Big differences between FIRECALC and New Retirement
- Thread starter palomalou
- Start date
COcheesehead
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
You likely have some inputs at odds.
qwerty3656
Full time employment: Posting here.
- Joined
- Nov 17, 2020
- Messages
- 762
I had crazy results from New Retirement (as opposed to Firecalc and Fidelity).
Here is a post where I asked a similar question to yours
https://www.early-retirement.org/forums/f28/new-retirement-pessimistic-scenario-114840.html
Here is a post where I asked a similar question to yours
https://www.early-retirement.org/forums/f28/new-retirement-pessimistic-scenario-114840.html
I’m not familiar with that SW but I would expect that if they want you to pay for a license, they need to say your current plan is not financially safe.
If the free versions of the SW says your plan works, what incentive do you have to pay for the premium version?
I’ll stick with FIRECalc. It’s transparent, we know how it works, we know its limitations, and most importantly, we know how to interpret the results.
If the free versions of the SW says your plan works, what incentive do you have to pay for the premium version?
I’ll stick with FIRECalc. It’s transparent, we know how it works, we know its limitations, and most importantly, we know how to interpret the results.
walkinwood
Thinks s/he gets paid by the post
Does New Retirement use historical data to give you a probability of success? Firecalc does.
SecondCor521
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
All other things being equal:
Average returns will have the highest probability of success. Mostly because they ignore SORR.
Historical calculators (like FIREcalc) will be in the middle.
Monte Carlo calculators will have the lowest probability of success. Mostly because they can overemphasize SORR.
Below that, I guess, are various purveyors of doom who make assumptions most here would find extreme.
Average returns will have the highest probability of success. Mostly because they ignore SORR.
Historical calculators (like FIREcalc) will be in the middle.
Monte Carlo calculators will have the lowest probability of success. Mostly because they can overemphasize SORR.
Below that, I guess, are various purveyors of doom who make assumptions most here would find extreme.
Thanks so much, Qwerty! i feel so much better seeing your experience and that of others who parallel mine.
Dtail
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Good accurate post.
Alex The Great
Recycles dryer sheets
In a firecalc, there is an important entry called "Portfolio". I always had a hard time to understand how does it work. Someone's portfolio for example can be $1.1M one day and $1.3M other day whatever market conditions are. Especially if the majority is in stock. How firecalc results can be trusted in this situation?
COcheesehead
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
It’s just a starting point for the Monte Carlo analysis. You have to have a starting balance. It will then apply the random outcomes from that amount.
Last edited:
Alex The Great
Recycles dryer sheets
Yes that's correct. But this starting point greatly affect the outcome. I tried different amounts which I had today, yesterday and the day before yesterday and all these results has a significant difference.
SecondCor521
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
FIREcalc does not do Monte Carlo AFAIK. It's historical only.
TripleLindy
Recycles dryer sheets
- Joined
- Aug 30, 2022
- Messages
- 296
That’s the nature of a Monte Carlo analysis. It’s going to run through repeated random sampling to determine an outcome. It’s likely that whatever variance you had between days 1-3 was accounted for in the sampling, so just be aware of that if you choose to plug in a new starting point. In the end, it’s an academic exercise to provide guidance, but nothing more.
SecondCor521
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
There are two related questions:
First, FIREcalc is just applying historical patterns to your balance, AA, portfolio expenses, budget, and so forth. It's just math applied with computer code; it could be applying the math incorrectly, but the code and results have been reviewed quite a bit. I trust the code and the math based on this.
Second, FIREcalc is not providing any sort of guarantee about the future. Retiring before you die will always have some risk that cannot be eliminated. All you can do is look at your overall situation and the output of retirement projections you understand and trust (FIREcalc or otherwise) and decide if you want to work longer or not. Although there is risk to retiring, there is also risk to remaining at work.
Most questions like this regarding FIRECalc can be answered by spending a little time reading these two pages:
FIRECalc: A different kind of retirement calculator
FIRECalc®: How long will your money last?
FIRECalc: A different kind of retirement calculator
FIRECalc®: How long will your money last?
I believe NewRetirement use Monte Carlo. I have the paid version of NewRetirement.
