Show HN: I built a simulator for personal finance
projectionlab.com
projectionlab.com
This idea of FIRE has been spreading beyond techies/Internet nerds and into other highly paid professionals like doctors. [0]. The main difference is that non-programmer professionals generally don't get a thrill out of min/max-ing financial projections.
Such people might be willing to pay a few hundred dollars a year for someone to generate this kind of analysis with them and simply execute on it each year. Similar to how people pay for CPAs. My employer actually pays Goldman Sachs to provide a similar service for its employees (no commission, fiduciary advisors) but their tools suck compared to this.
[0] For example https://www.whitecoatinvestor.com/, https://www.physicianonfire.com/
At a certain net-worth, they'll start calling you on a regular basis trying to get you to do the same.
It's all pretty annoying. Perhaps do that, and after a 6 months of so of calls and prompts, then offer to stop calling / prompting for $10 a month.
IIRC it's robo-investing. I got a pitch from them, and it was mostly "high expense ratio funds are bad, we put you in low cost funds with expense ratio of < .10% to save you money".
"OK... but your expense fee is .90% of funds under management - this puts me at approximately 1%/year in fees."
"But we're actively rebalancing 2x per year and we do tax loss harvesting".
I was close to giving them a small $ to manage/test with, but backed out.
"But we're actively rebalancing 2x per year and we do tax loss harvesting".
Beat them all: Put all of your money is an ETF that tracks S&P 500 index. iShares IVV and Vanguard VOO are excellent. Expenses are less than 5bps per year! After 3/5/10 years, you are nearly guaranteed (statistically) to be ahead of the investment advisor when including expenses.
I agree 100%. I think it is a good idea to pay for at least one hour of financial advice per year. Do not buy any products recommended if they receive a direct or indirect commission. An outsider who studies the financial situation of others for a living will usually have helpful advice, even if limited to: "This is good." or "Have you considered...?"
Another bundling is to help users execute on their projections, similar to how https://www.helloplaybook.com does with tax advantaged investing.
More generally, I think there is an opportunity to market the product not as a tool (“ProjectionLab builds the best financial projections”) but as a solution to a problem (“ProjectionLab lets you retire early/plan to pay down your debt/buy a house”).
I know you’re doing this as a side project and maybe what I’m suggesting is too startup-y. Best of luck either way, it’s a really great product.
Agreed about turning off existing users, but you can carefully orchestrate this to help users find appropriate advisors as a completely opt-in service. I think you can find a scalable monetization model in match making. Not sure what your plans are, but yeah unlikely you'll be able to pull it off in solo mode.
Totally impressed with what you have built.
I am looking to build a similar tool for engineers who are switching jobs and have multiple job offers. Essentially a financial projection of options 1, 2 and 3 taking into account inflation, location and salary growth.
Happy to brainstorm and collaborate if you are up for it.
the same goes for CPAs - the moment they see that your tax situation becomes too complicated or a potential liability or they can't make a good profit (read: cookie cutter tax situation which can be handled by fresh grads but can be invoiced at partner rate) they tell you to go elsewhere.
Anyway, my advisor works via a larger "network" company that provides their white-label tools to run monte carlo sims, etc and game out retirement. None of them are this accessible / slick. Definitely something to this.
1. You've got RSU's (at first it wasn't clear to me where this would be addressed, but I think it made sense once I came across it) but I don't see a dedicated strategy for dealing with ISO/NSO's. That'd be helpful for the tech community especially.
2. I spent the 10-15 minutes working on a plan knowing that I'd need to upgrade to pro for every feature, but it seems like I need to upgrade to save my plan. I ended up not paying because I don't have enough time right now to fully evaluate this, but I might look back later. It would have been great if I could have entered my email and saved my plan without paying, and then had to later pay to access it. This gives you the benefit of getting my email address, and then sending me an email to access my plan so that I've got a second touchpoint to your product when I check my email in the future.
Looking forward to using this more in depth in the future!
But overall I think you're right on both points. I've been wanting to add better support for modeling options for a while; for anyone who'd like to bump priority there, feel free to upvote this item in changemap: https://changemap.co/projectifi/projectifi/task/5735-better-...
For #2, I'll do some thinking to gauge level of effort in rigging up a mechanic like that with the current stack. Losing data sucks, and I should take more steps to reduce the chances of that happening to anyone in the onboarding funnel.
I liked this enough where I made a model during my lunch break, then remade it when I got home, and then remade it a third time after I had accidentally pressed the "back" button on my mouse talking to someone on Discord, taking me back to the landing page and losing all my data.
