018 3:56 pm

Re: How much do you trust FIRECalc?

Post by rich126 »

dbr wrote: Wed Jul 14, 2021 11:15 am
rich126 wrote: Wed Jul 14, 2021 11:09 am With FireCalc's spending number, is that really "withdrawal" as opposed to spending since it doesn't include taxes in the calculations?

So if your spending is $60,000 you really need to estimate taxes and bump that up higher?
It's withdrawal and it's based on index returns. Included in spending is taxes and all investment costs.

That's why a person paying 1% AUM to an advisor and 1% in fund costs and taking a safe withdrawal of 4% can only spend 2%, and that 2% includes taxes which may also be too high if his investment managers have him in tax inefficient funds. It is not even inconceivable that investment costs reach 3%. The good news would be that our poor investor has retired in a year when 7% is possible and he can still get away with it.
Thanks.

In my case I can actually start with a fairly high withdrawal (7.5%) because of a pension that starts at 65 and social security at 65/70 that will greatly reduce the SWR (1-3%) as I get older. This is with a 100% back test success rate. I also left out a small pension my wife gets at 65.
----------------------------- | If you think something is important and it doesn't involve the health of someone, think again. Life goes too fast, enjoy it and be nice.
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: How much do you trust FIRECalc?

Post by dbr »

rich126 wrote: Wed Jul 14, 2021 11:34 am
dbr wrote: Wed Jul 14, 2021 11:15 am
rich126 wrote: Wed Jul 14, 2021 11:09 am With FireCalc's spending number, is that really "withdrawal" as opposed to spending since it doesn't include taxes in the calculations?

So if your spending is $60,000 you really need to estimate taxes and bump that up higher?
It's withdrawal and it's based on index returns. Included in spending is taxes and all investment costs.

That's why a person paying 1% AUM to an advisor and 1% in fund costs and taking a safe withdrawal of 4% can only spend 2%, and that 2% includes taxes which may also be too high if his investment managers have him in tax inefficient funds. It is not even inconceivable that investment costs reach 3%. The good news would be that our poor investor has retired in a year when 7% is possible and he can still get away with it.
Thanks.

In my case I can actually start with a fairly high withdrawal (7.5%) because of a pension that starts at 65 and social security at 65/70 that will greatly reduce the SWR (1-3%) as I get older. This is with a 100% back test success rate. I also left out a small pension my wife gets at 65.
Yes, it is important to note that what all of these models add to the basic observations of a Trinity type study is simulations of real life in which income streams come on line, expenses may be added or removed, money may be taken from or added to the portfolio, and so on.
BogleFan510
Posts: 1039
Joined: Tue Aug 04, 2020 2:13 pm

Re: How much do you trust FIRECalc?

Post by BogleFan510 »

I think 'trust' is the wrong term to use. The right question to ask IMHO is 'How well do I understand the FIRECalc Model?

No model pretends to actually predict the future. All use past data to create scenarios which may or may not occur. The probability figures are backwards looking models of what succeeded in the past.

So what one is 'trusting' is that we can model the future and that the future investment return patterns will be consistent with patterns of the past.

This is not certain, but can be a reasonable assumption to make for planning purposes.
rich126
Posts: 4475
Joined: Thu Mar 01, 2018 3:56 pm

Re: How much do you trust FIRECalc?

Post by rich126 »

BogleFan510 wrote: Wed Jul 14, 2021 11:40 am I think 'trust' is the wrong term to use. The right question to ask IMHO is 'How well do I understand the FIRECalc Model?

No model pretends to actually predict the future. All use past data to create scenarios which may or may not occur. The probability figures are backwards looking models of what succeeded in the past.

So what one is 'trusting' is that we can model the future and that the future investment return patterns will be consistent with patterns of the past.

This is not certain, but can be a reasonable assumption to make for planning purposes.
Like many, I try not to trust any one thing and even in the sims I've been running, I've only used my core account, pensions, social security as inputs but left out another 20-30% of investments that I consider my emergency/black swan type event of funds. I think anytime after 60, I should be fine to retire.

It is nice that people have developed various tools to play with (firecalc, portfolio visualizer, Fidelity has a retirement planner thing, etc.).
----------------------------- | If you think something is important and it doesn't involve the health of someone, think again. Life goes too fast, enjoy it and be nice.
Post Reply