“An Investment in Knowledge Always Pays The Best Interest” – Benjamin Franklin
March 2020. The portfolio was dropping hard week after week, headlines worse than the week before. I had cash I’d been planning to deploy for months. And I kept not deploying it, telling myself I was waiting for a better entry point.
What I was actually doing was something that has a name in investing circles: catching a falling knife. Every time I put money in, it dropped further. Fear shrank each subsequent purchase. By the time the actual bottom arrived, I was trickling in amounts that barely registered not because I was out of cash, but because each loss reinforced the certainty that it would get worse before it got better.
It didn’t. The market bounced hard off the March 23rd lows, and I was still sitting in cash, paralyzed by the question I’d created for myself, when do I get back in? That question has no clean answer. Which is exactly why you shouldn’t put yourself in a position where you have to answer it.
Ben Carlson described this perfectly: “Buy low, then buy lower, then buy even lower, and once you really hate yourself, buy lower than you thought was possible.”
That is exactly what it feels like from the inside. And it is exactly why the conversation about portfolios cannot stop at what you own. It has to include the systems, the rules, and the structures you build to protect the plan from the person running it.
That’s what this post is about.
Two Buckets, Two Different Jobs
The most clarifying thing we’ve done for our financial life is draw a hard line between two completely separate pools of money — each with its own purpose, its own rules, and its own emotional weight.
The first is the core portfolio. VTSAX and VBTLX, held in Roth IRA accounts. This is the machine. It is not for spending. It is not for experimenting. It is not for reacting to headlines or responding to economic news cycles. Its only job is to compound over decades without being disturbed.
The second is everything else: the cash buffer, the Maryland rental held inside a Wyoming LLC, the ESPP, the three Fundrise accounts, the HSA. These serve different purposes and operate under different rules. Some generate income. Some are building equity. Some are insurance against timing risk.
The separation is the point. When the core portfolio is clearly defined and walled off this is the thing that doesn’t get touched — every other decision becomes less emotionally loaded. You’re not asking “should I sell some VTSAX to fund this?” because VTSAX isn’t part of that conversation. The machine runs. Everything else serves around it.
Real estate exposure, rental income, employer stock, P2P lending these are all real assets worth building. But they belong in a different mental bucket from the long-term compounding engine. Mixing them creates confusion about what each piece is for. Keeping them separate creates clarity.
What We Hold, And Why The Allocation Is Where It Is
The core portfolio: 90% VTSAX, 10% VBTLX.
VTSAX is a stake in every publicly traded company in the US market over 3,500 companies, expense ratio 0.04%, meaning $4 per year on every $10,000 invested. VBTLX holds thousands of investment-grade bonds at 0.05%. Both held almost entirely inside Roth IRA accounts, meaning every dollar of growth, every dividend, every rebalancing transaction generates no taxable event and exits tax-free.
We’re staying in VTSAX. The reason is not that we think 100% domestic is mathematically superior. It’s that we know ourselves. Adding a third fund adds another variable to track, another allocation decision to make each month, and another data point for the brain to react to when international underperforms domestic for five straight years as it did between roughly 2010 and 2020. The efficient frontier math is real. The behavioral cost of complexity is also real. For us, the second number outweighs the first.
A plan you actually stick to through good years and bad is worth more than the theoretically optimal plan you’ll tinker with every time something diverges from expectations.
The 10% in VBTLX is not primarily an investment decision. It is a behavioral one. Bonds don’t generate great expected returns over long time horizons. That’s not why they’re there. They’re there because when equities fall hard 30%, 35%, 40% bonds tend to hold steady, and that cushion keeps the portfolio from landing at a number that triggers panic. The 10% costs some expected return. What it buys is the ability to stay in the game when the game gets hard.
It also gives us something to rebalance against without selling anything. When equities drop and become underweight relative to the 90% target, new money flows toward them. When equities run up, new money flows toward bonds. The allocation creates a mechanism for buying what’s cheap and adding to what’s expensive not as a market call, but as the automatic consequence of following a predetermined plan.