After that last one I decided to at least get the free trial so I could save, but I had already planned pay for Premium so I wasn't too bothered.
- You can add a "dependent expense" (have a kid) at a certain age
- You can run Monte Carlo simulations of your portfolio performance
- Expenses can be configured to increase linearly, with inflation, up to an amount, or using a custom function that the user can input graphically
- Adding a house purchase automatically cancels out your existing rent expense if you indicate that it will be your residence
- Easily visualize all of your projected earnings from portfolio growth and losses due to inflation and taxes
As an avid YNAB-er, this is heaven.
The top 5% salary in the UK is £81,000 -> after tax is £54,817. £4568 a month. Say you spend £2k a month. That's £2568 for investments per month. Using the calc, it seems you would be able to retire at age 50.
I quickly checked the US salary percentiles and the top 5th percentile of £81,000 in the UK is equal to the top 15th in the US. Thats 3x ceiling you could climb through not to mention taxes in the US are much much better?
Am I crazy?
https://www.gov.uk/government/statistics/percentile-points-f... https://www.thesalarycalculator.co.uk/salary.php https://dqydj.com/average-median-top-individual-income-perce...
I suspect there are other high paying jobs/markets with similar restrictions. It'll be interesting to see how the dust settles over the next few years as geographical location within a country and without impacts the labor market.
However, personal experience and friendship group makes me think that US tech salaries have only accelerated over the last couple of years. I went looking for some data to back up the idea to tech salaries have accelerated, but https://spectrum.ieee.org/engineer-salary was the best I found, and was less dramatic than I expected.
This is complicated. Taxes in the US are different. For example, a lot of higher earners are concentrated in California and New York, which have fairly high taxes; in New York or San Francisco a single person making $105k would take home about $75k (slightly less in SF, slightly more in NY) or about £3k more. I would suspect (but do not know) that the UK government pension and health-care is at least slightly better than the US version.
On top of that, people of a given income level spend more in the US (either because things are actually more expensive, or people in the US are more spendthrift; probably a bit of both), so if you come to spend like a typical American[1] your savings won't necessarily be higher in the US. Bay area rent, in particular is insane (rent alone would be more than the £2k a month you proposed for expenses, though $105k/year in the bay area would be a very low salary for a bay-area tech worker).
Now, you can do better. I have a friend who works as a DBA in a very inexpensive part of the country; he makes more than the $105k per year. In fact his yearly salary is higher than what he paid for his house. When remote became normal during COVID a lot of people moved to cheaper parts of the country while keeping their high salaries. Also, everything else being equal, a higher top-line will always provide more opportunities for savings
1: Before you dismiss this completely, there are real social costs to spending less than your peer group. IMO they are worth it, but they should be factored into any planning you do.
That's income taxes; states aren't that different when you actually count "taxes" ie property, sales, income, and whatever else is left over.
Texas is theoretically a little worse for the middle class because property taxes are much higher. It's better if you're rich though.
I know they are higher in relative terms, but are they higher in absolute terms? I live in a single-family house in SoCal and pay over $12k/year in property taxes.
[edit] absolute <-> relative
Average property tax rate for the Houston area is 2.31%. There's lots of things that come into the equation (homestead caps, exemptions, etc.), but for an average priced home of, say $350,000 you'll be owing ~$8,000 in property taxes.
The trickiest part is that your property values are reassessed every year, so for a long time our tax bill has just gone up, whether your income has or not.
I'll still take Texas economics over California economics, but everywhere has its tax pros and cons.
I kind of wish it was that way here; the previous owner of the house I am in was paying $1400 per year in taxes, after purchasing it (pre-covid price boom) we pay $12000 in taxes (note the extra zero). I get people not wanting to be priced out of their housing by taxes, but:
1. If your property actually went up by a large fraction in value, so did your equity, and if you have a mortgage, then the equity goes up faster than the value due to leverage.
2. MIN(2%, inflation) is an absurdly low cap. Since it is applied annually, even if inflation averages 2% per year, the property tax increase will lag inflation.
Ultimately when far more people want to live in a place than there is housing, bad things happen, and Prop 13 and rent control just try to shift the advantage from "has lots of money" to "got there first" with lots of annoying second-order effects.
My previous residence in particular was "interesting" because under the now-current zoning rules it would only be allowed to be 1/2 the square-footage, but as long as it stays as the current floor-plan it's grandfathered in. If you make it illegal to improve the usage of a property, then an LVT has no effect!
Also the most expensive states by this metric are almost all in the Northeast, which matches my intuition.
1: https://wallethub.com/edu/states-with-the-highest-and-lowest...
