Comparing Investment Options - Charts
Investment decisions are hard. Broadly, you have two choices: let someone else manage your money, or manage it yourself. If someone else does, you should at least benchmark their returns — we wrote about that here. If you do it yourself, you need to know how to compare your options before you can decide between them.
Comparing investment options is a broad topic, but an important part of it is comparing past performance. We will look at how charts help in comparing returns. Past performance is no guarantee, but it is one indicator of what might lie ahead.
So, how do you actually compare options? Judging by what most sites show, the usual answers are CAGR (over set periods like 1y, 3y, 5y) and the NAV chart. Both summarize past performance, and both are useful — but they're biased summaries, not the only way to look. The same performance reads differently from different angles. We'll walk through a few of those angles here, and build mental models so you're not lost in front of any benchmark.
Underneath the Metrics
Fundamentally, every investment option is just price changing with time. Some have price for each day, like NAV for mutual fund. Others could have their price changing by the second, or even faster (whenever a seller and a buyer agree on a price), like ETFs, stock prices and other traded assets. But FinBodhi only tracks daily prices for all assets. E.g. for stocks, we pick end of day price. So, mostly it comes down to: since the asset started trading, on each date, a unit of the asset has some price.
That is the raw material behind every comparison. Visually, it is just a price chart, with time on x-axis (horizontally) and price of a unit, on the y-axis (vertically).
Most of the performance analysis are based on the price history. We will call this the NAV Chart (price of an asset changing over time).
Returns
A NAV Chart on its own is just what one unit costs. It becomes a return only when you overlay a decision on top of it: a start date when you put money in, and an end date when you take it out. The return is how much you got back: price on end date - price on start date.
But what you really care about is the return multiple (equivalently, return percentage). In other words, how much did you get for a rupee invested, for any given time period. Let's say over two years:
- Fund A goes from ₹100 to ₹121
- Fund B from ₹3,200 to ₹3,872 and
- Gold from ₹5,800/g to ₹7,018/g
The absolute moves (₹21, ₹672, ₹1,218) look nothing alike, but for the buyer all those options are similar — for every ₹1 you put in, you get back ₹1.21, in every case. Whatever money you had put in any of the funds at the starting of the 2 years, would grow 1.21× by the end of the 2 years.
Putting every option on this same footing (what one rupee becomes) is called normalizing. We rebase each fund so it starts at the same value (₹1, or sometimes 100), and from then on the curves track the multiple on that rupee rather than the raw price. A ₹100 fund and a ₹3,200 fund, impossible to eyeball side by side, become directly comparable once normalized. This is the trick behind every comparison chart later in the article.
So a return for an investment option (a NAV Chart) is a consequence of: a start date, an end date. Now we have covered 3 things: NAV Chart, start date and end date. But you also have to consider how long the investment period was (end date - start date). Why? You can't compare a 5 year return multiple with a 1 year return multiple. That's where CAGR comes in.
CAGR
CAGR calculates a yearly return multiple, for any period. E.g. A return multiple of 1.61 for 5 years, means CAGR yearly multiple of 1.1. To calculate this, we need the duration, which can be derived from start and end date. The CAGRs that you come across in wild, pick them for you. The convention is to take today (or the report date) as the end date, step back , , , or years for the start, and quote what each of those windows produced.
Below is a sample five-year price history. With the end date pinned at the right edge, each window length carves a different CAGR out of the same data.
Try a different window: · · ·
If you bought 2 years back and sold now, the simple return is:
CAGR annualizes this return multiple: it finds the constant yearly rate that would have taken you from entry to exit over those years.
Our sample chart goes ₹100 → ₹161 over — a 61% total return, or a CAGR of 10%/yr (since 1.10⁵ ≈ 1.61×). CAGR, (which represents annual growth) helps you compare performance of a fund across different periods. E.g. how has the fund done in last vs last .
But even then, comparing CAGR with a CAGR is missing a lot of context. Why? Because to compare funds you have to compare how they did in similar circumstances. There might have been a slump back, which meant most investments made at the time would show great returns now — buying in a slump means a low entry price, so the same recovery works out to a bigger multiple. When you invest and when you take money out is important. Not just the duration of your investment. It's reasons like these, that I prefer charts.
Two Dots
To understand how we can analyze returns with charting we have to start again from NAV Chart. Below, it traces a fund from 2020 to 2025. The price history is the wiggly line, and return is calculated from two dots picked from it.
The highlighted window is ₹110 in 2021 to ₹133 in 2023.
A 1.21× (21.0%) return multiple over 2 years. Two dots, one number. The full chart has all the wiggles in between, but the CAGR calculation keeps only the two ends. This is an interactive chart, move the gray dots around to calculate the return and CAGR.
Charting Returns
Now remember we want to understand returns not just for one fund, but across funds. So, how do we chart this? If we let both start date and end date change, we need a chart which can show two variables and one result. Charting that needs three axes (start, end, return) or a colour-coded heatmap — interesting, but hard to compare across funds (see the appendix). So instead we restrict what changes: chart return against one variable, which lets us stay in 2d.