The Rule That Protects Everything Else: Never Sell
There is one rule that sits above all the others in how we manage this portfolio: we do not sell from the core portfolio.
Not when things feel expensive. Not when something looks like it’s about to drop. Not when a compelling case is made at dinner about why this time really is different.
The reason is not confidence in our predictions. It’s exactly the opposite we’ve accepted, genuinely accepted, that we cannot reliably predict what markets will do. Nobody can. And the cost of acting on the illusion that we can is not just the bad trade. It’s the chain of impossible decisions that follows.
Here’s how that plays out: you sell because you think things are going lower. Maybe they do go lower and you feel briefly smart. But then you’re facing the hardest question in investing, when do I get back in? At every price on the recovery, there’s a convincing reason to wait. Too early and you’re calling the bottom. Too late and you’ve already missed meaningful gains. Most people never get back in cleanly. The real cost of selling is not the exit. It’s the re-entry that never happens cleanly.
The rule eliminates the problem entirely. There is no exit to reverse. There is only the question of what to buy this month.
We have never sold from the core portfolio. The discipline is not in the holding. It’s in having decided, in advance and in writing, that selling is not an option so that the decision never has to be made in the moment, under conditions specifically designed to produce bad decisions.
Rebalancing With New Money, Not Sales
When the portfolio drifts from its target allocation which it does constantly, because markets move we don’t sell what’s overweight to buy what’s underweight. Auto balancing direct new contributions toward whatever is underweight until balance is restored.
This approach has several compounding advantages. It creates no taxable events. It never puts us in the position of having to decide when to re-enter a position we’ve trimmed. It uses the one variable we actually control monthly contributions as the instrument for maintaining the allocation.
Practically, it works like this: a simple tracking sheet shows current allocation versus target. When I run it each month, it tells me where the gap is. If VTSAX has run up and is now 93% instead of 90%, the next contribution goes to VBTLX. If both are near target, my investment tool splits it proportionally. Simple, mechanical, no judgment required. All done on the backend.
The psychological effect is worth naming. When equities are underweight because they’ve dropped and the sheet says to buy more VTSAX, it doesn’t feel like buying low in any triumphant sense. It feels uncomfortable. But the discomfort is absorbed by the process the decision was already made, weeks or months ago, when things were calm. The emotional charge drains out of it. That is a significant and under appreciated gift that a written system gives you.
When The Market Falls Fast: The Plan For Cash
Buying into a declining market does not get easier the more it falls. It gets harder. Each loss reinforces the fear that things will get worse before they get better, and that fear shrinks each subsequent purchase. This is not a character flaw it’s a near-universal human response to losing money in real time.
Which is why the plan for what to do with available cash in a downturn has to be written in advance, not improvised in the middle of one.
The structure we use: when cash sits outside the core portfolio proceeds from an asset sale, income that’s accumulated, any lump sum waiting to be deployed we don’t hold it open-endedly “waiting for the right moment.” We set predetermined thresholds. At what level of decline from recent highs does additional deployment make sense? How much goes in at each stage? Do purchase sizes increase as prices fall further, or stay flat?
The critical design principle is that purchase sizes should hold steady or increase at deeper price levels not shrink, which is what instinct pushes toward. More money goes in at -25% than at -15%, because the discount is larger and the fear is higher. The plan has to override the instinct, which is why it has to exist before the instinct kicks in.
Limit orders in Vanguard make this automatic. A standing buy order at a specific price executes without requiring any action in the moment. When markets open lower than expected and the order fills at a price below the target, that is not luck it is structure working exactly as designed. The best prices I’ve gotten have come from orders I set when calm and forgot about until the confirmation email arrived.
Asset Location: Where Things Live Matters As Much As What You Own
Different assets have different tax characteristics, and those characteristics interact differently with different account types. A little intentionality about which fund lives in which account reduces your tax bill every year compounding over decades into a real number.