Source on the all-inclusive assessment? I am familiar with Texas and California and New York. One of these things is not like the others.
Otherwise, it's too expensive for cheap fire-minded folks like myself. Adding a subscription to this would push back our FIRE dates :)
Persistent storage (even if it's just a text file I have to copy-paste every time) for $12/ a year would add enough value to be worth it for me, and for me to replace my spreadsheet with this.
I think I have a mental friction on paying monthly based on how much I interact with the software, so it might be me.
I'm not sure how I feel about the persisting being a premium feature yet, I have to play with the software a bit more. It might be that a tier in-between with just persisting would solve the problem, but it could just be that I haven't realized the software' s full potential.
You get a lot of points for making it work well with the phone, I was really surprised.
What made it surprising is that usually diagrams and tables display horribly on phone in most apps. This app is centered around those and works well, which is literally amazing.
I put in some serious numbers and it looks great. Now I'm a bit concerned for the last 3 years of my life, lol
I will definitely want the "tax adjustment" feature, I'm very happy you provided the "lifetime" option.
Do you have a "plan" in case everything goes wrong and you don't have enough funds? I wouldn't want to lose all the planning. That being said, the software is incredibly good, I don't expect that to be a thing.
However, having that sheet meant I could enter in all my accounts, assets, debts into ProjectionLab in about 30 minutes, and spent the rest of the time fiddling with the projection tool, which far exceeds what I would have built.
The cash flow priorities I simply haven't found in another tool, and was one of the big reasons for me wanting to build my own, since I get about 50% of my total comp annually via RSUs vesting, bonuses and share sales, and route those directly into e.g. my mortgage.
Is there any chance to have the Pay Extra goals be annually (I mean, I can do N/12, but annually probably is more accurate from a charting perspective).
I've never forked out my credit card so quickly, well done.
And to your point on data entry, I should probably go back and look at all the forms and make more of the inputs multi-type so you can always specify in whatever frequency you like.
1. It would have been great if I could make other variables used in Monte Carlo simulations than the market. It would allow one to model uncertainty. For example, an event could happen following a random distribution (kid arrival, move between 3 and 5 years), or a future salary could follow some normal distribution, etc.
2. The main dashboard for a plan is GREAT. UI/UX-wise, considering the number of dimensions to factor, I find it surprisingly simple to iterate over a plan.
3. Having the option to show expenses on the main plan chart would be great. Expenses are a big part of optimizing personal finances.
4. The plan comparison may deserve a different UI than the simple plan analysis. Seeing differences on the right is not completely intuitive.
5. I'm currently living in France, and we have some accounts with no amount limit but with an annual transfert limit. For example, you cannot add more than 20K€/year to the account. I believe it's currently not possible to set a cash allocation such that it maximizes transfert amount/year instead of overall amount.
6. Related to 5., but we cannot withdraw from that account without paying taxes until retrieving, then it's tax-free up to some amount per month. Maybe similar to a 401k. But it's impossible to set that up on a custom account.
7. I have set that model up, still in trial, but now I'm really scared that my browser might bug or hit refresh by habit and loose everything. The find the saving feature to be really harsh. It'd be nice to find some alternative so that it can be saved, and maybe retrieve it after subscribing? Or limit the number of parameters/events that can be added to the model? Or show in much simplified mode. In my case, I'd like to show my partner and discuss it before subscribing: I want to make sure it contains enough features needed to model our finances.
Really good job overall, I'll certainly follow your work!
1. Definitely looking to add more degrees of freedom to Monte Carlo scenarios over time. A couple examples: https://changemap.co/projectifi/projectifi/task/5736-flex-on... https://changemap.co/projectifi/projectifi/task/5926-allow-s...
3. Have you seen the Plot Builder? (open plot selector and hit new). Expenses are one of the metrics you can use.
4. Totally agree. Eventually I want this to show more than the deltas in the yearly summary, just haven't gotten to it yet.
5. I actually just added some more options for savings type goals to the early access site here: https://projectifi-201a2--dev-xsmyy9im.web.app. let me know if that helps at all.
7. I'd like to make some improvements here too. Losing data sucks, and I should take more steps to reduce the likelihood of that ever happening. I'll see if I can come up with a good approach that's not easy to take advantage of.
3. I haven't seen that, wow! It can do a lot.
4. Makes sense. Just switching between plans works, anything more is nice to have, not must have.
5. I can see a Yearly Contribution Limit, that rocks! It's exactly that.
7. Good luck with that! I read you got couple ideas and remarks on here, I'm sure there is some solution ticking most boxes.
I'll probably be a customer very soon! Good luck and again: well done.