So, the question becomes, what do we fix. A convenient choice is start date. This way, only end date changes and we can chart return against end date. Let's take it step by step. Here is our trusty old NAV Chart.
Here is an arbitrary start and end date in the same chart, showing return multiple within the chart. The start date is fixed. Move the end date to see how return changes. We will use as an indicator of whats fixed.
Now instead of showing it within the chart, we can show the return in a separate chart, with y-axis as return and x-axis as the end date.
And that brings us to last step. Draw a line for return, as end date changes.
To play around with this, drag the two points on the NAV Chart. You can also drag the fixed point (). Changing the fixed point, changes the whole Return Chart. In other words, changing the start date might change how you look at return. Below are a few more observations. You can click on them to show the scenario on the chart above.
- Exit timing: Here are two exits with the same entry date sells into a peak; sells into the dip right after it. In other words, selling at peak gives you better returns.
- Entry timing: Now consider same exit but with different entry dates. buys near the early trough; buys at the peak a little later. Notice how the whole result chart moves up if you buy at dip.
- Slope: Notice how the slope of the line corresponds to return multiple. Higher slope means bigger returns(). Negative slope means you lose money ().
I find it useful to think in terms of what question does the chart answer. In this case, the question is If I invest on a certain date, how would my investment grow with time?. I call this the Leading Return Chart. The left, leading, edge of the window is what's pinned.
For the rest of this article we will continue drawing the NAV chart and the normalized chart as a stacked pair: NAV charts on the top, and the return chart comparing the NAVs below. Same data, two views. But remember the important one is Return Chart. The NAV Chart is just for building your understanding.
Compare Funds
Now why did we spend effort into building Return Chart? So we can compare multiple funds. Here we see NAV Charts of two funds and their returns in a single Return Chart.
Again, you can drag both the fixed start date, and end date to play around with the chart. Changing start date changes the return chart.
The Leading Return Chart is not the end of the story. The pinned left edge is often arbitrary: when you opened the factsheet, when a blog post was written, when a reporting period began. Anything that depends entirely on one chosen slice of history is one Tuesday away from a different headline. Investing in the covid downturn would have made almost any fund's return look good. That leads to the other two chart types.
Three Views to NAV Chart
We saw how fixing start date gives us a Return chart. There are two other ways to get Return Charts. One is to fix the end date, where y-axis would be start date. And the last one is to fix the duration (end date - start date). Let's say 2yr. The x-axis would represent 2yr period, with (by convention) the ending date as x-axis.
To summarize, there are three things you can hold constant, so that the x-axis can represent time, y-axis the return.
- Start time: left edge of the chosen window is fixed, so the x-axis is the end time. Leading Return Chart.
- End time: right edge of the chosen window is fixed, so the x-axis is the start time. Trailing Return Chart.
- Duration: here what's held constant is a length of time, not a point — both edges move together, keeping the same gap, and the x-axis is the end time of each window. Rolling Return Chart.
We already saw the Leading chart. Now, let's look at Trailing, where the end date is fixed.
What does the trailing chart answer? If I sell now, what would be my return for money invested anytime in the past?. It's an interesting question that I don't see asked almost anywhere. This is the curve form of the familiar "1y / 3y / 5y / 10y" numbers on a factsheet. Those headline figures are just a few selected points from the much larger trailing curve. It's a difficult chart to parse. The line has -ve slope in general. Given the current value, it helps you understand how much timing the entry would have mattered. Higher values in the past means, investments done at those times would have done better now.
Now, the last one. Rolling Return Chart. This answers the question If I invest for 2yrs, what would be my return over the history.
Leading pins the left edge. Trailing pins the right edge. Rolling removes that special status entirely. Instead, keep the window width fixed, say 2 years, and slide the whole window through history.
Rolling window: · ·
Each point here is the return multiple of one 2-year block. So, it asks: across history, what kind of 2-year outcomes has this asset produced?
That is why rolling is often the better view for comparing long-holding behavior. Leading and trailing each show one path through history. Rolling shows a distribution of paths of a fixed length.
In FinBodhi, the leading and trailing views share one chart with a draggable pivot. The pivot is the date being held fixed, and dragging it slides smoothly between a pure leading view (pivot at the left edge) and a pure trailing view (pivot at the right edge). The rolling chart is a separate view, with a window-length slider from 1 year up to 10 years (capped at whatever the underlying data supports).
What the Charts Let You See
A few patterns are common enough to be worth naming. Once you've seen them, you'll spot them everywhere.
Recovery dressed up as performance. A fund that bottomed out in March 2020 and ran up afterward will look spectacular on a leading chart pinned to March 2020. Move the left edge of the window back to January 2020 and you first see the crash, then the same recovery. The headline number is real, but it is mostly a story about where the window began.
A 5-year CAGR that depends on one Tuesday. Trailing returns are sensitive to where the right edge of the window sits. A "5-year CAGR of 14%" today might be 11% in a month if the market dips, or 17% if it rallies. The trailing curve makes that instability visible.
