SatoshiMacro

$BTC There are two ways to trade Bitcoin from Australia: own the coin on an exchange, or trade its price through a crypto CFD with an ASIC-regulated broker. I spent years on an institutional derivatives desk, and the pattern I see with retail traders is that almost nobody chooses between the two deliberately. They end up in one or the other by accident, and the accident is expensive. This is the deliberate version of the choice.
What a crypto CFD actually is
A contract for difference is a leveraged derivative. You never own Bitcoin. You open a position against a broker, and your profit or loss is the difference between your open and close price, settled in your account currency. That one design fact drives everything else: the leverage, the costs, the tax treatment, and the situations where a CFD is the right tool.
The ASIC rulebook
Australia runs one of the stricter retail CFD regimes in the world, and in my view that is a feature. Retail crypto CFD leverage is capped at 2:1, the lowest cap of any asset class (major forex pairs get 30:1, shares get 5:1). Negative balance protection means you cannot lose more than what is in the account. Client money sits in segregated trust accounts, and AFCA membership gives you a dispute channel that can make binding awards. The 2:1 cap exists because crypto routinely moves 10 percent in a session; at the old pre-2021 leverage levels those sessions simply deleted accounts.
What it really costs
Two costs matter. The spread is the visible one. The invisible one is overnight funding: a daily financing charge on leveraged positions that compounds for as long as you hold. For a trade measured in hours or days, funding is noise. For a position held for months, it becomes the dominant cost and quietly eats the thesis. This is the single most common mistake I see: using a trading instrument to make an investing bet.
When the CFD is the right tool
Three cases. First, the short side: a CFD lets you sell first with one click, which is the cleanest way for an Australian retail trader to be short Bitcoin. Second, hedging: if you hold spot and want to protect against a drawdown without selling (and without triggering a CGT disposal event on your coins), a short CFD against the holding does exactly that. Third, short-term tactical trades where custody, wallet security and on-ramp friction are not worth the setup cost for a position you plan to close within the week.
When spot wins, and it usually does
If your time horizon is measured in months, buy the coin. Two reasons. The first is the funding drag above. The second is tax: for an individual investor, spot crypto held longer than 12 months qualifies for the 50 percent CGT discount. CFD profits are ordinary income under the ATO's treatment, taxed at your full marginal rate with no discount, ever, because you never own an asset. At the 47 percent top marginal rate, that difference is not a rounding error; it is the largest single number in the whole comparison.
The honest part
ASIC-mandated disclosures show 70 to 85 percent of retail CFD accounts lose money. That statistic is about leverage and sizing behaviour, not about the instrument being rigged, but it should calibrate you. My read: most people should hold spot, and should only touch the CFD side once they have a tested short-term edge or a genuine hedging need. The instrument is a scalpel, and most accounts die from using it as a hammer.
I keep a full plain-language guide to crypto CFD trading in Australia (leverage maths, worked cost examples, the tax split, and how the ASIC protections actually work) here: https://satoshimacro.com/guides/forex/crypto-cfd-trading-australia/
Not financial advice. I am a former institutional trader, not your adviser, and CFDs are a high-risk product.
#SatoshiMacro #CryptoCFD #RiskManagement #Bitcoin

$BTC On the institutional desks I worked on, the week started with a pack: one document, every chart that mattered, reviewed in a fixed order before any position discussion was allowed. Most retail crypto traders do the opposite. They graze - a funding-rate screenshot on X here, a Glassnode free-tier chart there, a dominance chart from a third site with different date ranges and different currencies. Then they wonder why their view changes with whichever chart they saw last.
I rebuilt the desk-pack habit as a single free page: the SatoshiMacro Crypto Charts Dashboard. This is how I actually use it on a Monday morning, in reading order.
First pass: valuation. Where are we in the cycle?
The cycle indicators come first because they set the context every other chart gets read inside. Mayer Multiple (price against the 200-day moving average), Pi Cycle Top and Pi Cycle Bottom, MVRV Z-Score, Benjamin Cowen's Risk Metric, the Power Law regression, the Rainbow bands, and the 200-week moving average heatmap. No single one of these is reliable alone - Pi Cycle famously missed the November 2021 echo top - which is exactly why the page shows them side by side. Confluence over any single chart is the whole discipline.