Coincidentally, I've just finished a year-long (side-)project building a similar app for a friend.
Similar, I say, if you only read the one-sentence description of what they do. ProjectionLab is 100x prettier, smarter and more featured than anything I could have done.
You have my full admiration, and perhaps my subscription :) I have to come back to it on the weekend, on the computer (although it looks/works fantastic on mobile.)
I’d love to model ranges of uncertainty on, e.g., inflation, recessions, major events, etc. — I’d love to be able to play out scenarios around stock market dynamics, housing market dynamics, and more — and this interface is beautiful.
Part of me wants to make some mocks for you, in the hopes that you’ll add some of this… :)
Have you seen the Advanced editor yet that you can use to model how various things change over time? That can be helpful for playing out some specific scenarios. And the compare tool might also be worth checking out. I think I see what you mean about ranges of uncertainty though; I can imagine some cool UI possibilities.
Two things I noticed, one is a minor visual bug and the other is a potential win (unless I missed it entirely, in which case disregard):
- it'd be nice to have a feature to do one-off expenses with greater ease, applied to a subset or all plans right off the bat, the idea being I'd like to see the effect of a spontaneous / fun buy and its greater impact on my plans at that moment (thinking really casual FI where me being off 1-2 years isn't a major dealbreaker);
- the promo image under the `Plan Together` summary for mobile has the app layout _behind_ the device notch when it handles it correctly.
And thanks for the heads-up! That's what the mock-up generator I used produced and I don't have an iPhone so I wasn't 100% sure how it _should_ look haha. I'll take a note to go back and adjust.
Would love some integration with [plain text accounting](https://plaintextaccounting.org/) so that I don't have to re-document all my finances though, even just a simple `ledger equity` import.
- The initial import can be done pretty easily, `ledger bal` reports my current balances, so for example I can do `ledger bal Assets` and it will report back totals on my `Assets:<Bank>`, `Assets:Retirement:401k`, `Assets:Retirement:Roth 401k`, etc.
- You could potentially get a very in-depth "progress points" import with `ledger reg`, which just shows the history of my ledger. `ledger reg Assets:<Bank> --monthly --collapse --total-data` gives me a monthly tally of my savings since forever.
- It would be possible but harder to infer future events & spending I think. I can somewhat figure out how much I spend monthly looking at my data, but you'de need heavy processing to get anything useful out of that.
I think maybe a more reasonable request rather than supporting ledger is to just support a general purpose plaintext import. The above `--total-data` command for example just gives me:
2018-06-01 10000
2018-07-01 11000
...I like simple projections. Most calculators out there on the web suck or are advertisements. They account for too much and I just need PERT but for various asset classes.
I tried doing the walkthrough and it told me I'd go bankrupt. I think there's a bug somewhere as I have 0 debt and over a mill in liquid assets meaning net worth positive.
The sandbox was a bit confusing with the seeded data and I wasn't able to make much sense so I had to start over with the walkthrough.
I would pay a few hundred dollars a year for a simplified version of this where I can just play around with fixed percentages based on my expected returns for different asset classes. Right now it's a bit complex for me and I'd just go back to pen/paper & excel.
If I could show something simple like this to my friends and family, I'm sure they would enjoy it too.
To simulate different asset classes, you can always add different account types, rename them, and set custom growth rates / change-over-time plots.. but I agree that something more built-in could make sense, e.g.: https://changemap.co/projectifi/projectifi/task/5670-asset-s... ... though I suppose this moves in the direction of adding even more features rather than simplifying haha.
Not helping my "11 months till 30" anxiety/depression. I don't know why it's messing with me so much. The internet loves to make is seem like your 20s are where it's all at but I spent it building a career rather than partying or dating... Leaving the "young adult" phase and just entering "adult". Also mega pissed that I started doing better in 2019 (international travel, lots of time spent with friends) and then the pandemic came and turned the world on its head and trapped me at home for the last bit of my 20s.
Oh, and paragraph 2 hit me hard. The countdown to 30 is... happening.
Overall, awesome tool though!
Made me a bit sad to realize that even with a tech salary, most bay area houses are out of reach of a single income family unless you're an L5 at Google or a similarly paying company.
And yikes, happy to be east coast for now =)
Sure, you'll change your priorities and you'll have a few years of extra stress while the kids are small.
Then consider that in your final decade or so, your options are extremely limited because of severe mental and physical decline. 1 in 9 adults will spend one or two of their final decades in a long goodbye due to Alzheimer's alone.