Smooth leading, jagged rolling. Two funds can finish at the same multiple on the leading chart and look completely different on rolling. One may have delivered a tight band of 5-year outcomes; the other may have swung from excellent to mediocre depending on the window. Rolling surfaces what the smoothed leading curve hides.
Sitting above or below a baseline. Add a constant line at your hurdle rate: FD, inflation, 8%, whatever your alternative is. On the rolling chart, that baseline is flat, so you can quickly see how often an asset stayed above it.
Short history flattering a curve. A 3-year rolling chart can look far better than a 5-year rolling chart simply because the difficult years fall outside the shorter window.
A few more come straight from the shape of each chart.
Volatility is the height of the rolling band. The taller the band, the more your return depended on when you happened to enter. A flat band means timing barely mattered; a tall one means it mattered a lot, even if the average looks fine. And how low the band dips tells you whether any N-year hold ever ended underwater.
Compounding fans out on the leading chart. Two curves sitting a hair apart early on spread into a wide gap by the right edge, because a small edge in slope compounds year on year. That widening gap between two leading curves is the visible price of a slightly lower CAGR over a long hold — the reason "only 2% behind" is rarely as small as it sounds.
A crash shows up as a cliff in the trailing curve — and the two funds rarely cliff the same way. A crash interrupts the curve with a sudden step: enter just before it and your return-to-today drops, enter just after and it jumps. Overlay two funds and the cliffs line up at the same dates, so you can read which one fell harder and which one climbed back faster off the same crash. A deeper cliff or a slower recovery is the kind of difference a single CAGR hides.
In FinBodhi
The Compare page bundles the three cases into two charts:
- A Leading/Trailing chart with a draggable pivot. Pivot at the left edge gives the leading view; at the right edge, the trailing view; anywhere in between, you see some of each.
- A Rolling chart with a window-length slider that runs from 1 year up to 10 (or whatever the underlying data supports, whichever is shorter). In the app, rolling windows are shown as CAGR instead of raw return multiple.
Both accept any mix of curves — mutual funds, indices, metals, stocks, and constant baselines — normalized to a common reference so they're comparable on a single axis. The chrome around the charts (colour palettes, date range, zoom and pan, curve hide/show, editable labels, hover sync between the two charts) is covered in the Compare page reference in the docs.
What Else Lives in the Price History
The Compare charts in FinBodhi answer one specific question: what return do I see when I hold the left edge, the right edge, or the window width constant? They do not answer everything. The same price history can also be used for:
- Volatility. The exact standard deviation of daily or monthly returns — the day-to-day jaggedness of the path — isn't isolated by these charts. (The rolling band's height shows a related but distinct thing: how much the N-year outcome swung with timing, not how bumpy the ride was getting there.)
- Drawdowns. The deepest peak-to-trough fall, and how long the asset took to recover. A leading curve at 1.6× could have dipped to 0.7× and climbed back; the curve shows the path, but doesn't isolate the worst stretch.
- Beta and alpha. Regress a fund's returns against a benchmark. Beta is how much it moves with the benchmark; alpha is what's left over after accounting for that.
- Risk-adjusted returns. Sharpe, Sortino, information ratio — return divided by some measure of risk.
- Correlation. Whether two assets move together or offset each other. Important for diversification, doesn't show up on a side-by-side multiple comparison.
- Money-weighted returns. What you actually earned, given when and how much you invested. This needs your cashflows on top of the price history. FinBodhi has a separate Benchmark page with NAV and Value charts for that.
- Costs and taxes. NAVs are post-expense, but exit loads, capital gains tax, and tracking error against an actual ETF aren't in the curve. The chart compares the theoretical price series, not what reaches your account.
Compare is intentionally narrow. We had to start somewhere. In future we might add support for the other metrics.
In Summary
A return always has a left edge, a right edge, and a window width. In shorthand, those are start, end, and duration. Pin the left edge and you get a leading chart. Pin the right edge and you get a trailing chart. Keep the window width fixed and you get a rolling chart. Leading and trailing each show one path through history; rolling shows a distribution of paths of a fixed length.
A single "5-year CAGR" on a factsheet is just one point on one trailing curve. Seeing the same asset through leading, trailing and rolling views gives a fuller picture, and often a less flattering one. The point of these charts is not to replace the headline number, but to show what the headline number is hiding.
Appendix: The Two-Variable View
In Charting Returns we noted that letting both the start and end date vary needs more than a 2d line — you have to show two inputs (start, end) and one output (return) at once. There are two honest ways to draw that:
- A 3d surface: start date and end date on the floor (x and y axes), return as height (z axis).
- A heatmap: start date and end date on the two axes, return encoded as colour.
Here is the heatmap form for a single fund. The NAV history is synthetic, which is why the bands look so regular — ignore the regularity and focus on the structure.
Every cell is one (start, end) pair, coloured by the return multiple for holding between those two dates. The leading, trailing and rolling charts in the main article are each just a slice through this surface. Leading fixes the start date, trailing fixes the end date, and rolling walks a fixed-width diagonal.
It's a comprehensive picture, but a hard one: the eye can't compare two of these side by side. That's exactly why the article restricts to one variable at a time and charts the result as a line.