Allocation is a snapshot pretending to be a strategy
Every portfolio has an allocation. Not every portfolio has a target allocation, and the difference between those two is the difference between measuring and deciding.
A pie chart of current holdings shows a fact: this is what you hold now. It does not show whether that is where you meant to be. If you never set a target, a drift view is describing motion with no reference point.
Weight is not risk
The common assumption is that a large position is a large risk. That is true for one class of asset and badly wrong for another.
| Situation | Weight suggests | What actually drives the result |
|---|---|---|
| 60% in a large established asset | High exposure | Often the majority of the variation comes from here anyway |
| 25% in a thin small asset | Moderate exposure | A modest position can dominate returns in either direction |
| 10% in a stablecoin | Low exposure | Small variation, but a structural rather than market effect |
| 80% in one asset, the rest stable | Extreme concentration | Genuinely concentrated, but the risk may be intentional |
Position size and risk contribution are different measurements, and a view that shows only the first is giving you half the picture while looking complete.
Rebalancing drift
Without a target, allocation moves on its own. Assets that rise grow, assets that fall shrink, and the portfolio becomes more concentrated in whatever already worked — without anyone deciding that.
That is a defensible strategy, but only if it was chosen. The question a drift view should prompt is not "is my allocation wrong" but "did I ever state what I wanted".
Return per asset versus return per holding
Two numbers that get presented together are not the same measurement and do not add up.
A return measured from first purchase treats the entire holding period as one investment. That is rarely how it went. Money went in, some came out, more went in. A single figure over the whole period hides all of it, and the same asset can show a very different number depending on which entry you attribute it to.
A per-trade or per-lot view is more honest and considerably harder to read. Most dashboards show the first because the second does not fit on a screen.
Benchmarks flatter almost everyone
Comparing a portfolio against a broad index is the most common performance view on any dashboard, and the least informative for a typical crypto holder.
A concentrated portfolio will usually look worse against a broad index for years, and that says nothing about whether its owner made good decisions. Equally, a portfolio that happened to hold one asset for the whole period will beat the index through no skill whatsoever.
What a useful analytics view states
- Realised and unrealised separately. Not merged into one number that cannot be interpreted.
- A target, or an explicit note that there is none. Otherwise drift has nothing to be measured against.
- Risk contribution, not just weight. Size in the portfolio and size of the effect are different quantities.
- The method behind each figure. Average cost or lot-based, time-weighted or money-weighted, named on the screen rather than in documentation.
- Coverage. Which accounts are in, which are not, and when each was last updated.
Every one of those is a disclosure rather than a feature. Their absence is the reason so many portfolio screens can be entirely wrong while looking plausible.
Next: when portfolio numbers do not match