Every article in this Bitcoin series eventually defers to one question it's time to answer head-on: how much? The question gets terrible answers from both poles — the maximalist's "as much as possible" (a conviction, not a size) and the dismisser's "zero" (a verdict, not an analysis) — and surprisingly few households ever derive their number the way they'd derive any other financial quantity: from their situation, their risk budget, and the asset's actual statistical character. This article runs that derivation: what the portfolio research on Bitcoin allocations actually shows (a younger, thinner literature than gold's, read honestly), the volatility math that converts "how much risk can this slice add?" into a percentage, the two human tests (survivable-size and sleep) that override every model, the life-situation adjustments that move the number — and the output format that matters more than the number itself: a written band with edges and rules, reviewed annually, governing a position instead of expressing a mood.
What the research and the math actually support
The evidence base, graded honestly: the portfolio studies: the academic and industry literature on Bitcoin allocations (mean-variance analyses, risk-parity frameworks, the fund-manager whitepapers) keeps finding that small allocations — commonly 1% to 5% — historically improved risk-adjusted returns of traditional portfolios (the asymmetric-payoff logic: an asset that can multiply contributes meaningfully at small size, while its crashes stay survivable), with the honest caveats attached: the backtests lean on Bitcoin's spectacular early returns (the past decade's entry prices no longer exist — forward returns are structurally more modest as the asset matures, per the market-cap article's arithmetic), the correlation story is unstable (Bitcoin traded like an uncorrelated orphan in some eras and like a leveraged tech stock in others — the diversification benefit is real but unreliable, unlike gold's steadier near-zero record), and the sample is one asset's single life (fifteen years containing a handful of full cycles — statistically thin next to gold's centuries, which is why every responsible framework sizes Bitcoin below gold, not beside it); the volatility math — the method that derives your number: the risk-budget approach: decide first what a total loss of the slice may cost the household (the risk budget — say, "we can absorb losing 3% of net worth without any life change"), and since Bitcoin's realistic drawdowns run 70–85% (and the honest planning case is 100% — the asset's tail includes zero in a way gold's doesn't), the allocation ≈ the risk budget: the household willing to lose 3% outright can hold ~3% (or slightly more if calibrating to the 80%-drawdown case rather than zero) — the derivation that replaces "what feels right" with one honest question about loss tolerance; the volatility-parity cross-check: Bitcoin's volatility runs several multiples of gold's (roughly 3–5× across recent eras), so a household comfortable with a 10% gold band gets equivalent risk-contribution from a 2–3% Bitcoin band — the arithmetic behind this blog's standing structure (gold the anchor, Bitcoin the satellite, sized at a fraction); and the range the methods converge on: for households that hold it at all, the frameworks keep landing in the 1–5% of net worth region (the aggressive tail stretching toward high single digits for young, hard-currency, high-conviction profiles; the conservative tail at 0–1%) — with the honest note that zero remains a fully defensible answer (the asset's tail risks are real, the sleep test below is sovereign, and this series' machinery never requires participation — only that whatever number is chosen be chosen deliberately).
The human tests that override every model
The math proposes; two tests dispose: the survivable-size test — the master control: the question asked with full honesty: if this position went to zero in a bad year — the exchange failure, the protocol catastrophe, the regulatory kill — would the family's life change? — the goals delayed? the obligations strained? the marriage stressed? — and the allocation set where the answer is genuinely "no": the test that catches what models can't (the household whose 5% is the emergency fund's twin sibling versus the one whose 5% is genuine surplus — same percentage, different survivability), and the reframe that makes it work: the position sized as pre-paid, non-refundable exposure — money the household has mentally spent on asymmetric optionality, whose return to zero would disappoint but not damage; the sleep test — the behavioral governor: the empirical check the mindset article demands: does the position's daily existence cost attention, checking compulsion, or actual sleep? — with the honest reading of failures: the position that demands hourly monitoring is oversized for this owner regardless of what any model blessed (risk tolerance is revealed by behavior, not claimed by questionnaire — the holder who panic-checks at 3 a.m. has already voted), and the resizing rule: sleep-test failures cut the position until the checking stops, and the number where calm returns is the household's true capacity — discovered, as it only ever is, in live conditions; and the couple's test joining them: the sizing conversation run jointly per the couples article (the consultation threshold covering crypto absolutely), with the practical rule that the household's size is the more conservative partner's size — the position both can hold through a 75% drawdown without the marriage renegotiating it monthly, because the mindset article's storms are weathered by households, not individuals, and the allocation one partner evangelized past the other's comfort is a divorce-grade drawdown waiting for its cycle.