The framework is straightforward:
Bonds generate income taxed as ordinary income the highest rate available. They belong inside a tax-deferred or Roth account, not a taxable brokerage.
International stock funds may generate a foreign tax credit, but only if held in a taxable account. If you want to capture that credit, the international position belongs in taxable.
US equity index funds with low dividend yields and low turnover are tax-efficient and can sit anywhere. They’re the most flexible.
Almost everything we hold sits in Roth IRA accounts. Every dollar grows tax-free. Every dividend and capital gain inside the account generates no current tax. At withdrawal, everything comes out clean. For a portfolio we expect to compound over decades, this structural advantage is enormous.
The HSA gets treated as a long-term investment account, not a healthcare spending reserve. We invest the balance in VTSAX rather than leaving it in cash or money market. The triple tax advantage pre-tax in, tax-free growth, tax-free out for qualified medical expenses makes it the most efficient single account in the entire stack. Every dollar invested there works harder than anywhere else.
The ESPP sits in its own mental category. Employer stock is single-company concentration risk. It does not belong in the same bucket as broadly diversified index funds. We watch the position and periodically diversify it into VTSAX rather than letting it accumulate to a size where it represents a meaningful portion of the total portfolio.
The Maryland rental and Fundrise accounts are real assets real estate exposure that serves a different function from the equity core. Different time horizon, different risk profile, different rules. They belong in a different mental bucket, managed separately from the compounding machine.
The Portfolio Right Now
For anyone who finds the specifics useful, here is where things actually stand as of mid-2026:
Core Roth IRA portfolio: 90% VTSAX, 10% VBTLX. Held in Roth accounts for both Leslie and me. Automated monthly contributions, rebalanced with new money, never sold.
Kids’ Roth IRAs: Lessie, MG, and Niko each have Roth IRAs opened when they were young, fully invested in VTSAX. Each earns modest household income that funds the contribution legally. The point is not the annual dollar amount it’s the 50-plus-year compounding runway those accounts now have.
HSA: $7,500 per year, invested in VTSAX. Treated as long-term wealth, not a healthcare spending account.
Roth 401k: $23,500 per year. Low-cost, broad-market index funds consistent with the core philosophy.
ESPP: Employer stock treated as a bonus accelerant, not a foundation. Diversified into the portfolio periodically.
Fundrise (three entities): Personal, OVI Realty LLC, OVI Consulting LLC $100/month each. Real estate exposure across three quiet streams.
Maryland rental (Wyoming LLC): $2,500/month in rent against a $1,416 mortgage. Net cash flow of roughly $300/month redirected to principal. The LLC cost $150 to set up and creates structural separation between this asset and our personal balance sheet.
Cash buffer: One to two years of living expenses in a high-yield savings account. Not invested. Its one job is to exist between us and any scenario where we’d be forced to sell investments at a bad moment.
Total invested annually: Approximately $81,900 across all vehicles on a $160,000 income. Savings rate roughly 51%.
What This Is Really About
The allocation ranges, the limit orders, the bear-market buying plan, the never-sell rule none of these exist because they’re mathematically optimal. They exist because they keep me from sabotaging myself.
I am not a detached rational actor standing outside the market, calmly executing the correct sequence of trades. I am a person who watched the portfolio drop in March 2020 and felt genuine fear. Who has held extra cash too long because re-entering felt impossible to time correctly. Who has wanted, more than once, to do something just to feel like I was responding to what was happening.
The systems exist for that version of me. Not the calm version sitting at the kitchen table writing this post. The version that will be staring at a 25% portfolio decline at 7:00am on a Tuesday while three kids need breakfast and the news is explaining why this time really is different.
The plan written in advance is better than the decision made in that moment. Every time. Not because the plan is perfect it isn’t. Because it was made when the information wasn’t distorted by fear, and because the act of having a plan removes the decision from the moment when decisions are most likely to go wrong.
We are still building. The machine is running. The rules are written. The work happening underground right now is the work that will eventually be visible.