Racing your equity curve against something is the only way to find out whether a year was good. On its own a rising curve says almost nothing, because most of what moves a portfolio is the market moving, and you did not do that. The trouble is that the choice of comparison line decides the answer, and the line usually gets chosen after the reader already knows how the year went. That is not a benchmark. It is a way of feeling a specific way about a number you already had.
A word on what I am working from. The capture of the Performance tab below shows the view controls, the summary tiles and the top of the Account Equity panel. I cannot see a benchmark picker in it, so I am not going to tell you it offers four options or name them, because I would be making that up. What I can give you is the rule for choosing the line, which is the part that actually decides whether the comparison is worth anything, and it applies to whatever options your screen turns out to offer.
Three tests a comparison line has to pass
A benchmark is a claim about what you would have done instead. That gives you three tests, and a line has to pass all of them.
It has to be investable without you. If you could not actually have bought and held it for the whole period, with real money, through the same accounts, it is not an alternative you gave up. This is the test that eliminates most exotic comparisons and quite a few index constructions that assume free rebalancing.
It has to carry roughly the risk you carried. Beating a bond index with a leveraged crypto book is not information. If your drawdowns are three times the benchmark's drawdowns, you are being paid for risk rather than for skill, and the race is rigged in your favour in a way that will reverse violently at some point.
It has to be fixed before the period starts. Write it down. This is the test everyone fails, and it is the only one that costs nothing to pass. Pick the line in January, note it somewhere you cannot quietly edit, and read the result in December whichever way it goes.
The line most people should race first
Before you reach for an index, race yourself. Take whatever you were holding at the start of the period, freeze it, and ask what that untouched book would be worth today. No trades, no rotations, no tactical exits. That is the do nothing benchmark, and for a retail reader it answers the only question that has a decision attached to it, which is whether the activity was worth doing.
Every trade you place costs a spread, a commission and, on a leveraged position, funding. Every trade also costs attention, which is the part nobody prices. If your active curve has not beaten your own frozen starting book, the correct response is not a better strategy. It is fewer trades. That is a change you can make this week, and it is the only benchmark result that translates directly into an action rather than a feeling.
Work out roughly what activity costs you before you run the comparison, because it sets the bar. Say you place four trades a week and each round trip costs you thirty dollars all in between spread and fees. That is a hundred and twenty a week, call it six thousand a year. On a fifty thousand dollar book that is twelve percent of capital consumed annually before a single decision has to be right. Your active management has to beat the frozen book by more than that just to break even on the effort.
A mixed book needs a blended line, not a famous one
If you hold both crypto and equities, neither of the obvious lines works. Race a mixed book against bitcoin and you will lose in every crypto bull run and win in every crash, for reasons that have nothing to do with you. Race it against a broad equity index and the same thing happens in reverse. Both comparisons are mostly measuring the fact that your asset mix is not the benchmark's asset mix.
The fix is arithmetic rather than cleverness. Work out your average allocation over the period, not today's allocation, and blend the two lines in those proportions. If you ran roughly sixty percent crypto and forty percent equities, your benchmark is sixty percent of the crypto line plus forty percent of the equity line. Say crypto returned thirty percent over the year and equities returned ten. Your blend is 0.6 times 30 plus 0.4 times 10, which is twenty two percent. That is the number your curve has to beat, and it is a very different bar from either thirty or ten.
Two details make the blend honest. State the rebalancing rule, because a blend that resets to sixty forty monthly is a different line from one that drifts all year, and in a volatile mix the difference is not small. And use your average weight rather than your ending weight. If crypto ran from twenty percent of the book to sixty percent because it went up, benchmarking at sixty is grading yourself against the outcome rather than against the decision.
The period button decides the winner more often than the line does
Look at the period controls in the screenshot. They run 1D, 7D, 14D, 1M, 3M, 1Y, ALL and CUSTOM, and the capture is sitting on 1Y. Every one of those produces a different race, and against a volatile benchmark the ordering flips regularly.
This is where benchmark comparisons quietly become self flattery. You look at 1Y, you do not like it, so you look at 3M, and the 3M window happens to start just after the drawdown that hurt you. Nothing dishonest happened at any individual step, and the conclusion is still worthless.
Two rules keep it clean. Pick the window before you look, and use the same window every time you check. And whatever window you use, also look at ALL, because the full history is the only one you cannot have selected for. If the story only exists at 3M and disappears at ALL, the story was the window.
A balance line and a return line are not the same thing
There is one more distinction to settle before racing anything, and the capture shows why. The Account Equity panel calls itself the real cumulative balance summed across your connected accounts, and it reports a net figure of positive 396.07% for the selected window. The tiles directly above it show a win rate of 25.29%, a profit factor of 0.08 and expectancy of negative 1.20% across 87 trades.
Those readings pull in different directions, and both can be perfectly correct. Figures on the same page legitimately differ for a handful of ordinary reasons. They can cover different measurement windows. They can be computed on different underlying series, and a balance series and a list of closed trades are genuinely different objects, since a balance also moves when you deposit or withdraw and a trade list only moves when a position closes. One may be annualised and the other stated per trade. And the sample sizes are nothing alike, because a daily balance series has hundreds of points while the trade list here has 87.
I am not going to tell you which of those explains this particular pair, because working that out requires knowing exactly how each figure is built and I do not. There is also no reason to assume anything is being calculated wrongly. What you can do is find out for your own account, by changing one control at a time and watching which numbers move. Switch the period from 1Y to ALL and see which tiles respond. Switch the account selector between the cumulative view and a single named account and watch again. Whichever line responds to your deposits is the balance line, and that is the one you must not race against an index, because a chunk of its rise was your own money walking in the door rather than anything you traded.