Second pass: rotation. Bitcoin dominance and the Altcoin Season Index tell you where risk appetite sits inside crypto. Rising dominance with weak alt breadth is a very different market from the same Bitcoin price with alts running.
Third pass: positioning. Funding rates, aggregate open interest, and the futures basis. These are the fast-moving charts and the ones most likely to mark a local extreme rather than a cycle turn. I read them last among the crypto charts precisely because they are the noisiest.
Fourth pass: macro. DXY, M2 money supply, the yield curve, VIX. Five minutes, not thirty. Crypto is a liquidity asset; the macro pass is about knowing which way the tide is leaning, not about forecasting the Fed.
The distillation layer
The dashboard also carries the SatoshiMacro Model, the 48-signal confluence model I built across six weighted tiers, which compresses this whole review into one calibrated 0-to-100 reading. Backtested against every major Bitcoin cycle inflection since 2013, the calibrated model lands all seven in their target zones. But I still read the individual charts weekly, because the model is a position classifier, not a forecaster - an honest limitation that applies to every chart on the page. None of this predicts next week. It tells you where you are, so your sizing matches the environment.
Two Australian-specific details, because the page is built AUD-native: every price chart is denominated in AUD (with a one-click USD toggle that remembers your preference), and for AU tax residents the cycle context feeds directly into the 12-month CGT discount decision - knowing where you sit in the cycle is half of deciding whether a parcel is worth holding past its discount date.
The whole page is free, no signup, no paywall tiers: https://satoshimacro.com/tools/crypto/dashboard/
My read on process, after years of doing this professionally: the traders who survive are not the ones with the best single indicator. They are the ones who look at the same fixed set of evidence every week, in the same order, and size accordingly. A dashboard does not give you an edge. It gives you a discipline, which is rarer.
Not financial advice. Former institutional trader, Sydney.
#SatoshiMacro #BitcoinCycleAnalysis #OnChain #Bitcoin

$BTC There are two ways to trade Bitcoin from Australia: own the coin on an exchange, or trade its price through a crypto CFD with an ASIC-regulated broker. I spent years on an institutional derivatives desk, and the pattern I see with retail traders is that almost nobody chooses between the two deliberately. They end up in one or the other by accident, and the accident is expensive. This is the deliberate version of the choice.
What a crypto CFD actually is
A contract for difference is a leveraged derivative. You never own Bitcoin. You open a position against a broker, and your profit or loss is the difference between your open and close price, settled in your account currency. That one design fact drives everything else: the leverage, the costs, the tax treatment, and the situations where a CFD is the right tool.
The ASIC rulebook
Australia runs one of the stricter retail CFD regimes in the world, and in my view that is a feature. Retail crypto CFD leverage is capped at 2:1, the lowest cap of any asset class (major forex pairs get 30:1, shares get 5:1). Negative balance protection means you cannot lose more than what is in the account. Client money sits in segregated trust accounts, and AFCA membership gives you a dispute channel that can make binding awards. The 2:1 cap exists because crypto routinely moves 10 percent in a session; at the old pre-2021 leverage levels those sessions simply deleted accounts.
What it really costs
Two costs matter. The spread is the visible one. The invisible one is overnight funding: a daily financing charge on leveraged positions that compounds for as long as you hold. For a trade measured in hours or days, funding is noise. For a position held for months, it becomes the dominant cost and quietly eats the thesis. This is the single most common mistake I see: using a trading instrument to make an investing bet.
When the CFD is the right tool
Three cases. First, the short side: a CFD lets you sell first with one click, which is the cleanest way for an Australian retail trader to be short Bitcoin. Second, hedging: if you hold spot and want to protect against a drawdown without selling (and without triggering a CGT disposal event on your coins), a short CFD against the holding does exactly that. Third, short-term tactical trades where custody, wallet security and on-ramp friction are not worth the setup cost for a position you plan to close within the week.