Below are the rates in the free version of NewRetirement which you can't change. In the paid version, you are able to adjust. Maybe that is causing the differences.
These are the rates used as defaults in the Planner:
Rates of return
Optimistic: 5% Pessimistic: 2%
Work & Passive Income Growth:
Optimistic: 3% Pessimistic: 2%
Pension COLA:
Optimistic: 0% Pessimistic: 0%
General inflation
Optimistic: 2% Pessimistic: 3%
Inflation Data
Social Security COLA
Optimistic: 2% Pessimistic: 0.5%
Housing Appreciation:
Optimistic: 3% Pessimistic: 2%
Medical inflation:
Optimistic: 2.5% Pessimistic: 5.5%
Last edited:
Dtail
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
I believe that the "Random Performance" choice is effectively a Monte Carlo type simulation.
SecondCor521
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Right. I vaguely remembered it was there but didn't find it on a quick perusal.
I'm not a fan of MC, so I really don't pay that option any attention. Nice that it's there though for those who want to do things that way.
I had New Retirement (paid version) a few years ago when it first came out. It was decent, but there were some obvious problems. Steve, the owner, was always good at getting back to me with some suggestions.
For the fun of it, I pulled up the free version a couple of weeks ago. Apparently, I need to get a job and keep it until I'm dead as I'm nowhere close to making it based on the calculations. Both Firecalc and Fidelity Retirement Planner say I'm in good shape. I rechecked my inputs and can't see anything wrong. It wasn't worth my time to dig in any deeper, but something wasn't right.
Sunset
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
My sanity check was deciding if 3.5% of my savings was enough to live on for a year. Basically the 4% rule with a little safety factor built in.
PatrickA5.. I also have the paid version of NewRetirement. I also use the Fidelity Retirement Planner as well.
Which retirement tool do you like the best and feel is accurate?
Any others you would recommend trying?
The returns in the New Retirement (free version) look horrible at 5% in the best case average (before inflation adjustment) and even lower in the pessimistic case. So it is generating lots of incredibly poor sequences, presumably some that have strongly negative real returns - hopefully that doesn't happen!
I like Pralana Gold (paid Excel sheet) as a planning tool. It's super flexible, handles all kinds of accounts, has a good tax model, has historical and tunable Monte Carlo modeling, allows various withdrawal methods and so forth. A rare feature is it allows you to hold different asset allocations in different types of accounts (so allows preferentially holding bonds in IRA, stocks in Roth for instance). The owner told me he now has partner developing a web based version, hopefully they'll have that some time next year.
As an aside, I think it's difficult for developers to make a living building good tools as so many folks gravitate to less powerful but free tools. I never quite understood why people would handicap their understanding of their finances by refusing to pay a few $, but that's human nature I guess.
I like Pralana Gold (paid Excel sheet) as a planning tool. It's super flexible, handles all kinds of accounts, has a good tax model, has historical and tunable Monte Carlo modeling, allows various withdrawal methods and so forth. A rare feature is it allows you to hold different asset allocations in different types of accounts (so allows preferentially holding bonds in IRA, stocks in Roth for instance). The owner told me he now has partner developing a web based version, hopefully they'll have that some time next year.
As an aside, I think it's difficult for developers to make a living building good tools as so many folks gravitate to less powerful but free tools. I never quite understood why people would handicap their understanding of their finances by refusing to pay a few $, but that's human nature I guess.
Koolau
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Heh, heh, and in keeping with our reusing-dryer-sheets mentality, FIRECalc is FREE!
Firecalc, Fidelity, New Retirement and Quicken Planner are the only ones I've used. I think I like Fidelity's the best. I stopped New Retirement a few years ago.
I might look into some new ones now that DW is planning on retiring early next year and we'll be squarely in the draw down phase.
daylatedollarshort
Give me a museum and I'll fill it. (Picasso) Give me a forum ...
- Joined
- Feb 19, 2013
- Messages
- 9,358
Before we retired, I made my own retirement planner using a spreadsheet, with inflation and real interest rates as parameters. I used the Fidelity planner as a reasonableness check. The results came out fairly close, so I trust the Fidelity planner to be making fairly reasonable assumptions, at least as reasonable as future planning can be.
Last edited:
