Firecalc success rate comfort level

[lutorm]

For exactly the reasons Malcat outline, I think it's a lot more instructive to pick one of the dynamic withdrawal scenarios that never run out of money, like the "withdraw the lower of 4% of the initial stash, inflation adjusted, or 4% or the current stash", which sort of mimic how a real person would behave. Then look at what the minimum withdrawal rate was across the scenarios and see if you'd be OK with that number.

Instead of "you ran out of money in 5% of the realizations", it's a lot more instructive to know whether your $200k spend had to be reduced to $190k or $10k. If the "failure" is the former, you'd probably be OK with it. Not so much with the latter.

[Metalcat]

For exactly the reasons Malcat outline, I think it's a lot more instructive to pick one of the dynamic withdrawal scenarios that never run out of money, like the "withdraw the lower of 4% of the initial stash, inflation adjusted, or 4% or the current stash", which sort of mimic how a real person would behave. Then look at what the minimum withdrawal rate was across the scenarios and see if you'd be OK with that number.

Instead of "you ran out of money in 5% of the realizations", it's a lot more instructive to know whether your $200k spend had to be reduced to $190k or $10k. If the "failure" is the former, you'd probably be OK with it. Not so much with the latter.

My argument isn't even to pick a variable WR, my argument is that most people will naturally vary their withdrawal rate given ongoing market conditions that threaten to erode their principle.

I don't recommend putting faith in any output from a FIRE calculator. Although, it is very helpful to compare fixed WR to variable to understand the magnitude of difference.

I don't really recommend anything beyond being realistic about what any of the math actually means.

[lutorm]

For exactly the reasons Malcat outline, I think it's a lot more instructive to pick one of the dynamic withdrawal scenarios that never run out of money, like the "withdraw the lower of 4% of the initial stash, inflation adjusted, or 4% or the current stash", which sort of mimic how a real person would behave. Then look at what the minimum withdrawal rate was across the scenarios and see if you'd be OK with that number.

Instead of "you ran out of money in 5% of the realizations", it's a lot more instructive to know whether your $200k spend had to be reduced to $190k or $10k. If the "failure" is the former, you'd probably be OK with it. Not so much with the latter.

My argument isn't even to pick a variable WR, my argument is that most people will naturally vary their withdrawal rate given ongoing market conditions that threaten to erode their principle.

I don't recommend putting faith in any output from a FIRE calculator. Although, it is very helpful to compare fixed WR to variable to understand the magnitude of difference.

I don't really recommend anything beyond being realistic about what any of the math actually means.

I mean, there's some balance point between "f it, I'm sure it'll be fine" to "the calculator shows 0.64% risk of failure so it won't work". You need some tool to help tell you what the outcome would have been in different situations to give you an idea of whether you need to worry about a 2008-level recession or a 1930-level depression, or whatever. If you don't have that tool you're just as likely to react to minor market swings by excessively cutting spending as you are to thinking everything is fine until you've eroded your principal to an unhealthy point.

[Villanelle]

For exactly the reasons Malcat outline, I think it's a lot more instructive to pick one of the dynamic withdrawal scenarios that never run out of money, like the "withdraw the lower of 4% of the initial stash, inflation adjusted, or 4% or the current stash", which sort of mimic how a real person would behave. Then look at what the minimum withdrawal rate was across the scenarios and see if you'd be OK with that number.

Instead of "you ran out of money in 5% of the realizations", it's a lot more instructive to know whether your $200k spend had to be reduced to $190k or $10k. If the "failure" is the former, you'd probably be OK with it. Not so much with the latter.

My argument isn't even to pick a variable WR, my argument is that most people will naturally vary their withdrawal rate given ongoing market conditions that threaten to erode their principle.

I don't recommend putting faith in any output from a FIRE calculator. Although, it is very helpful to compare fixed WR to variable to understand the magnitude of difference.

I don't really recommend anything beyond being realistic about what any of the math actually means.

I mean, there's some balance point between "f it, I'm sure it'll be fine" to "the calculator shows 0.64% risk of failure so it won't work". You need some tool to help tell you what the outcome would have been in different situations to give you an idea of whether you need to worry about a 2008-level recession or a 1930-level depression, or whatever. If you don't have that tool you're just as likely to react to minor market swings by excessively cutting spending as you are to thinking everything is fine until you've eroded your principal to an unhealthy point.

I agree that without "4%", I'd be completely lost.  Frankly, that one thing--that one blog post--completely transformed the way I look at retirement.  Before that it was "just save as much as you think you reasonably can and keep doing that and then... something will happen that I guess will indicate it's time to quit working".  It was an entirely nebulous concept.  "4%" gave me goals and understanding and clarity, even though I knew instantly that it would likely never be *exactly* 4%.  So I very much need the specific, exact math, to give me that.

But if FIREcalc tells me that the math gets me to 90%, I have tremendous faith in myself to get that other 10%.  I'm resourceful, sensible, aware, thoughtful, and intelligent.  That's more than enough to make of for a ten percent gap in the math.  IOW, my dollars get me to 90%, and if that's not enough, myself will get me to 100%.

[Metalcat]

For exactly the reasons Malcat outline, I think it's a lot more instructive to pick one of the dynamic withdrawal scenarios that never run out of money, like the "withdraw the lower of 4% of the initial stash, inflation adjusted, or 4% or the current stash", which sort of mimic how a real person would behave. Then look at what the minimum withdrawal rate was across the scenarios and see if you'd be OK with that number.

Instead of "you ran out of money in 5% of the realizations", it's a lot more instructive to know whether your $200k spend had to be reduced to $190k or $10k. If the "failure" is the former, you'd probably be OK with it. Not so much with the latter.

My argument isn't even to pick a variable WR, my argument is that most people will naturally vary their withdrawal rate given ongoing market conditions that threaten to erode their principle.

I don't recommend putting faith in any output from a FIRE calculator. Although, it is very helpful to compare fixed WR to variable to understand the magnitude of difference.

I don't really recommend anything beyond being realistic about what any of the math actually means.

I mean, there's some balance point between "f it, I'm sure it'll be fine" to "the calculator shows 0.64% risk of failure so it won't work". You need some tool to help tell you what the outcome would have been in different situations to give you an idea of whether you need to worry about a 2008-level recession or a 1930-level depression, or whatever. If you don't have that tool you're just as likely to react to minor market swings by excessively cutting spending as you are to thinking everything is fine until you've eroded your principal to an unhealthy point.

I feel like I'm saying the same thing over and over again and never managing to say it clearly.

First, I wasn't saying that I I'm against recommending a planned variable WR, I meant that I never personally recommend any universal approach to anything. I was clarifying my previous point that I wasn't previously making a recommendation. That's all.

Second, sure there's a balance. There's a balance between understanding what the calculator can do and what it can't.

The calculator can indicate the relative impact of a market drop, and the relative impact of a fixed WR from a variable rate, but it cannot actually output a predictive value.

I spent ages with these calculators and they were invaluable to me in terms of understanding how all of this works, but the more I fiddled with them, the more I understood that the percentage output is only meaningful as a comparison.

The percentages could be changed to any scale from red-->blue, or grades F-->A+, or even the pain scale faces at the doctor's office. They would still be just as informative, and probably more helpful because they would remove the illusion of predictive value.

The only thing an output of 100% says compared to 80% is "given the inputs that you chose, there is a significantly higher magnitude of positive impact on projected outcome compared to your previously selected inputs."

If I'm being totally unrealistic my ability and willingness to cut my spending in down years, it doesn't matter if the calculator spits out 100% compared to 80%.

Put more simply, if the calculator has no inter-user predictive validity, then it has no predictive validity period. Given the same inputs, it can't tell the difference between an unrealistic person and an overly cautious person. One might be very likely to end up needing to go back to work and the other might be very likely to die rich, and the calculator might spit out the same 90%.

So is a variable WR a good idea? Sure, for some people.
Is a bond tent a good idea? Sure, for some people.
Is a sub 4% WR a good idea? Sure, for some people.
Is being prepared to have to generate money in retirement a good idea? Oh absolutely...for some people.

That's why I don't make generalized recommendations beyond very, very vague concepts, because the factors that determine which people those "some people" are are very personal and individual in ways that calculators can't effectively account for.

[chevy1956]

Someone who is Lean FIRE does not have the wiggle room, so 100% would be necessary.

I really don't think this is the case but this is just a play on expenses ala the ERE way to retire. If you manage your expenses really strictly and get to a really high success rate than you should be good to go. The point being if your expenses increase than because you've saved to such a low WR you should be okay.

To me you can't manage your expenses that tightly. It's just a guide which makes the math a guide. It's interesting to think about but I don't think it helps that much in RE.

I suppose one thing you can do is just updated your figures annually and see how it is looking. Then you decide if and how much to withdraw. I think this is what I will be doing. Keep tracking the numbers unless the amount of money becomes way more than I can spend in my lifetime. I fully expect this to happen rather than running out of money.

[Metalcat]

For exactly the reasons Malcat outline, I think it's a lot more instructive to pick one of the dynamic withdrawal scenarios that never run out of money, like the "withdraw the lower of 4% of the initial stash, inflation adjusted, or 4% or the current stash", which sort of mimic how a real person would behave. Then look at what the minimum withdrawal rate was across the scenarios and see if you'd be OK with that number.

Instead of "you ran out of money in 5% of the realizations", it's a lot more instructive to know whether your $200k spend had to be reduced to $190k or $10k. If the "failure" is the former, you'd probably be OK with it. Not so much with the latter.

My argument isn't even to pick a variable WR, my argument is that most people will naturally vary their withdrawal rate given ongoing market conditions that threaten to erode their principle.

I don't recommend putting faith in any output from a FIRE calculator. Although, it is very helpful to compare fixed WR to variable to understand the magnitude of difference.

I don't really recommend anything beyond being realistic about what any of the math actually means.

I mean, there's some balance point between "f it, I'm sure it'll be fine" to "the calculator shows 0.64% risk of failure so it won't work". You need some tool to help tell you what the outcome would have been in different situations to give you an idea of whether you need to worry about a 2008-level recession or a 1930-level depression, or whatever. If you don't have that tool you're just as likely to react to minor market swings by excessively cutting spending as you are to thinking everything is fine until you've eroded your principal to an unhealthy point.

I agree that without "4%", I'd be completely lost.  Frankly, that one thing--that one blog post--completely transformed the way I look at retirement.  Before that it was "just save as much as you think you reasonably can and keep doing that and then... something will happen that I guess will indicate it's time to quit working".  It was an entirely nebulous concept.  "4%" gave me goals and understanding and clarity, even though I knew instantly that it would likely never be *exactly* 4%.  So I very much need the specific, exact math, to give me that.

But if FIREcalc tells me that the math gets me to 90%, I have tremendous faith in myself to get that other 10%.  I'm resourceful, sensible, aware, thoughtful, and intelligent.  That's more than enough to make of for a ten percent gap in the math.  IOW, my dollars get me to 90%, and if that's not enough, myself will get me to 100%.

Yes, 4% is a FANTASTIC basic starting point.

It's knowing your personal risks, which the calculator can't account for, that's really important. My personal risk factors are highly complicated and absolutely impossible to account for with a FIRE calculator. It's taken me years to formulate a tentative plan that I think works for me.

[chevy1956]

I feel like I'm saying the same thing over and over again and never managing to say it clearly.

Personally I think your advice is spot on. I have the same perspective so I get it.

Given the same inputs, it can't tell the difference between an unrealistic person and an overly cautious person.

This is part of the issue. The next part is that the future won't turn out exactly as you plan. So some years you may spend some more cash whereas other years you won't. It's really hard to be completely 100% clear on your expenses. I'm currently at a 5% WR but I am spending 5k on my oldest son this year. That cost will start to drop from next year on. I have a younger son who I don't spend anywhere near that amount on but I may at some point. We have a car that is great at the moment but at some point I have to upgrade it. I may have to paint the house. I may do that myself or I may pay someone. It's impossible for me to be 100% accurate with my future expenses.

[Metalcat]

I feel like I'm saying the same thing over and over again and never managing to say it clearly.

Personally I think your advice is spot on. I have the same perspective so I get it.

Given the same inputs, it can't tell the difference between an unrealistic person and an overly cautious person.

This is part of the issue. The next part is that the future won't turn out exactly as you plan. So some years you may spend some more cash whereas other years you won't. It's really hard to be completely 100% clear on your expenses. I'm currently at a 5% WR but I am spending 5k on my oldest son this year. That cost will start to drop from next year on. I have a younger son who I don't spend anywhere near that amount on but I may at some point. We have a car that is great at the moment but at some point I have to upgrade it. I may have to paint the house. I may do that myself or I may pay someone. It's impossible for me to be 100% accurate with my future expenses.

Yes, the lumpiness of year over year spending is another drum I bang on incessantly. I said it already earlier that the calculator makes the assumption of exactly X% WR +estimated inflation spent year over year, which no one does, which makes its own assumptions fundamentally inaccurate.

[lutorm]

For exactly the reasons Malcat outline, I think it's a lot more instructive to pick one of the dynamic withdrawal scenarios that never run out of money, like the "withdraw the lower of 4% of the initial stash, inflation adjusted, or 4% or the current stash", which sort of mimic how a real person would behave. Then look at what the minimum withdrawal rate was across the scenarios and see if you'd be OK with that number.

Instead of "you ran out of money in 5% of the realizations", it's a lot more instructive to know whether your $200k spend had to be reduced to $190k or $10k. If the "failure" is the former, you'd probably be OK with it. Not so much with the latter.

My argument isn't even to pick a variable WR, my argument is that most people will naturally vary their withdrawal rate given ongoing market conditions that threaten to erode their principle.

I don't recommend putting faith in any output from a FIRE calculator. Although, it is very helpful to compare fixed WR to variable to understand the magnitude of difference.

I don't really recommend anything beyond being realistic about what any of the math actually means.

I mean, there's some balance point between "f it, I'm sure it'll be fine" to "the calculator shows 0.64% risk of failure so it won't work". You need some tool to help tell you what the outcome would have been in different situations to give you an idea of whether you need to worry about a 2008-level recession or a 1930-level depression, or whatever. If you don't have that tool you're just as likely to react to minor market swings by excessively cutting spending as you are to thinking everything is fine until you've eroded your principal to an unhealthy point.

I feel like I'm saying the same thing over and over again and never managing to say it clearly.

Sorry, that was not my intent.

I think we are in agreement. It's essentially "the map is not the territory" argument (https://en.wikipedia.org/wiki/Map%E2%80%93territory_relation) in that the scenarios in the calculator is not the same as ones travel through retirement but, like a map, it's useful for plotting a path. Having a made a path on a map, though, is not the same as traveling though the terrain, both because the map is only a facsimile of the terrain but also because you might not be as good at traveling as you thought you would be...

My day job is in modeling and simulation, so I'm quite familiar with the risk of thinking that the simulator is reality. The entire trick is figuring out what you should and should not rely on it for.

[chevy1956]

I think we are in agreement. It's essentially "the map is not the territory" argument (https://en.wikipedia.org/wiki/Map%E2%80%93territory_relation) in that the scenarios in the calculator is not the same as ones travel through retirement but, like a map, it's useful for plotting a path. Having a made a path on a map, though, is not the same as traveling though the terrain, both because the map is only a facsimile of the terrain but also because you might not be as good at traveling as you thought you would be...

My day job is in modeling and simulation, so I'm quite familiar with the risk of thinking that the simulator is reality. The entire trick is figuring out what you should and should not rely on it for.

This is in my opinion the right way to view these calculators. Use them to get some confidence in your numbers. Understand what those numbers mean. I mean if you really need to get to a 3% WR you are costing yourself a lot of extra time at work. Those expenses are going to change in retirement and it's impossible to get the exact number worked out in advance. Do you really need to budget for that once in a lifetime trip to Paris that you may or may not even care about in 20 years time.

To answer the OP I'm pretty cool with an 80% success rate but I have back-up options available to me. I also don't believe I am taking the path of deluding myself via thinking that these models are cast in stone and going forward I have to act perfectly in sync with my plan. My plan is a very rough guide but I'm fine with it.

[MasterStache]

I feel like I'm saying the same thing over and over again and never managing to say it clearly.

Personally I think your advice is spot on. I have the same perspective so I get it.

Given the same inputs, it can't tell the difference between an unrealistic person and an overly cautious person.

This is part of the issue. The next part is that the future won't turn out exactly as you plan. So some years you may spend some more cash whereas other years you won't. It's really hard to be completely 100% clear on your expenses. I'm currently at a 5% WR but I am spending 5k on my oldest son this year. That cost will start to drop from next year on. I have a younger son who I don't spend anywhere near that amount on but I may at some point. We have a car that is great at the moment but at some point I have to upgrade it. I may have to paint the house. I may do that myself or I may pay someone. It's impossible for me to be 100% accurate with my future expenses.

Yes, the lumpiness of year over year spending is another drum I bang on incessantly. I said it already earlier that the calculator makes the assumption of exactly X% WR +estimated inflation spent year over year, which no one does, which makes its own assumptions fundamentally inaccurate.

Absolutely. Although my spouse has yet to join me in early retirement we've been monitoring expenses for years. We've seen swings of over 10K from year to year. This year is looking low on expenses. Last year we dealt with a car replacement (car wreck) and a trip to the ICU for our dog. Obviously neither were predictable. But it helps us understand the variability and flexibility needed in retirement.

[norajean]

Estimating what amount you need saved to quit working is fundamentally different from an actual withdrawal strategy.  Calculators are intended to help with the former, but not the latter.

[chevy1956]

Estimating what amount you need saved to quit working is fundamentally different from an actual withdrawal strategy.  Calculators are intended to help with the former, but not the latter.

The problem is that it's pretty hard to follow a withdrawal strategy perfectly to the letter. The same as it is pretty hard to have your expenses remain stable year on year. When I say pretty hard I really mean it's probably impossible.

This means that your inputs are not perfectly accurate and therefore all the modelling isn't perfectly accurate which means that the whole argument over getting to 3.5% or a 4% or whatever withdrawal rate isn't really going to be completely accurate. Then when you add it's just historical data that is being modeled it throws a lot of potential variability into the mix.

I get that we like to use these models and they are great but the point is that you can fool yourself by what you use as your expenses or your timeline or your withdrawal strategy. They don't really offer the certainiy that people think they do. So people are arguing about a 4% WR when their inputs are built on quicksand.

I think you get to a figure you think is good enough but the more you know yourself and your blind spots the better your plan will be. It won't be the 3.5% WR versus the 5% WR that is the reason for your failure. It's going to your expenses and options going forward.

I know a couple of things that have stood out to me. You can work for a long time trying to gain some safety level that is really a psychological issue and at some point you have to make a leap of faith if you want to retire.

[reeshau]

I feel like I'm saying the same thing over and over again and never managing to say it clearly.

Personally I think your advice is spot on. I have the same perspective so I get it.

Given the same inputs, it can't tell the difference between an unrealistic person and an overly cautious person.

This is part of the issue. The next part is that the future won't turn out exactly as you plan. So some years you may spend some more cash whereas other years you won't. It's really hard to be completely 100% clear on your expenses. I'm currently at a 5% WR but I am spending 5k on my oldest son this year. That cost will start to drop from next year on. I have a younger son who I don't spend anywhere near that amount on but I may at some point. We have a car that is great at the moment but at some point I have to upgrade it. I may have to paint the house. I may do that myself or I may pay someone. It's impossible for me to be 100% accurate with my future expenses.

Yes, the lumpiness of year over year spending is another drum I bang on incessantly. I said it already earlier that the calculator makes the assumption of exactly X% WR +estimated inflation spent year over year, which no one does, which makes its own assumptions fundamentally inaccurate.

Absolutely. Although my spouse has yet to join me in early retirement we've been monitoring expenses for years. We've seen swings of over 10K from year to year. This year is looking low on expenses. Last year we dealt with a car replacement (car wreck) and a trip to the ICU for our dog. Obviously neither were predictable. But it helps us understand the variability and flexibility needed in retirement.

I actually think there are some straightforward ways to deal with your two examples.  For many replacements / repairs, you can build sinking funds to save for the eventuality--in effect, smoothing out the cost.  Foe the latter, if the variability puts a dent in your plan, you can get medical insurance for your pet.

Of course, not everyone does this.  Some would be just as happy to reduce discretionary spend for a while, or otherwise improvise an answer.  Either is OK, as a match for your personality.  But it isn't rocket science, either.  If not entirely common, It's a solved problem, easily observed in any number of case studies.

[jeroly]

I actually think there are some straightforward ways to deal with your two examples.  For many replacements / repairs, you can build sinking funds to save for the eventuality--in effect, smoothing out the cost.  Foe the latter, if the variability puts a dent in your plan, you can get medical insurance for your pet.

Of course, not everyone does this.  Some would be just as happy to reduce discretionary spend for a while, or otherwise improvise an answer.  Either is OK, as a match for your personality.  But it isn't rocket science, either.  If not entirely common, It's a solved problem, easily observed in any number of case studies.

When it comes to budgeting, I use the sinking fund concept.  For example, I set up a new auto expense item that is 1/10 of the price of the car I want to buy (ten years from the last time I bought a car), and a travel expense item based on my expected travels for the year, which are not evenly distributed over the months.  However I don't specifically fund those sinking funds but rather just withdraw from my Vanguard accounts an amount each month or two that will fund what I am planning to do for that time interval. 

When I was living with my ex, we had separate finances and funded a joint expense fund from which all household expenses were withdrawn.  I would examine spending on each line item in that household budget in an annual ritual which also involved recalculating what the joint expense fund would require from each of use each month; our contributions were assessed based on a modified ratio of our incomes. We treated that joint expense fund like an escrow account at a bank, with sinking funds for things like repairs, a new roof, etc., and would adjust the total required fund deposits based on the adjustments to estimated new year expenses versus prior year.

Now that I don't have to do that annual household calculation, I don't look at year-to-year variation in line item spending closely, but overall it's pretty notable how even my spending has been over the last 20 years.  It's been in a range from $ (X),000 to $ (1.2 * X),000  more or less every year for the last 20 years, even when there have been some big lump sum outlays (car, roof, education).  I haven't necessarily consciously reduced spending in other areas when those big-ticket costs have hit, but I either have done so subconsiously or have run into a weird set of coincidences.

[MasterStache]

I feel like I'm saying the same thing over and over again and never managing to say it clearly.

Personally I think your advice is spot on. I have the same perspective so I get it.

Given the same inputs, it can't tell the difference between an unrealistic person and an overly cautious person.

This is part of the issue. The next part is that the future won't turn out exactly as you plan. So some years you may spend some more cash whereas other years you won't. It's really hard to be completely 100% clear on your expenses. I'm currently at a 5% WR but I am spending 5k on my oldest son this year. That cost will start to drop from next year on. I have a younger son who I don't spend anywhere near that amount on but I may at some point. We have a car that is great at the moment but at some point I have to upgrade it. I may have to paint the house. I may do that myself or I may pay someone. It's impossible for me to be 100% accurate with my future expenses.

Yes, the lumpiness of year over year spending is another drum I bang on incessantly. I said it already earlier that the calculator makes the assumption of exactly X% WR +estimated inflation spent year over year, which no one does, which makes its own assumptions fundamentally inaccurate.

Absolutely. Although my spouse has yet to join me in early retirement we've been monitoring expenses for years. We've seen swings of over 10K from year to year. This year is looking low on expenses. Last year we dealt with a car replacement (car wreck) and a trip to the ICU for our dog. Obviously neither were predictable. But it helps us understand the variability and flexibility needed in retirement.

I actually think there are some straightforward ways to deal with your two examples.  For many replacements / repairs, you can build sinking funds to save for the eventuality--in effect, smoothing out the cost.  Foe the latter, if the variability puts a dent in your plan, you can get medical insurance for your pet.

Of course, not everyone does this.  Some would be just as happy to reduce discretionary spend for a while, or otherwise improvise an answer.  Either is OK, as a match for your personality.  But it isn't rocket science, either.  If not entirely common, It's a solved problem, easily observed in any number of case studies.

Sure. None of this really impacted us too much because technically we are still accumulating. But you can't plan for every scenario. My spouse had just purchased a the car a couple years before the accident. What if this new(er) car is wrecked as well this soon? You can make rainy day funds but there is no guarantee if, when and how often you will need them. The dog in the ICU can be prevented by not having a pet. We have also been fighting a legal battle with my son's High School in which we had to obtain a lawyer. That's thousands of more dollars. You just can't predict every scenario. My spouse and I both agree that having a lot of flexibility is the key to being good in full retirement. Some years are going to be worse than others. Maybe even a lot worse.

To relate this back to the topic, I don't put too much faith into any retirement calculator. My spouse will call it quits when she is ready.

[Metalcat]

I'm rereading all of Taleb's books right now, and this excellent example came up.

If someone tosses a coin 100 times and the first 99 times it comes up heads, what are the chances is will come up heads again?

The mathematician will say 50%, applying the most basic of statistical calculations to the problem.

The person with even a fraction of common sense will say the coin is obviously weighted.

The calculation will only be accurate if the assumptions it is working on are accurate.

A calculator can only generate it's outputs based on the known risks built into the equation, but it's utterly useless for assessing actual risk.

[jeroly]

I'm rereading all of Taleb's books right now, and this excellent example came up.

If someone tosses a coin 100 times and the first 99 times it comes up heads, what are the chances is will come up heads again?

The mathematician will say 50%, applying the most basic of statistical calculations to the problem.

The person with even a fraction of common sense will say the coin is obviously weighted.

The calculation will only be accurate if the assumptions it is working on are accurate.

A calculator can only generate it's outputs based on the known risks built into the equation, but it's utterly useless for assessing actual risk.

No mathematician will say that.
If the question was, “Someone tosses a _fair_ coin 100 times” ... then yes, that’s the answer a mathematician would give.  However, using statistical inference, that same mathematician would reject an assumption that the coin was fair unless it’s stated as a given.

[Metalcat]

I'm rereading all of Taleb's books right now, and this excellent example came up.

If someone tosses a coin 100 times and the first 99 times it comes up heads, what are the chances is will come up heads again?

The mathematician will say 50%, applying the most basic of statistical calculations to the problem.

The person with even a fraction of common sense will say the coin is obviously weighted.

The calculation will only be accurate if the assumptions it is working on are accurate.

A calculator can only generate it's outputs based on the known risks built into the equation, but it's utterly useless for assessing actual risk.

No mathematician will say that.
If the question was, “Someone tosses a _fair_ coin 100 times” ... then yes, that’s the answer a mathematician would give.  However, using statistical inference, that same mathematician would reject an assumption that the coin was fair unless it’s stated as a given.

True, the example shouldn't say "mathematician", and I'm not positive Taleb specified mathematician now that I think of it. He did specify an academic of some kind. 

[lutorm]

I really liked "The Black Swan" (once you get past his abrasive writing style) and the main point is along exactly these line of "don't mistake the model for reality".

I take Taleb's point to be that if someone gives you a fair coin and you toss it 100 times and the first 99 times comes up as heads, your conclusion should be "I've been given wrong information, the coin is not fair." This may sound obvious to the point of silliness, because it should be clear after 10 tosses that the model is questionable.

But in most real cases, the failure is in the wings of the distribution such that the model will be correct 999 times of a thousand, or more, but when it fails it fails by the 1000th time being an event that should not happen in a billion cases. It's precisely those events that can't be tested, because you can't gather enough evidence to find them out in the wings.

I think one of the examples he uses is the 1984 stock crash, which according to the models at the time should not have happened once in the lifetime of the universe. In our context, the black swan failure probably in the conclusion that "the worst retirement scenario that can happen is if you retired in 1972". No, that's only the worst scenario that's played out in our very limited sample (only the US, over the past century), which means we have zero evidence about anything that's had about a less than 1% chance of happening.

I personally suspect that everyone in the US over the past century have been spectacularly lucky and the fraction of scenarios in a properly averaged sample that end in war, genocide, runaway inflation, societal collapse, asteroid impact, zombie apocalypse, etc, where running out of money in retirement will be a small worry, is significant.

[Metalcat]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

The chances of finding out that your mom has early onset Alzheimer's and needs expensive care, or your kid needs financial help so they don't go bankrupt from their divorce, or your grandkid gets into some crazy elite program or commits a crime and needs an expensive legal defense (either way works), or your LCOL plans go tits up because you can't stand the isolation or smaller town culture, or you have a small stroke in a part of your brain that makes you less responsible with your money, etc, etc.

None of the above are wildly improbable possibilities. None of them are rare enough to be considered Black Swans. I came up with all of them from examples within my patient population, which is just a few thousand of upper middle class Canadians.

[lutorm]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

Well, sure, but they don't claim to cover that. If you put in that you'll use $40k/year and you use $80k, that's a straightforward case of Garbage In - Garbage Out. The failure modes of the model itself aren't so plainly visible.

[Villanelle]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

The chances of finding out that your mom has early onset Alzheimer's and needs expensive care, or your kid needs financial help so they don't go bankrupt from their divorce, or your grandkid gets into some crazy elite program or commits a crime and needs an expensive legal defense (either way works), or your LCOL plans go tits up because you can't stand the isolation or smaller town culture, or you have a small stroke in a part of your brain that makes you less responsible with your money, etc, etc.

None of the above are wildly improbable possibilities. None of them are rare enough to be considered Black Swans. I came up with all of them from examples within my patient population, which is just a few thousand of upper middle class Canadians.

Yes.  During the "how much E fund do I need conversations", my answer is always that you need to look at your life.  And I think the answer is the same for "what is my safe WR?".  While your list isn't comprehensive (and I know you didn't intend it to be), I look at those things and since I don't have kids or grandkids, I can cross of many of the possibilities.  I don't plan to FIRE to a small town, or a town significantly different than ones I've lived in, so that probability is lower.  My parent's have, frankly, gobs of money and excellent health care, so those possibilities are small.  My relationship seems rock solid (and "empty nest" seems to be a major driver in divorces that would come around this stage of life; also not an issue for us). 

To be very clear, I'm not saying my life is crisis-proof.  I could be sued.  My seemingly rock-solid relationship could turn out to be water-liquid instead.   I could be a victim of a financial crime.  Shit can always happen.  But as I evaluate my life in an individual level, rather than a broad demographically based perspective, I can decide whether my odds seem better or worse than average, and I can adjust accordingly.  That's the step that needs to be taken, after understanding the 4% rule.  Great, now that I understand what 4% means, I need to honestly evaluate MY life to see what puts me at higher or lower risk than average.  I have a special-needs kid who will never be self-supporting and parents barely getting by on social security but whom I'd gladly assist?  Maybe I need to be more conservative.  I'm a contended loaner, with no living family?  I'm probably on the lower scale of risk.

It's not as definitive, or as Shockingly Simple as "4%", but it's still well worthwhile. 

[Metalcat]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

Well, sure, but they don't claim to cover that. If you put in that you'll use $40k/year and you use $80k, that's a straightforward case of Garbage In - Garbage Out. The failure modes of the model itself aren't so plainly visible.

Entirely my point.

I don't fault the calculator. I'm a huge fan actually. I criticize the thinking that allows someone to angst over 90% vs 100% as if they actually represent a 10% or 0% chance of portfolio failure.

I've made that very very clear in my extensive ranting in this thread.

[Metalcat]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

The chances of finding out that your mom has early onset Alzheimer's and needs expensive care, or your kid needs financial help so they don't go bankrupt from their divorce, or your grandkid gets into some crazy elite program or commits a crime and needs an expensive legal defense (either way works), or your LCOL plans go tits up because you can't stand the isolation or smaller town culture, or you have a small stroke in a part of your brain that makes you less responsible with your money, etc, etc.

None of the above are wildly improbable possibilities. None of them are rare enough to be considered Black Swans. I came up with all of them from examples within my patient population, which is just a few thousand of upper middle class Canadians.

Yes.  During the "how much E fund do I need conversations", my answer is always that you need to look at your life.  And I think the answer is the same for "what is my safe WR?".  While your list isn't comprehensive (and I know you didn't intend it to be), I look at those things and since I don't have kids or grandkids, I can cross of many of the possibilities.  I don't plan to FIRE to a small town, or a town significantly different than ones I've lived in, so that probability is lower.  My parent's have, frankly, gobs of money and excellent health care, so those possibilities are small.  My relationship seems rock solid (and "empty nest" seems to be a major driver in divorces that would come around this stage of life; also not an issue for us). 

To be very clear, I'm not saying my life is crisis-proof.  I could be sued.  My seemingly rock-solid relationship could turn out to be water-liquid instead.   I could be a victim of a financial crime.  Shit can always happen.  But as I evaluate my life in an individual level, rather than a broad demographically based perspective, I can decide whether my odds seem better or worse than average, and I can adjust accordingly.  That's the step that needs to be taken, after understanding the 4% rule.  Great, now that I understand what 4% means, I need to honestly evaluate MY life to see what puts me at higher or lower risk than average.  I have a special-needs kid who will never be self-supporting and parents barely getting by on social security but whom I'd gladly assist?  Maybe I need to be more conservative.  I'm a contended loaner, with no living family?  I'm probably on the lower scale of risk.

It's not as definitive, or as Shockingly Simple as "4%", but it's still well worthwhile.

Of course!

All estimates of future personal risk are by definition going to be very hand-wavy and kind of pulled out of our asses in terms of what they might cost.

That would be the entire point of what I've been trying to say.

I'm not saying the calculator should be better, I'm saying that it CAN'T be better and we shouldn't expect it to be.

The weather Channel is very valuable to me even if it can't reasonably predict the actual weather very well.

ETA: I *just* got a weather Channel alert on my phone as I was typing that

[lutorm]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

Well, sure, but they don't claim to cover that. If you put in that you'll use $40k/year and you use $80k, that's a straightforward case of Garbage In - Garbage Out. The failure modes of the model itself aren't so plainly visible.

Entirely my point.

I don't fault the calculator. I'm a huge fan actually. I criticize the thinking that allows someone to angst over 90% vs 100% as if they actually represent a 10% or 0% chance of portfolio failure.

I've made that very very clear in my extensive ranting in this thread.

Yeah, but saying "What's even worse though about FIRE simulators ..." makes it sound like you are faulting the calculators, when what you mean is "what's even worse is how people use FIRE simulators". ;-)

[Metalcat]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

Well, sure, but they don't claim to cover that. If you put in that you'll use $40k/year and you use $80k, that's a straightforward case of Garbage In - Garbage Out. The failure modes of the model itself aren't so plainly visible.

Entirely my point.

I don't fault the calculator. I'm a huge fan actually. I criticize the thinking that allows someone to angst over 90% vs 100% as if they actually represent a 10% or 0% chance of portfolio failure.

I've made that very very clear in my extensive ranting in this thread.

Yeah, but saying "What's even worse though about FIRE simulators ..." makes it sound like you are faulting the calculators, when what you mean is "what's even worse is how people use FIRE simulators". ;-)

Sure, you could interpret it that way if taken completely out of context of everything else I've written in this thread...

;)

[Villanelle]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

The chances of finding out that your mom has early onset Alzheimer's and needs expensive care, or your kid needs financial help so they don't go bankrupt from their divorce, or your grandkid gets into some crazy elite program or commits a crime and needs an expensive legal defense (either way works), or your LCOL plans go tits up because you can't stand the isolation or smaller town culture, or you have a small stroke in a part of your brain that makes you less responsible with your money, etc, etc.

None of the above are wildly improbable possibilities. None of them are rare enough to be considered Black Swans. I came up with all of them from examples within my patient population, which is just a few thousand of upper middle class Canadians.

Yes.  During the "how much E fund do I need conversations", my answer is always that you need to look at your life.  And I think the answer is the same for "what is my safe WR?".  While your list isn't comprehensive (and I know you didn't intend it to be), I look at those things and since I don't have kids or grandkids, I can cross of many of the possibilities.  I don't plan to FIRE to a small town, or a town significantly different than ones I've lived in, so that probability is lower.  My parent's have, frankly, gobs of money and excellent health care, so those possibilities are small.  My relationship seems rock solid (and "empty nest" seems to be a major driver in divorces that would come around this stage of life; also not an issue for us). 

To be very clear, I'm not saying my life is crisis-proof.  I could be sued.  My seemingly rock-solid relationship could turn out to be water-liquid instead.   I could be a victim of a financial crime.  Shit can always happen.  But as I evaluate my life in an individual level, rather than a broad demographically based perspective, I can decide whether my odds seem better or worse than average, and I can adjust accordingly.  That's the step that needs to be taken, after understanding the 4% rule.  Great, now that I understand what 4% means, I need to honestly evaluate MY life to see what puts me at higher or lower risk than average.  I have a special-needs kid who will never be self-supporting and parents barely getting by on social security but whom I'd gladly assist?  Maybe I need to be more conservative.  I'm a contended loaner, with no living family?  I'm probably on the lower scale of risk.

It's not as definitive, or as Shockingly Simple as "4%", but it's still well worthwhile.

Of course!

All estimates of future personal risk are by definition going to be very hand-wavy and kind of pulled out of our asses in terms of what they might cost.

That would be the entire point of what I've been trying to say.

I'm not saying the calculator should be better, I'm saying that it CAN'T be better and we shouldn't expect it to be.

The weather Channel is very valuable to me even if it can't reasonably predict the actual weather very well.

ETA: I *just* got a weather Channel alert on my phone as I was typing that

We are in violent agreement.

Perhaps these planning tools need to at least offer up a link to something reminding people of these things--that they are tools (the planning devices, not the people who are planning!) and have value, but that their calculations need to be viewed in the context of real life and real people, and one needs to consider one's own circumstance along side the cold, hard math. 

And I hope you stay dry/warm/moderately weathered. 

[Metalcat]

What's even worse though about FIRE simulators is that they aren't just missing tail risks, they're missing extremely common individual risks that they just can't account for, like the example I gave of divorce. That's why I just don't take them seriously at all as a predictive model.

The chances of finding out that your mom has early onset Alzheimer's and needs expensive care, or your kid needs financial help so they don't go bankrupt from their divorce, or your grandkid gets into some crazy elite program or commits a crime and needs an expensive legal defense (either way works), or your LCOL plans go tits up because you can't stand the isolation or smaller town culture, or you have a small stroke in a part of your brain that makes you less responsible with your money, etc, etc.

None of the above are wildly improbable possibilities. None of them are rare enough to be considered Black Swans. I came up with all of them from examples within my patient population, which is just a few thousand of upper middle class Canadians.

Yes.  During the "how much E fund do I need conversations", my answer is always that you need to look at your life.  And I think the answer is the same for "what is my safe WR?".  While your list isn't comprehensive (and I know you didn't intend it to be), I look at those things and since I don't have kids or grandkids, I can cross of many of the possibilities.  I don't plan to FIRE to a small town, or a town significantly different than ones I've lived in, so that probability is lower.  My parent's have, frankly, gobs of money and excellent health care, so those possibilities are small.  My relationship seems rock solid (and "empty nest" seems to be a major driver in divorces that would come around this stage of life; also not an issue for us). 

To be very clear, I'm not saying my life is crisis-proof.  I could be sued.  My seemingly rock-solid relationship could turn out to be water-liquid instead.   I could be a victim of a financial crime.  Shit can always happen.  But as I evaluate my life in an individual level, rather than a broad demographically based perspective, I can decide whether my odds seem better or worse than average, and I can adjust accordingly.  That's the step that needs to be taken, after understanding the 4% rule.  Great, now that I understand what 4% means, I need to honestly evaluate MY life to see what puts me at higher or lower risk than average.  I have a special-needs kid who will never be self-supporting and parents barely getting by on social security but whom I'd gladly assist?  Maybe I need to be more conservative.  I'm a contended loaner, with no living family?  I'm probably on the lower scale of risk.

It's not as definitive, or as Shockingly Simple as "4%", but it's still well worthwhile.

Of course!

All estimates of future personal risk are by definition going to be very hand-wavy and kind of pulled out of our asses in terms of what they might cost.

That would be the entire point of what I've been trying to say.

I'm not saying the calculator should be better, I'm saying that it CAN'T be better and we shouldn't expect it to be.

The weather Channel is very valuable to me even if it can't reasonably predict the actual weather very well.

ETA: I *just* got a weather Channel alert on my phone as I was typing that

We are in violent agreement.

Perhaps these planning tools need to at least offer up a link to something reminding people of these things--that they are tools (the planning devices, not the people who are planning!) and have value, but that their calculations need to be viewed in the context of real life and real people, and one needs to consider one's own circumstance along side the cold, hard math. 

And I hope you stay dry/warm/moderately weathered.

Or I can just keep ranting like a crazed woman...either way I'm cool with it.

[secondcor521]

I wondered if FIREcalc's data screens addressed the point that Malcat is raising.

Not sure if the language at the bottom applies, but it seems like it might:

"How can FIRECalc predict future returns from past performance?

It can't. And it doesn't try. In fact, it tries to predict what will not happen. This might sound confusing, but it's really simple.

Consider an analogy: Suppose you are building a house in Honolulu. No one could predict the temperature for any given future date during the decades the house will be used. But if you know that it has never been under 52° in that location in all of recorded history, you could make an intelligent judgment about how much heating capacity is enough.

Planning for an Anchorage-style winter would be a true waste of money that could be better used elsewhere.

FIREcalc works the same way, using stock market history and your portfolio and spending plan instead of weather history and furnace capacity, to give you the information to judge if your savings, combined with your Social Security, pensions, and other resources, are sufficient to handle the winter."

I think it's a fair criticism that the UI for FIREcalc, especially that first page, is pretty wordy and lots of people wouldn't read all the way to the bottom.  So maybe the disclaimer should be made more prominent, but at least it is there.

[sparkytheop]

A wrench is pretty good at turning bolts, but makes a really shitty hammer.  Calculators are useful tools, when used for the proper purpose.  Anytime a tool is used by an idiot, or by a non-idiot, but in a way not intended, the results could be pretty shitty.  Or they may still work with some luck.  I've been in the trades a long time though, and can tell you you'll never get people to quit misusing tools.

[UnleashHell]

yeah - these things are nothing more than a guide. not a single one of them predicted that my wife would want a divorce as soon as I was ready to fire.

That'd fuck with your numbers....it did mine.

[Metalcat]

yeah - these things are nothing more than a guide. not a single one of them predicted that my wife would want a divorce as soon as I was ready to fire.

That'd fuck with your numbers....it did mine.

Ugh, that's brutal. I hear you on this. Sudden unexpected life circumstances fuckered my financial projections as well.

[DK]

So with talking about some of the limitations of firecalc, does anyone know some better models/simulation to use that allow more input? Eg, my tentative plan is to be all in stocks minus a 12 month cash backdrop, and checking where the market is at to decide to pull from stocks or cash. With that indicator being if the market is >20% off its high. So in bad cases I won't be locking in losses but also not keeping so much in cash it will be a drag on returns. In looking I haven't found flexible model/simulation to input something like this though.

[park10]

So with talking about some of the limitations of firecalc, does anyone know some better models/simulation to use that allow more input? Eg, my tentative plan is to be all in stocks minus a 12 month cash backdrop, and checking where the market is at to decide to pull from stocks or cash. With that indicator being if the market is >20% off its high. So in bad cases I won't be locking in losses but also not keeping so much in cash it will be a drag on returns. In looking I haven't found flexible model/simulation to input something like this though.

The real problem is with the inputs that user puts in the model. In Engineering, there is a very common saying "Garbage in Garbage Out". User has absolutely no idea what their expenses are going to be next year or in 10 years or 20 years and what life will throw at them, but model wants you to put in a static number. What is worse is some models then perform real sophisticated regression based on historical market performance etc, but results of whether you can retire or not are still based on bad inputs. So, the question - is there a better model, -No, but thats not the Biggest problem, the problem is the inputs, and there really is no real solution for that. All you can do is put in a massive range of inputs to see a range of solutions, but, even then, the range can get so wide, that it essentially has very little actionable value. This does not even include the inherent problems with model itself i.e past performance is indicative of future returns.

[chevy1956]

yeah - these things are nothing more than a guide. not a single one of them predicted that my wife would want a divorce as soon as I was ready to fire.

That'd fuck with your numbers....it did mine.

This is something I think about. So far the marriage is pretty strong but you can't predict the future. This is the classic example of why debating about WR's misses the point.

[chevy1956]

So, the question - is there a better model, -No, but thats not the Biggest problem, the problem is the inputs, and there really is no real solution for that.

Exactly. I don't know why some people think that you can micromanage your FIRE situation via getting to the ideal pre-retirement model success rate.

I should add that it's also nowhere near as simple as being conservative in your guesstimated future expenses as well. It's going to be a guesstimate no matter what because you cannot predict the future. If you though choose to be highly conservative in your estimates then you have failed FIRE because you've worked too long. This can have unintended consequences as well.

Do the work. Try and get as accurate a picture or your financial situation as possible. At some point though you just have to make a leap of faith because you are talking about the future. I hope in the future my kids don't become drug addicts and I need to raise their kids. I haven't planned for this situation. I haven't planned for a divorce. I haven't planned for future medial advancements that mean I can live forever. I don't know if any of these things will happen. I've just reached a level that appears okay to me and I'll see what happens from here on in.

What is interesting to me is that I never realized that the decompression stage would occur but I'm going through that now. I feel I'm starting to come out of it but it's only been a couple of months.

I should add that prior to retirement I went through all sorts of modelling scenarios as well as tracked my expenses accurately and in detail and that is why I felt confident enough to retire.

[ardrum]

I like looking at Firecalc from the perspective of two "stages" of independence.  The first is a LeanFI, bare-bones budget that wouldn't really be very exciting/interesting but "livable" too.  The second is a wish list, desirable spending included, kind of Fat FIRE dream spending amount. 

For the lean annual spending amount, I target at least the 100% success rate (I agree with all the criticisms of Firecalc, so this is just a way for me to set a general goal though).  Then, for the "wish list" higher annual spending goal that is nice but not necessary, I aim for a 90% success rate amount as my final goal.

I figure this is a generally flexible approach though since I can definitely cut way back from the higher spending amount if real life outcomes are disproportionately unfavorable over time.

[BlueHouse]

We have a mentally handicapped daughter and we're not young folks like some of you are.   So we had a need for a very high chance of success and with gobs of fallback positions that would protect us.    So I over-engineered our FI process.   Bigly.

The cost in extra years vs. the chance of leaving our daughter at the not-so-tender mercies and care of the government made that trade off really important to us.    We didn't want our mentally handicapped daughter to be a financial burden on our son, we wanted her to be a financial blessing instead.

I would have been happy to take higher risks without that.

@SwordGuy I love you for doing this and saying this.  Happy new year!