When spot wins, and it usually does
If your time horizon is measured in months, buy the coin. Two reasons. The first is the funding drag above. The second is tax: for an individual investor, spot crypto held longer than 12 months qualifies for the 50 percent CGT discount. CFD profits are ordinary income under the ATO's treatment, taxed at your full marginal rate with no discount, ever, because you never own an asset. At the 47 percent top marginal rate, that difference is not a rounding error; it is the largest single number in the whole comparison.
The honest part
ASIC-mandated disclosures show 70 to 85 percent of retail CFD accounts lose money. That statistic is about leverage and sizing behaviour, not about the instrument being rigged, but it should calibrate you. My read: most people should hold spot, and should only touch the CFD side once they have a tested short-term edge or a genuine hedging need. The instrument is a scalpel, and most accounts die from using it as a hammer.
I keep a full plain-language guide to crypto CFD trading in Australia (leverage maths, worked cost examples, the tax split, and how the ASIC protections actually work) here: https://satoshimacro.com/guides/forex/crypto-cfd-trading-australia/
Not financial advice. I am a former institutional trader, not your adviser, and CFDs are a high-risk product.
#SatoshiMacro #CryptoCFD #RiskManagement #Bitcoin

$BTC There are two ways to trade Bitcoin from Australia: own the coin on an exchange, or trade its price through a crypto CFD with an ASIC-regulated broker. I spent years on an institutional derivatives desk, and the pattern I see with retail traders is that almost nobody chooses between the two deliberately. They end up in one or the other by accident, and the accident is expensive. This is the deliberate version of the choice.
What a crypto CFD actually is
A contract for difference is a leveraged derivative. You never own Bitcoin. You open a position against a broker, and your profit or loss is the difference between your open and close price, settled in your account currency. That one design fact drives everything else: the leverage, the costs, the tax treatment, and the situations where a CFD is the right tool.
The ASIC rulebook
Australia runs one of the stricter retail CFD regimes in the world, and in my view that is a feature. Retail crypto CFD leverage is capped at 2:1, the lowest cap of any asset class (major forex pairs get 30:1, shares get 5:1). Negative balance protection means you cannot lose more than what is in the account. Client money sits in segregated trust accounts, and AFCA membership gives you a dispute channel that can make binding awards. The 2:1 cap exists because crypto routinely moves 10 percent in a session; at the old pre-2021 leverage levels those sessions simply deleted accounts.
What it really costs
Two costs matter. The spread is the visible one. The invisible one is overnight funding: a daily financing charge on leveraged positions that compounds for as long as you hold. For a trade measured in hours or days, funding is noise. For a position held for months, it becomes the dominant cost and quietly eats the thesis. This is the single most common mistake I see: using a trading instrument to make an investing bet.
When the CFD is the right tool
Three cases. First, the short side: a CFD lets you sell first with one click, which is the cleanest way for an Australian retail trader to be short Bitcoin. Second, hedging: if you hold spot and want to protect against a drawdown without selling (and without triggering a CGT disposal event on your coins), a short CFD against the holding does exactly that. Third, short-term tactical trades where custody, wallet security and on-ramp friction are not worth the setup cost for a position you plan to close within the week.
When spot wins, and it usually does
If your time horizon is measured in months, buy the coin. Two reasons. The first is the funding drag above. The second is tax: for an individual investor, spot crypto held longer than 12 months qualifies for the 50 percent CGT discount. CFD profits are ordinary income under the ATO's treatment, taxed at your full marginal rate with no discount, ever, because you never own an asset. At the 47 percent top marginal rate, that difference is not a rounding error; it is the largest single number in the whole comparison.
The honest part
ASIC-mandated disclosures show 70 to 85 percent of retail CFD accounts lose money. That statistic is about leverage and sizing behaviour, not about the instrument being rigged, but it should calibrate you. My read: most people should hold spot, and should only touch the CFD side once they have a tested short-term edge or a genuine hedging need. The instrument is a scalpel, and most accounts die from using it as a hammer.
I keep a full plain-language guide to crypto CFD trading in Australia (leverage maths, worked cost examples, the tax split, and how the ASIC protections actually work) here: https://satoshimacro.com/guides/forex/crypto-cfd-trading-australia/
Not financial advice. I am a former institutional trader, not your adviser, and CFDs are a high-risk product.
#SatoshiMacro #CryptoCFD #RiskManagement #Bitcoin

$BTC
Every June I get the same question from Australian crypto holders. "If I sell some Bitcoin before 30 June, what does the tax actually look like?" The honest answer is that most people calculate it wrong, and the errors are expensive.
Here is the framework I used on the desk and still run on personal capital.
Capital gains tax in Australia is not a separate tax. Your net capital gain gets added to your assessable income under section 6-5 ITAA 1997 and taxed at your marginal rate. For a top-bracket earner that is 45 percent plus the 2 percent Medicare levy, so 47 percent on the gain. There is no flat crypto rate here, unlike some offshore regimes that tax at a fixed 30 percent.
The single most valuable lever is the 12-month rule. Hold a parcel more than 12 months before you dispose of it and you get the 50 percent CGT discount as an individual. Sell at 11 months and you are taxed on the full gain. That one day of holding period can move the effective rate from 47 percent to 23.5 percent on the same parcel.
A worked example. Say you bought 0.5 BTC at A$40,000 and you dispose at A$100,000. The gross gain is A$60,000. Held under 12 months, the whole A$60,000 is assessable, roughly A$28,200 in tax at the top marginal rate. Held over 12 months, the discount halves the assessable gain to A$30,000, so roughly A$14,100. Same trade, A$14,100 difference, decided entirely by the calendar.
Three things people get wrong.
First, the disposal date is the contract date, not the settlement date and not the day the AUD lands via PayID. Crypto-to-crypto swaps are disposals too. Swapping BTC for ETH triggers CGT on the BTC even though no Australian dollars moved.
Second, cost base includes brokerage and acquisition costs, and you choose a parcel identification method. Specific identification lets you dispose of the highest-cost parcels first to minimise the gain, but only if your records are clean and consistent. The ATO expects consistency.
Third, capital losses. If you hold underwater alt positions, a crystallised loss offsets your BTC gain in the same year, and unused losses carry forward indefinitely. This is the legitimate version of tax-loss harvesting. Watch Part IVA: the ATO can deny a loss if the dominant purpose was the tax benefit and you rebought immediately. There is no fixed wash-sale day count in Australian law, but selling and rebuying within minutes invites scrutiny.
One honest limitation. The 50 percent CGT discount for individuals has been flagged for review in the forward Budget cycle, but as of 30 June 2026 it remains the law. Plan against the rules that exist, not the rules that might arrive.
What I would actually do before EOFY. Map every parcel by acquisition date and cost base. Flag which parcels cross the 12-month line before 30 June. Model the gain at your real marginal rate, not a round guess. Then size the disposal. A free Crypto CGT calculator that does the discount maths and marginal-rate layering is here: https://satoshimacro.com/crypto/cgt-calculator/
This is general information from a former institutional trader, not tax advice. Confirm your position with a registered agent before you act.
#SatoshiMacro #CryptoTax #EOFY #Bitcoin

The most expensive mistake I made in 2018 was running a single-indicator Bitcoin cycle framework. The indicator was MVRV Z-Score on the original Awe and Wonder formula. The reading was clear in late 2017 at Z above 11. The exit thesis was correct. What broke was 2018: holding through a -84 percent drawdown to USD 3,200 because the same indicator never re-printed a Z reading that matched a textbook cycle-bottom anchor. The framework worked at the top and failed at the bottom of the same cycle.
Single-indicator frameworks have a worse track record than the literature suggests. Survivorship bias is real - the Twitter accounts you see citing one magic indicator are the ones whose magic indicator happened to fire on the most recent cycle. The 2021 cycle proved this brutally: it ran as a double-top structure with a 5-month gap between the April peak and the November echo peak, and that structure broke the signal cleanly across every moving-average-based cycle indicator in continuous use. The piece below walks through the six BTC valuation signals I now treat as a confluence panel rather than as standalone triggers, the structural blind spots each one carries, and the multi-tier weighting framework that catches what any individual signal misses.
## Pi Cycle Top
The simplest classical cycle-top indicator. Signal fires when the 111-Day Moving Average crosses above the 350-Day Moving Average multiplied by 2. Two moving averages, one multiplier, one crossover. Pure price-history math, no on-chain data needed.
Historical fires: 2013-04 (first 2013 peak, fired ~17 days before local peak), 2017-12-17 (fired 1 day before the cycle top), 2021-04-12 (fired 4 days before the April peak). Extraordinarily precise when it does fire - the window has been zero to four days.
The 2021-11 miss: did NOT fire at the November echo peak even though spot made a fresh all-time high. The double-top structure with a 5-month gap allowed the 350DMA to catch up between peaks, suppressing the crossover. Documented blind spot that will repeat on any future double-top cycle structure.
Limitation: binary fire/no-fire mechanic. Cannot express partial conviction. Single Pi Cycle fire by itself is not the level of confidence at which I take cycle-exit action on real capital.
## Mayer Multiple
BTC price divided by 200-day moving average. One ratio, one input. Trace Mayer published the formula in 2013-2014.
Threshold bands per SatoshiMacro implementation: above 2.4 historically marks cycle euphoria, below 0.6 marks cycle accumulation. Documented ATH 2.42 (2017-11), documented ATL 0.52 (2022-06 Luna collapse).
What Mayer does well: cleanly continuous output. Supports graduated position-sizing logic. Replicable with five lines of Python. No on-chain data dependencies.
What Mayer cannot do: distinguish between stretched late-cycle and early-mania still going. Mayer at 2.6 in October 2017 looked the same as Mayer at 2.6 in February 2021 - both were 8 to 12 weeks pre-top. Reading is ambiguous on directional timing. Also compresses on echo-top structures where the 200DMA catches up between peaks - same 2021-11 blind spot as Pi Cycle.
## MVRV Z-Score
Awe and Wonder's 2018 formula: Z equals (MarketCap minus RealizedCap) divided by stddev of MarketCap on an expanding window. Measures how stretched aggregate cost basis is versus current market value.
True Glassnode/CoinMetrics implementation anchors: 2013-12 Z approximately 9.3, 2017-12 Z approximately 11.5, 2021-04 Z approximately 6.3, 2021-11 Z approximately 2.9 (documented miss), 2018-12 Z approximately -0.3, 2022-11 Z approximately 0.1.
The 2021-11 miss is structural and important. Coins moved during the May-July 2021 crash, partially catching up the RealizedCap baseline. By November the Z-Score had compressed even though spot made a fresh ATH. Same root cause as Pi Cycle and Mayer's 2021-11 misses - moving-average-based indicators do not handle double-top cycles cleanly.
## Power Law
Giovanni Santostasi's 2018 work on Bitcoin's log-time / log-price regression. BTC price-versus-time-since-genesis fits a power-law curve with cycle peaks and troughs printing as statistical deviations from the fair-value line.
SatoshiMacro implementation uses sigma units (standard deviations from the regression). Historical anchors: 2017-12 plus 2.50 sigma, 2021-03 plus 1.94 sigma, 2015-09 minus 1.71 sigma, 2022-12 minus 1.45 sigma. Pattern: cycle peaks have printed progressively smaller sigma deviations each cycle, consistent with the diminishing-cycle-amplitude thesis.
Limitation: long-horizon position-classifier, not tactical-timing forecaster. Most useful at the extremes (below minus 1.0 sigma equals aggressive sizing in, above plus 1.5 sigma equals aggressive sizing out). Less useful in the mid-band where it currently sits.
## Rainbow Chart
Logarithmic regression with nine colour-banded zones from Maximum Bubble (red, top) to Basically a Fire Sale (deep blue, bottom). The most-shared cycle visual on crypto Twitter.
What Rainbow does: visual quick-read on where price sits versus the long-term log-regression base. HODL band historically covers approximately 35-40 percent of cycle time. The two upper bands (FOMO + Maximum Bubble) cover roughly 8-12 percent of days, clustered at cycle tops.
What Rainbow does not do: forecast anything. The bands are a visual, not a probabilistic model. Useful as a sentiment-anchor when crypto Twitter is screaming "Maximum Bubble" but the chart shows HODL - that mismatch is the alpha signal.
## Bitcoin Risk Metric
Benjamin Cowen's 0-to-1 normalised cycle-position scale, in continuous use since 2018. Combines log-price deviation from a long-term trendline with a smoothed momentum overlay.
Historical anchors per SatoshiMacro implementation: 2017-12 Risk at 0.90, 2021-03 Risk at 0.78, 2015-09 Risk at 0.00, 2022-12 Risk at 0.06. Pattern: peaks above 0.78, bottoms below 0.10. Middle 0.10 to 0.78 band is mid-cycle.
Where Risk Metric differs from the binary indicators: continuous and graduated all the way through. Natively suited to position-sizing functions like "Risk above 0.80 equals sale-only stance; Risk 0.40 to 0.60 equals 40 percent position; Risk below 0.20 equals 100 percent sizing-in tranches".
## The six-tier confluence framework
No single one of the above six valuation signals survived the 2021-11 echo top cleanly. Pi Cycle did not fire. Mayer compressed below its prior cycle anchors. MVRV printed only 2.9 versus 11.5 in 2017. Power Law printed plus 1.0 sigma versus plus 2.5 in 2017. Rainbow showed FOMO band but not Maximum Bubble. Risk Metric at 0.78 (cycle-top territory but lower than the prior cycle's 0.91).
The lesson: structural blind spots are structural. Any cycle with double-top potential will produce these same compressions across all six. The right response is not to find a seventh magic indicator. The right response is to weight a basket of valuation signals as one tier, alongside five other dimensions (Cycle Timing, Sentiment, Rotation, Miner Economics, Macro) in a confluence model.
The SatoshiMacro Model wraps all six valuation signals into Tier 1 (25 percent weight) and combines them with Tier 0 (Cycle Timing, 30 percent), Tier 2 (Miner Economics, 10 percent), Tier 3 (Sentiment, 20 percent), Tier 4 (Rotation, 10 percent), and Tier 5 (Macro, 5 percent). The composite 0-100 reading classifies cycle position (Accumulation / Mid-cycle / Caution / Distribution / Cycle Top) and the historical calibration produces 7-of-7 in-zone cycle calls across all four documented BTC cycle inflections.
## For Australian residents specifically
The single highest-EV improvement I made to my cycle-exit framework after the 2018 mistake was wrapping the tactical signal in 12-month CGT-discount-eligibility logic. The 50 percent CGT discount at the 47 percent top marginal rate drops effective CGT from 47 percent to 23.5 percent on parcels held longer than 12 months. The asymmetry exceeds the tactical alpha from selling 1 to 8 weeks early on most realistic AU-resident scenarios.
Practical rule: a parcel at month 10 of its 12-month hold sitting at MVRV Z elevated, Pi Cycle fired, Risk Metric above 0.80, and three other tiers confirming distribution is almost always better held for the 2-month eligibility crossing than sold immediately. The tax-discount asymmetry is the dominant variable, not the tactical-timing alpha.
## Full model and calibration receipts
The complete six-tier confluence model with live composite reading, all 48 signals across the six tiers, and the 7-of-7 BTC plus 4-of-5 ETH historical calibration:
https://satoshimacro.com/tools/crypto/satoshimacro-model/
Disclosure: I built and maintain SatoshiMacro. The model is free and ad-supported (broker affiliate links on the main site, not in this Article). This Article is editorial, not financial advice.
#SatoshiMacro #BitcoinCycle #OnChain #Bitcoin

$BTC | Australia''s 2026 EOFY closes on 30 June. That is 33 days from today. For Australian-resident crypto holders sitting on unrealised losses, this is the window where the tax math materially changes by deliberate action rather than market direction. I will walk through the Part IVA framework, cost-base method choice, and the carry-forward record-keeping that 90 per cent of retail traders get wrong.