But the decline starts earlier, mental speed is high until age 60. This will affect your job and career prospects, and even if it doesn't for you, employers (esp. in tech) think it will and discriminate based on age anyway (roughly ~45yo based on some articles I've read). If you're a blue collar worker, a career of hard physical labor starts to hit a wall. If you're a woman, fertility starts declining (and childbirth health risks dramatically increase) fairly early (~30) and dating desirability follows rapidly.
Those are just the terminal issues, psychologically your life path more or less calcifies as you age. There's no age cutoff but, for example, it's said that the longer you stay in corporate work, the more likely you'll stay there and the less likely you'll succeed at entrepreneurship. I'd speculate there's maybe one or two more major life pivots in a 30yo.
Most of these factors don't exactly make life unbearable, but after a fairly "young" age (around 30 I suspect) a lot of doors start closing and it only accelerates from there. Depending on what "prime" or "best" means to you in terms of life goals, realistically you have much less than 6-7 decades to exploit them. I generally disagree with this attitude of oh don't worry you have plenty of time, life is much shorter than you think.
Nevertheless, once your doors are shut, there's little use in ruminating over the missed opportunity.
I think chronic conditions are also hard to plan for in your head, so some preset stats-based options for medical costs could be helpful.
No one intuitively imagines it'll happen to them but something like 60% of 65yo+ adults managed 2 or more chronic conditions, which are conditions like heart disease, diabetes, strokes, and of course cancer. It's more likely than you think. Stuff like cancer won't just kill you sooner, it'll bleed you dry towards the end of your life as you attempt all manner of surgeries and treatments to delay the end.
One drawback that occurs to me about the idea of projecting lifespan/healthspan is that with the data you'd need, you get into PHI territory pretty quickly right?
[0] https://github.com/MikePiper/open-social-security/blob/maste...
My attitude is certainly not of the 'oh don't worry you have plenty of time' because we really don't have that much time. But if the view is that in an 80 or so year life span, you got to get in 8-10 golden years as a healthy American, oof, that is pretty tough, and I probably won't change that mindset.
I "should have" never allowed myself to become obese in college. I "should have" lost the weight in my early 20s. I "should have" focused on fitness during the pandemic. But that's the past. Yesterday would have been a better choice but tomorrow is a worse choice than today ¯\_(ツ)_/¯
Got in a funk a few months ago and decided to make some changes. 5 months later I'm down 30 pounds. Fell off the bandwagon after Christmas but I'm pulling myself back onto it. Transferring offices to be near my friends, many of whom are fitness and health nuts who have all told me they're going to drag me kicking and screaming into the wonderful world of actually taking care of myself.
Who knows, maybe I'll actually feel better in my 30s than my 20s as a result. Not that I feel bad now. Even though I'm a 6'4 blob I can still go out and take a 6 mile hike without any pain or trouble.
Sounds like you know yourself well - which is a great start. I believe in you.
- https://mui.com/material-ui/getting-started/learn/
If this is not a bug, then the modal is confusing. The labels inside the modal don't correspond to anything on the dashboard. The investment modal should have a field for each investment, rather than some random fields that don't clearly map. If I click "Cryptocurrency" and a modal with no mention of it shows up, I'll assume the software is broken.
I'd also suggest having less memeish simulation descriptions for younger demographics. It adds friction for the average user when they think "well, I'm none of these things, what do I pick". Might also be taken as condescending. Comparatively, the older demographics all had very simple descriptions which would have mass appeal. Comes off like the developers don't take this demographic seriously, or don't understand how to market to it.
Cool looking dashboard, but crazy amount of UX friction makes it hard to use. I'd love to see it when it's tuned up.
Edit: which things jump out to you as meme-ish? The plan names in some of the sandbox examples? Milestone types? All of the above? lol
For the memeish stuff, it was under the example personas. While its fine to include some of these (ie: FIRE, looking for part-time), it'd be nice to have some examples for the middle of the road. A lot of what you mention isn't even on the mind of a lot of early career folks.
Something like "young professional, focused on building financial stability" or "trying to create more disposable income". Something that would have the same mass appeal as "has a defined-benefit pension, planning for standard retirement age" does to the older demographic.
Adding an export, with yearly/monthly resolution and customizable columns would be trivial probably.
https://projectionlab.com?help=enter-data
https://projectionlab.com?help=data-security
https://projectionlab.com?help=am-i-the-product
But more directly to your point, if you enable cloud sync, that uses Google Firebase. Firebase encrypts at rest and in transit and has the usual certifications, but the project owner does have an admin UI for Firestore. If that's a concern, there are always the other data persistence options as alternatives: localStorage only, and/or importing/exporting copies of your data to local JSON files.
Perhaps it would also be wise for me to look into integrating with something like GCP cloud kms, that way maybe the user could supply their own encryption key client-side?
Feature request: make this an Electron app so I can actually use local storage, and feel safe that my data (and work) is safe and won't be wasted.
It doesn't involve linking your financial accounts, you don't have to make an account to try it, the free version has a lot of features, and there's a sandbox mode if you just want to see how the interface works.
Last year I posted a prototype here (back when it was called ProjectiFi): https://news.ycombinator.com/item?id=27844194, and the early feedback from the HN community was extremely helpful! I've put in about 1,000 more hours of dev time on nights/weekends since then, and it's pretty much a whole new app. Here are just a few of the things you can do with it, many inspired by the HN commentary:
- Build detailed and flexible plans for your future that go beyond the standard online retirement calculators
- Backtest on historical data and run Monte Carlo simulations
- Model international scenarios with various account types and tax estimation presets
- Experiment rapidly: the simulation engine runs in your browser and doesn't need to send your data server-side
- Control how/where your data is stored: cloud sync, localStorage only, or manual import/export, (client-side by default, no persistence in the free tier)
- Plan for goals like achieving financial independence, taking time off for travel, home ownership, starting a rental empire, etc.
- Create granular models for how accounts/income/expenses/inflation/etc change over time using interactive plots
- Build dynamic configurations using milestones that support multiple criteria and conditional logic
- Create custom plots to visualize the metrics you care about
- Plan separately or as a couple
- Create and manage client accounts with the Pro version
- Track progress over time and see it overlaid on top of your projections
- Cross-compare between different plans, or stage + analyze multiple changes within a plan (and revert if desired)
- Choose your own icons to personalize things
- And a lot more; for a full breakdown of everything that's new, see the version history
If you feel like checking it out, I would love to hear what you think! Most of the functionality is free (minus data persistence), but for anyone interested in upgrading to Premium, you can use coupon code "HN-10" for 10% off any plan :)
One thing I couldn't work out how to do was to add one off purchases once you had already created a plan. Is that possible?
I've noticed in this thread a few other requests for import and I'd like to also chime in with that request.
I have a unique situation when it comes to income where I can reasonably predict my income overtime because of a collective bargaining agreement. Being able to import a table of income/year that I've already generated would save me a ton of time!
Thanks again for the amazing work!
One bit of feedback:
I found the timed accordion about your product to be distracting while I was reading. I felt pressure to read quickly and concentrate, which I’m guessing was the intended effect.
However, I’m a parent of two young children and have a slightly intense job, so my days are filled with time pressure already.
The added stress to read something when I’m trying to explore your content at my own pace produced a net negative experience for me.
One way you might resolve this is by giving users some sort of a “pause” button or opt out.
The reason I spent the time to give you this feedback was because I recognize how hard you must have worked and appreciate the effort you’ve obviously put in. Nice work!
That said, my go-to for FIRE calculations/simulations is https://www.firecalc.com/index.php - and I don't see that changing anytime soon.
The FIREcalc UI is very clunky compared to ProjectionLab but once you get used to it, IMO the options are a little more straightforward to understand and set. And FIREcalc is free and makes it easy to "save" your settings for future tuning/revisiting by generating a link with URL query parameters containing your setting values - I'm good with that approach.
ProjectionLab does provide more granularity for specifying investment accounts that are taxable vs tax-deferred vs tax-free, which allows it to factor in the effect of taxes in the projections. But I wonder if that's a level of complexity that's better left out of the calculations.
Effective tax rate (and yeah verily, taxes) - so damn complicated at least in the US. It will vary considerably over time even during your post-FIRE journey, depending on: your income (of course); your deductions/credits; changes in tax rates and codes; on the state(s) you live in (and how the state taxes social security, HSA earnings); etc, etc and so on and so forth.
Also many folks setup their asset allocation across all the various types of accounts to be as tax efficient as possible (https://www.bogleheads.org/wiki/Tax-efficient_fund_placement), and AFAICT there's no way to reflect that in ProjectionLab currently.
All that's a long way of saying that I appreciate that FIREcalc doesn't try to factor in taxes. It's easier to get my head around just allocating a certain amount for taxes in the yearly spending number, maybe based on my tax history or some coarse tax estimation calculator, maybe using an expected case and worst case tax amount in the yearly spending number and then running the FIREcalc simulations and comparing - without trying to account for the effect of taxes in a more granular way, given the inherent squishiness of taxes. YMMV of course.
Anyways, nice work on ProjectionLab, I can see where it would be a valuable tool for those who are trying to get a handle on their current finances and FIRE goals and may be a little less comfortable using tools like FIREcalc or home-grown spreadsheets (I geek out on this stuff so naturally I use both!).
And even if it's an advanced option, as mentioned before there be dragons with taxes that makes me kind of shrug on the utility. I guess it does help to remind folks that taxes are a consideration in their FIRE planning and drawdown strategy.
Folks might be surprised how low the federal effective tax rate could end up being after they FIRE, especially if all your income is passive - this is if you have have built up a nest egg that's enough, say 20-25x a reasonable (ie non-MMM levels of frugality) annual expense, but not like FU money amounts. And this is in the early retirement stage before Social Security, IRA withdrawal eligibility and RMDs come into play. State income tax rate, on the other hand ... (but of course income tax is only one component of your overall state tax burden to consider along with sales tax, property tax, etc).
That said, you have to decide who your target user base is - if it's the hardcore optimizer FIRE folks, then yes, building in the capability to at least play those kind of what-if scenarios and maybe do optimizations will probably be important for that user base.
There is another group of folks just looking for a path to lower-case FI, maybe with early retirement, and not treating FIRE as the end goal but rather as a means to an end, that want to follow a not-so-complicated path that will give them that FI but not spend all their time obsessing the numbers. This excellent post Life On Fire – What's Next For Me, You and Us[0] by Vicki Robin of _Your_Money_or_Your_Life_ fame sheds some light on this other group:
"AT THE SAME TIME, I SOON REALIZED THAT FIRE IS ACTUALLY NOT MY TRIBE. We applied the same methodology to money but it seems for different ends. For me, FI has simply been the freedom to pursue a higher purpose – to grow spiritually, to learn, to create and to serve. While I’ve met a lot of people reaching for relevance in their lives, not just independence, it’s not what people obsess about. They obsess about taxes and investments. There are probably tens of thousands more in my frame, but of necessity investing the majoring of their time and attention into the “getting out” part of the journey."
If you target that group, then I suspect they'll be looking for less in the way of these advanced capabilities to optimize their finances and more in the way of what your app already provides.
I could see some middle ground for offering tools to help make decisions for specific questions / scenarios. I started to sketch out an example of what this might look like for "Should I do a Roth Conversion?" and it might look something like this:
- It would ask for estimated federal and state income tax rates (0% for some states), pre- and post-conversion.
- It would calculate estimated RMDs for different percentages of the projected Traditional IRA balance representing varying Roth conversion percentages (0%, 25%, 50%, 100%), that would be useful and there are a bunch of such RMD calculators out there.
- It could estimate the delta on estimated taxes due to estimated RMD and total the additional taxes due to RMDs up to your life expectancy age (assumes tax your fed and state tax rates stay the same, kind of a wildcard)
- It could estimate how much on average and in aggregate you'd pay in taxes if you converted 25%, 50%, 100% and of your traditional IRA to a Roth over N many years starting at age A (N and A user changeable)
- It could estimate your balances remaining at your life expectancy age broken down by account type ($X in taxable accounts, $Y in tax-deferred, $Z in tax-free) if you converted 0%, 25%, 50%, 100%.
I think that would have to be good enough, though I suppose you could add all sorts of bells and whistles in terms of charts and visualizations to make it easier to understand the data. The data provided would be helpful to make decisions based on various personal judgment calls and preferences: the importance of the highest possible tax-free balance in your estate to be passed along tax-free to beneficiaries; the importance of minimizing overall taxes paid over the various scenarios; your interest in either paying more taxes now or more taxes later; your comfort in doing the conversion on your own; your willingness to pay a professional to do the conversion for you if you don't feel comfortable doing it on your own; your willingness to move from a state with high income tax to a state that has no income tax which would maybe change the numbers enough to make it more reasonable to do a conversion given your other preferences. Those personal preferences would be tough to factor into an optimization calculation, though I suppose a questionnaire to capture preferences like the above (eg "on a scale from 1-3, how important is ...) could be used with a rule-based algorithm to guide the user to a scenario that fits their preferences best.
Whew, sorry for the long-winded example.
[0] https://yourmoneyoryourlife.com/life-on-fire/
EDIT: to improve the bullet formatting, first time trying "bullets" in HN comment!
Also, and maybe this is a non-goal, you could help me optimize - let's say I've set my living expenses to $X/mo and income to $Y/mo, and I've set a goal for financial independence by age A, the model could suggest things I could do to achieve that goal, say by cutting X down to X' or putting more of Y into a different kind of savings account. It might not be easy to do, but again, as a complete novice I'd really love to get a plan that didn't require days of work to set up.
