Comparing Two Coins on One Chart

You want to know which of two coins has done better. It sounds like the simplest question you could ask a chart, and it is the question charts are worst at answering honestly.

Put both prices on the same axes and you get a picture that looks like a comparison and functions as an optical illusion. One line sits near the top and barely wobbles. The other crawls along the bottom, apparently flat, actually up 400%. Nothing about that image is wrong. It is just not answering the question you asked.

Why can't you put two prices on the same chart?

Because a token's unit price carries no information about the size of anything.

Price is total value divided by the number of units, and the number of units was chosen by whoever launched the token. A project that minted a hundred billion units trades at fractions of a cent. A project that minted two million trades in the hundreds of dollars. Those two facts tell you about supply decisions made before the market existed. They tell you nothing comparative.

That means a chart with a $0.15 coin and a $3,400 coin on shared axes has already lost. Either the small one is a flat line at the bottom, or you split the axes and the visual relationship becomes whatever the scaling makes it. Dual-axis charts are the standard way to make a comparison say what you want, and once you know to look for them you find them everywhere in screenshots.

Market cap fixes part of this and introduces its own issue. It multiplies price by circulating supply, so it does describe size rather than unit denomination. But circulating supply excludes everything still locked, which means a low-float token appears far smaller than the market's implied valuation of it. Market cap vs. fully diluted valuation covers why the two figures can differ by an order of magnitude and which one belongs in a comparison.

What does indexing to a common start point do?

It converts both series into the same unit: percentage change from a shared moment.

Pick a start date. Set both assets to 100 on that date. Plot each subsequent price as a percentage of its own starting price. Now a line at 180 means up 80% since the start, for either asset, regardless of what a unit costs. Both lines answer one question, and it is the question you were asking: what would a holding in this have done over this window.

This is not a crypto technique. It is how equity indices, currency pairs and commodity series have been compared for as long as those charts have existed, and the reason is the same everywhere. Absolute levels are incomparable across instruments. Percentage change is comparable across everything. It is also the only defensible way to put crypto next to an index or a metal, which is why the same discipline governs reading crypto against gold and the S&P.

Two practical points follow. Indexed comparisons are sensitive to the scale you plot them on, and a linear axis exaggerates recent moves in anything that has multiplied several times over. A logarithmic axis gives equal vertical space to equal percentage moves, which is what you want when one asset is up 40% and the other is up 900% on the same picture.

The second point is that percentage change is not risk-adjusted. An asset up 60% with three drawdowns of half its value along the way is not the same experience as an asset up 55% in a straight line, and the indexed chart draws them as near-neighbours at the right-hand edge. What happened between the endpoints is most of what you would have actually lived through, which is the same reason unrealised vs. realised profit matters more than a single endpoint number.

The Crypto App does not draw two coins on one set of axes. What it gives you instead is a watchlist reading down one screen with the same percentage window applied to every row, and a coin-to-coin converter that returns the current exchange rate between any two assets. That covers the practical version of both comparisons on a phone, and it is worth being straight about which is which rather than implying an overlay that is not there.

What is a ratio chart actually showing you?

Divide one asset's price by the other's and plot the result. ETH/BTC, or any pair you like. That single line is the relative performance of the first against the second, with the market-wide move removed entirely.

The removal is the point. In a week when everything falls 20%, the ratio shows whether one fell less than the other. In a week when everything rallies, it shows which rallied harder. A rising ETH/BTC line means ETH gained ground on BTC, and it means that identically whether both went up, both went down, or they diverged.

What a ratio chart deliberately will not tell you is anything about value. ETH/BTC can rise through a period in which both assets lost half their dollar value, and the chart looks like an uptrend the whole way. People misread this constantly. Ratio charts answer relative questions only, and they are silent by construction on absolute ones.

They are also noisier than they look. A ratio has two moving denominators of its own, and a sharp move in either asset produces a spike in the line that has nothing to do with a durable change in the relationship. That is a version of the same timeframe problem covered in chart timeframes: the shorter the window, the more of what you are seeing is mechanical.

Pricing a token in Bitcoin rather than dollars is the same idea with a different frame, and it changes what a good year looks like. An asset up 50% in dollars during a period when Bitcoin doubled has lost ground in Bitcoin terms. Both statements are true. They are answers to different questions, and neither is the honest one on its own.

How do you pick a start date that isn't cherry-picked?

By choosing it before you know what it will show.

Almost every misleading comparison chart in circulation is misleading through its start date alone. Begin the series at a local bottom for the asset being promoted and its line launches upward from bar one. Begin it at a local top for the asset being criticised and the picture reverses. No data is falsified. The selection did all the work.

There are start dates that are defensible because they mean something independent of the outcome. The beginning of a calendar year or quarter. The date a network went live or a major upgrade shipped. The date you actually bought. A fixed lookback like one year or three years from today, applied to both assets without adjustment. Each of those was chosen for a reason that exists outside the comparison.

Then test it. Move the start date a month earlier and a month later and see whether the conclusion survives. A relationship that holds across several reasonable starting points is describing something. One that inverts when you shift the start by three weeks was an artefact of the window, and now you know.

The stronger version of the test is to run several windows deliberately and look at the disagreement. Year to date, one year, and since each asset's most recent all-time high will frequently rank the pair differently, and the spread between those answers is itself the finding. A comparison that only works from one starting point is a chart about that starting point.

Set the two coins side by side on a watchlist with the same window applied to both, revisit it in a month, and you will have a comparison nobody constructed after the fact — including you. Watchlist alerts will tell you when one of them moves in the meantime, which saves opening the chart on days nothing happened.