## The mechanic
If you hold a crypto asset at a capital loss, you can crystallise that loss by disposing of it during the 2025 to 2026 financial year. The realised capital loss offsets capital gains in the same income year. Any unused balance carries forward indefinitely against future capital gains.
The disposal must be genuine. Sending the asset to another wallet you control is not a disposal. Swapping to a related coin under common control structures is not a disposal. The ATO treats these as the same asset for cost-base purposes. Genuine disposal means selling to a third party at market price.
## Where it goes wrong: ATO Part IVA wash-sale framework
Australia does not have a fixed 30-day wash-sale rule the way the US does. What we have is Part IVA of the ITAA 1936, an anti-avoidance framework. Taxation Determination TD 2008/29 and the ATO''s broader wash-sale guidance treat a transaction sequence as a wash sale if the dominant purpose of the disposal-and-reacquisition was to obtain the tax benefit, not a genuine commercial purpose.
There is no fixed waiting period. There is a substantive test of dominant purpose. The ATO''s case studies suggest that a same-day sell-and-rebuy is almost certainly Part IVA. A 14-day gap with a documented commercial reason is more defensible. A 30-day-plus gap with intervening market movement is generally fine.
The practical guidance: aim for at least 30 days between disposal and reacquisition, and document the commercial reasoning at the time of disposal, not retroactively when the ATO asks.
## Cost-base method choice
Australia allows multiple cost-base methods for crypto: First-In-First-Out (FIFO), Last-In-First-Out (LIFO), Highest-In-First-Out (HIFO), and specific-identification. The method must be applied consistently within an asset class and documented.
For tax-loss harvesting purposes, HIFO and specific-identification maximise the realised loss on disposal. But switching methods to optimise loss-realisation in the harvesting year and then switching back is a Part IVA red flag. Pick one method, apply it consistently, and document the choice in your tax records.
## Carry-forward records
Unused capital losses carry forward indefinitely against future capital gains in Australia. But ONLY if you correctly report them on the relevant year''s tax return. The ATO myGov pre-fill does not populate capital losses automatically. It pulls disposal data from the major exchanges but does not compute the loss-vs-gain math. You must claim the loss on Schedule of Capital Gains and carry the unused balance forward year-on-year via the same line item.
I have seen Australian-resident traders lose access to AUD 50,000-plus in carry-forward losses because the loss was not reported in the year of realisation, and the ATO''s amended-return window had closed by the time the gain arrived to offset against.
## The 12-month CGT discount layer
If you have a long-held position you are considering selling at a loss to harvest, ask first whether the gain side has positions held over 12 months that would qualify for the 50 per cent CGT discount on disposal. Realising a loss before realising a 50-per-cent-discounted gain is suboptimal. The loss netting against a discounted gain only reduces the gain at half the dollar value of the loss. In some configurations, harvesting the loss in this year while deferring the gain disposal until a year you have less ordinary income produces a better after-tax result than netting them in the same year.
## Worked example
A holder has AUD 100,000 unrealised loss in altcoin position A and AUD 80,000 unrealised gain in BTC held over 12 months. If both are realised in 2025 to 2026:
- Loss: AUD 100,000 capital loss.
- Gain: AUD 80,000 reduced to AUD 40,000 after 50 per cent CGT discount.
- Net: AUD 60,000 unused capital loss carries forward.
Versus realising only the loss in 2025 to 2026 and deferring the gain to 2026 to 2027 when the holder will be in a lower marginal bracket:
- 2025 to 2026: AUD 100,000 unused capital loss carries forward.
- 2026 to 2027: AUD 80,000 gain, half discounted to AUD 40,000, fully offset by carried-forward loss. AUD 60,000 unused loss carries forward to 2027 to 2028.
Same total tax outcome but different cash-flow timing and bracket-shifting flexibility.
The free EOFY-timed calculator I built runs this math end-to-end including the Part IVA timing buffer:
https://satoshimacro.com/crypto/tax-loss-harvesting-calculator/
The tool is free, ad-supported (broker affiliate links on the main site, not in the calculator).
Disclosure: I built and maintain SatoshiMacro. This is editorial, not financial advice. Talk to a registered tax agent before executing.