For optimization, I could see that getting a little bit tricky, since for now I want to steer clear of anything that could be misconstrued as formal financial advice. Definitely understand where you're coming from though, and there may be some opportunities to add different kinds of optimizers (e.g. a button that runs the permutations for drawdown order and stops on the one with the best outcome).
There are some common tricks that I don't see how to model with ProjectionLab:
1. Do your tax-deferred 401k,403b,IRA saving in a high earning job, state, then move to a low or no income tax state to do the withdrawals in retirement.
2. Retire early so there is some time before taking social security payments to do Roth Conversions. I-ORP[0] turns this into a branch and cut linear optimization problem. User Indyhou at Bogleheads[1] has built a spreadsheet that uses a solver plugin. It may be possible to build a model in CBC[2] and compile it to WASM and run it in the browser.
3. After turning 63, watch the Roth Conversions to make sure you don't trigger IRRMA medicare surcharges.
4. Are you trying to stay under income limits for ACA subsidies? It's not quite the sharp cliff that it was, but can be important for some.
5. Are you trying to balance regular income and capital gains to take advantage of the 0% cap gains rates? You've got to plan ahead on your contributions to the taxable and tax deferred accounts for this to work. Jeremy at Go Curry Cracker has written about using this to pay $0 in US Federal Income taxes[3].
6. Paying full rate for health insurance will likely get you over the 7.5% limit for tax deductions.
7. Social Security claiming strategies can be complex for married couples.
A feature that would be useful during accumulation is life insurance planning for the death of a spouse.
The death of a spouse can throw a wrench in some of the strategies since the single tax bracket is much smaller. Tax law changes can also upset highly optimized strategies. So any highly optimized strategy should also have a monte carlo simulation around a spouse dying and tax law changes to understand what disruptions are possible and maybe accept a non-optimal strategy that is better in these adverse cases.
[1] https://www.bogleheads.org/forum/viewtopic.php?t=365518
[2] https://github.com/coin-or/Cbc
[3] https://www.gocurrycracker.com/go-curry-cracker-2020-taxes/
2-7: all great points; I'd love to add more strategies over time to help optimize scenarios like these, though I'll have to think carefully about what can be generalized and what is strictly US-specific.
Thanks for writing these up -- I'll be looking into them in more detail :)
I would be really interested to hear how this product is different - maybe there’s a different monetization strategy or a different marketing strategy?
Yes, I think the most important difference is that, as far as I can tell, you're looking for a bit of side income rather than trying to jump start a whole company.
Basically they had a few B2C customers who had a more pressing problem than forecasting, and the company provides an API for other companies to solve that problem for consumers as a feature.
Only criticism is that after playing with the sandbox data prior to signing up it stuck around and I didn't see an obvious way of removing it besides overwriting it with my own data. The plan settings was non-obvious for getting my data input started. I began with the Current Finances, rather than overriding the plan settings which made things a little odd.
I've been thinking quite a bit lately on what I could possibly hack together in a few julia scripts, where I can at least munge my own transaction data.
I'd never wind up with this level of polish on presentation -- visually it's very pretty. Manually inputting a bunch of details is such a chore though.
But beyond balance updates, it seems hard to avoid having a ton of input fields in an app where you build a detailed model of your future. It feels like there's a tradeoff between being able to build a super detailed model and having a comparatively frictionless experience where you don't need to enter much.
It's a good product so it's nice to see that it's still getting some love.
ii) Owning my data.
Currently, I have my own tool that gets balances from credit cards and some bank accounts. I save all the data in a random xyt format which I export/import to my web tool/app. I am looking to replace this tool with something but without giving my data.
How do you envision i) working in a self-hosted version? Something you wanted to code up on your own? And does that imply you'd expect the self-hosted version to be easily editable, e.g. the actual source code?
My current setup -> My web server connects to bank accounts every 6 hours and creates a xyt file which is encrypted and synced. I take that file and load my web tool to get the visualization, simulation etc.
My ideal solution is something similar.
Great work with projectionlab nonetheless. I have tracked the progress since it was called project fi or something.
Or if you're saying you think there's a bug, happy to help troubleshoot; I know there are lots of inputs haha
Aside from that, this kind of low-effort dismissal ("has/lacks X, closed the tab") plagues a lot of Show HNs, and it's not constructive. If you really feel so strongly against signing up, it would be more constructive to explain why you feel that way and suggest ways OP could improve their project. As it is this comment doesn't contribute to any meaningful discussion, it just comes off as rude.
Also, I feel strongly against signing up for stuff because I already get tons of spam marketing from sites that I have signed up. I also am very protective over who knows what about my money.