The adjustments: what moves your number up or down
The base range personalized by situation — each factor a deliberate nudge, not a new model: horizon and age: longer horizons absorb Bitcoin's cycle-length drawdowns (the 30-year-old's 4% defensible where the 60-year-old's isn't — retirement's sequence-of-returns logic from the retirement article cutting crypto allocations hardest in the final working decade and the withdrawal years); income stability: the stable-salary household holds more volatility than the freelancer whose income already swings (the risk budget is shared between income risk and asset risk — the volatile-income household's crypto band sized after the buffer articles' machinery is fully funded, never before); currency situation — the regional factor: the soft-currency household's refuge budget (the currency-risk article's two-refuge weight) is real but already allocated — hard currency and gold carrying the refuge job with track records Bitcoin can't yet match — so Bitcoin's slice comes from the risk budget, not the refuge budget (the framing that prevents the common regional error: fleeing a melting pound into an asset that can drop 70% is exchanging one volatility for a larger one — the refuge layer wants boring; the satellite wants small); obligations load: the debt-service reality check (households above the comfortable debt-ratio thresholds hold 0% while expensive debt exists — the guaranteed return of retiring 20%+ card debt beating any asset's expected return, per the standing arithmetic), and the near-term goals exclusion (money with a date inside five years never rides the satellite — the bucket logic that outranks every allocation); knowledge and custody capacity: the honest self-assessment (the security articles' tiers demand real competence at self-custody scale — the household unwilling to run the discipline caps its size at what custodial-and-ETF convenience safely holds, which is a legitimate ceiling, not a failure); and the conviction adjustment, bounded: genuine researched conviction may move a household toward its range's top — with the mindset article's guardrail welded on: conviction expressed as +1–2% within a written band is analysis; conviction demanding 15% "because it's the future" is identity wearing a spreadsheet, and the band's ceiling exists precisely for that voice.
The output: a band with rules, not a number with feelings
The derivation's product, formatted for survival: the written band: the target with edges ("3%, band 1.5–4.5%") — because a point-number generates constant micro-decisions while a band generates rare, rule-driven ones: inside the band, do nothing (the default state, by design); above the ceiling (the bull market's gift — the position that grew past 4.5% by appreciation), trim back to target per the rebalancing rule (the discipline that harvests manias mechanically — selling into euphoria being emotionally impossible and procedurally trivial); below the floor (the crash's aftermath), the schedule continues and the review decides any top-up (the floor triggering attention, not automatic buying — the distinction that respects the drawdown clause); the funding and flow rules: the position built by schedule (the DCA articles' machinery — never the lump that concentrates regret), funded from designated surplus (the budget line that comes after obligations, buffers, and the core allocations — the sequencing that makes the survivable-size test true by construction), and the trims' destination pre-named (proceeds to the under-target layers per the review — the satellite feeding the portfolio, which was always its job); the review's annual questions: the percentage recomputed on the household's true total (the tracking article's unified inventory making the number honest), the band itself re-derived when life moves (the new child, the income change, the approaching goal — each one a legitimate band amendment through the annual process, never a mid-drawdown improvisation), the thesis hour run (the mindset article's steelman ritual — the position held by analysis or released by it), and the tests re-passed (survivable-size against the new net worth, sleep against the lived year); and the closing calibration the whole derivation serves: the difference between the households that Bitcoin helped and the ones it hurt was almost never the asset's performance — both groups rode the same charts — it was the size: the helped held positions small enough to keep through the crashes and meaningful enough to matter in the recoveries; the hurt held conviction-sized positions that the first real drawdown converted into forced sellers and sworn enemies — and the entire technology for landing in the first group is one honest afternoon: the risk budget named, the tests passed, the band written, and the number thereafter defended from every voice — including your own best-performing year's — that suggests it should be bigger.
Frequently asked questions
My Bitcoin grew from 3% to 9% of my net worth. Selling feels like betraying the winner. Do I really trim?
You wrote the band for exactly this feeling: the position tripled relative to everything else, which means the household now carries triple the crypto risk it chose on a calm day — and 'the winner' framing hides what a 9% Bitcoin slice does to your drawdown math (the next 75% decline now moves your net worth by ~7 points instead of ~2). The trim is not a verdict on Bitcoin's future; it's the harvest the strategy was designed around (sell-high is what uncorrelated volatility is FOR), the proceeds fund the layers the same review found under target, and the position remains at full target size for whatever comes next. The betrayal framing has it backwards: the band IS the strategy — abandoning it at the first profitable test is betraying the only decision-maker in the room with a track record: calm-day you.
Shouldn't young people go much higher — 20, 30% — since they have time to recover?
Time absorbs drawdowns; it doesn't absorb the behavioral and life realities that break big positions: the documented pattern is that oversized crypto allocations fail through humans, not math — the 25-year-old's 30% position meets a 78% drawdown as a 23-point net-worth crater, exactly when life's first big obligations arrive (the wedding, the visa, the down payment — young horizons are long but young LIQUIDITY needs are near), and the panic-sale near the bottom converts theoretical time-to-recover into realized loss. The defensible youth adjustment is the range's top, not its abandonment: high single digits for the genuinely surplus-funded, tested-nerves, obligations-light profile — built by schedule, banded, and trimmed — with the reframe that actually serves ambition: the young person's edge is DECADES of schedule contributions compounding, not one oversized bet's outcome; the salary's future purchases dwarf today's allocation, and protecting the habit beats maximizing the position.
Is 1% even worth the hassle? It can't change anything.
Run the honest arithmetic in both directions: at 1%, a total loss is a rounding error (the survivable-size test passed trivially) while the asymmetric case still registers — the historical multiplications that occurred would turn 1% into a portfolio-relevant contribution (and the band's trims along the way compound it further) — so 1% buys genuine optionality at negligible risk: the textbook asymmetry position. What 1% actually can't survive is the psychology trap: the holder who watches a small position 'work' and concludes the mistake was size, doubling and tripling outside any process (the gateway pattern the mindset article maps). If 1% is your number, its guardian rule is that increases route through the annual review only — and if the hassle objection is really about custody overhead, the modern answer is honest: at 1% scale, the vetted-custodial or ETF route from the wrappers article makes the position nearly maintenance-free, which is exactly the effort it deserves.
How does the Bitcoin band relate to my gold band — do they share a budget?
They share a philosophy and split a risk budget: both are no-issuer assets in the household's alternative layer, but their statistical characters differ enough to size separately — gold (lower volatility, centuries of refuge record, the crisis-behavior anchor) carries the layer's weight per the portfolio article's 2–15% situational bands, while Bitcoin (3–5× the volatility, one lifetime of record, the asymmetric-optionality profile) takes the satellite fraction — the practical ratio most derivations land near being gold at 3–6× the Bitcoin weight (e.g., 8% gold / 2% Bitcoin), which the volatility-parity math above independently supports. They interact at the review: both bands computed on the same true total, trims from either feeding the household's under-target layers, and the combined alternative-layer total itself sanity-checked (the two bands' SUM staying inside what the household's risk budget and refuge needs jointly justify) — one architecture, two instruments, each doing the job its track record actually supports.
Key takeaways
- The evidence supports small: portfolio studies keep landing at 1–5% (thin literature, honest caveats), volatility-parity math sizes Bitcoin at a fraction of the gold band, and zero remains fully defensible — the only indefensible number is an unchosen one.
- Derive, don't vibe: name the risk budget (what total loss costs nothing structural), set the allocation approximately equal to it, and cross-check against the 3–5× gold-volatility arithmetic.
- The human tests are sovereign: survivable-size (zero tomorrow changes no plans), sleep (checking compulsion means oversized), and the couple's rule (the more conservative partner's number governs).
- Adjust by situation: horizon, income stability, obligations load, custody competence — and for soft-currency households, fund Bitcoin from the risk budget, never the refuge budget: the refuge layer wants boring.
- Output a band with rules: target with edges, schedule-funded, trims pre-routed, annually re-derived — because the households Bitcoin helped held sizes they could keep, and size was always the entire game.
The closing image: two friends buy Bitcoin in the same month with the same enthusiasm. One sizes by conviction — 20% and climbing, because the research he read was all written by believers and the number felt like commitment — and the cycle does what cycles do: the drawdown arrives carrying a fifth of his net worth, the checking compulsion arrives with it, and the bottom finds him selling to make the feeling stop, education complete, position zero. The other spent one afternoon with a spreadsheet and a spouse: risk budget 3%, band written, schedule started, ETF wrapper chosen for the maintenance-free scale — and the same cycle passes through her household as a few calendar entries: two trims into the euphoria (proceeds to the tuition fund), the drawdown clause read once with coffee, the schedule never missing a month. Same asset, same cycle, same information available. He allocated a belief. She allocated a percentage — and the percentage, unlike the belief, was still there compounding when the next cycle began.
How Wajib AI helps
A percentage only governs what it can see: Wajib AI computes the household's true total — the crypto beside the gold, cash layers, and obligations — so the Bitcoin band's number is real, its drift is visible, and the rebalancing rule fires on facts instead of feelings.
Download Wajib AI free and keep every commitment, price, and payment in one place.