#SatoshiMacro #CryptoTax #EOFY #Bitcoin

$BTC | The Bitcoin Power Law is the cycle-confluence stack''s oldest valuation gauge that still works. Giovanni Santostasi and Harold Christopher Burger popularised the formulation in research from 2018 onward: take Bitcoin''s price history, plot it against time on a log-log axis, fit a linear regression, and you get a slope of approximately 5.7. Above the regression mean by one standard deviation marks cycle-top territory. Below by 0.5 standard deviations marks accumulation territory.
I run it inside Tier 1 Valuation at 25 per cent weight in the SatoshiMacro Model, alongside MVRV Z-Score and Mayer Multiple. Here is what the model has caught at every documented BTC cycle inflection since 2013, and where the methodology stops being load-bearing.
## What the +1 standard-deviation band has caught
December 2013 cycle top: price at approximately USD 1,100 pinned the +1 band almost exactly. The model fired cycle-top territory in the December 7 to 17 window.
December 2017 cycle top: USD 19,500 at the peak, the +1 band sat at approximately USD 18,800 at that date. The model flagged cycle-top from late November onward and held through the December peak.
April 2021 cycle top: USD 64,800 print versus +1 band at approximately USD 62,000. Cycle-top window flagged from late March through mid-April.
Late 2025 cluster: BTC ATH at USD 126,198 on 2025-10-06 versus +1 band at approximately USD 118,000. Cycle-top territory flagged from September through October.
## What the -0.5 standard-deviation band has caught
January 2015 cycle bottom: USD 178 print, model in accumulation zone from late 2014 through early 2015.
December 2018 cycle bottom: USD 3,200 print, model in accumulation from October 2018 through February 2019.
November 2022 cycle bottom: USD 15,700 print, model in accumulation from FTX-collapse window through January 2023.
That is every confirmed BTC cycle inflection since 2013, caught by one model with two band thresholds. Most other valuation indicators (Mayer Multiple, MVRV Z-Score) have a similar record but each has a documented miss. The Power Law''s closest near-miss is the November 2021 echo top. Mayer fired, Pi Cycle Top did not, Power Law was already inside the +1 band from April. The honest framing is that none of these indicators is a precision instrument; they are position classifiers that work in confluence, not in isolation.
## Honest limitations
The slope claim of approximately 5.7 is descriptive, not predictive. If Bitcoin''s adoption trajectory bends, and history says adoption trajectories of every previous technology bend somewhere, the slope drifts. The model assumes the past 15-year regression line continues, which it might not.
The standard-deviation bands compress over time as more data points fill in. A standard deviation in 2014 with five years of price history is wider than a standard deviation in 2026 with sixteen years. This means the model has been getting more precise mechanically, not because Bitcoin has gotten more predictable.
The 2021-11 echo top sat inside the +1 band from April through November. A trader using only Power Law to time the top would have been holding through the entire summer drawdown and the November round-trip. This is why confluence matters.
## Australian-resident framing
The +1 / -0.5 band structure maps cleanly onto a 12-month CGT discount ladder. Tranche your distribution across the +1 band window rather than trying to nail the single peak date. Keep accumulation tranches purchased outside the previous 12 months so the 50 per cent discount stays available on disposal. The Power Law band gives you a structured cycle-position framework; the ATO''s 12-month rule gives you the tax math overlay.
## Where Power Law sits inside SatoshiMacro
The Power Law deviation feeds Tier 1 Valuation at 25 per cent of total model weight. Tier 1 also carries MVRV Z-Score (currently at 0.27 in near-realised-value band), Mayer Multiple, and Pi Cycle Top and Bottom signals. When three or more Tier 1 signals fire simultaneously the model surfaces a cycle-position warning above the gauge.
The full chart with current reading and historical zone overlays:
https://satoshimacro.com/tools/crypto/cycle-indicators/bitcoin-power-law/
Disclosure: I built and maintain SatoshiMacro. The model is free and ad-supported (broker affiliate links on the main site, not in this post). This post is editorial, not financial advice.
#SatoshiMacro #BitcoinPowerLaw #OnChain #Bitcoin