Trader vs Researcher Is the Wrong Question: The Five Axes That Actually Decide Your Quant Career

Hey, this is Stanley. I’m a quant researcher, and my first job out of school looked like a clean win on paper. The pay was high. Everybody knew the firm’s name, and it had a reputation as one of the best places in the market to make money. The whole package you get told to go chase, basically.

The part I’d missed didn’t surface until I was on the desk. The seat ran on domain knowledge. You had to know one industry cold, every bit of the institutional plumbing, the way it really cleared and settled. I’m just not wired that way. I’m a technical person, and what I’m good at is the modeling and the engineering. Running a book there, making PM, meant five to ten years of grinding myself into a domain expert, and probably longer if I’m honest with myself. The job also kept asking me to react to fundamental news on the fly, to form a view and trade it by feel. My brain doesn’t do its best work like that.

So I left. I moved to a different firm to land a seat where the thing I’m actually good at was the thing that got paid, and I sat out a non-compete to do it. That’s when the rest of the lesson hit me. Switching isn’t free. A lot of what I’d built didn’t come with me, because the years of skill I’d put together were welded to a domain I was now walking away from.

None of this had occurred to me when I signed the first offer. The title on it said “quant researcher.” That phrase told me almost nothing about the actual job. (If you’re earlier than that and still mapping the landscape, the free lessons Welcome to Quant Finance and The Quant Industry Landscape are a decent primer on the players before you read the rest of this.)

Trader, researcher, developer: the label isn’t the job

Every junior burns a month agonizing over trader versus researcher versus developer. Put it down. Those three words are real categories, but they barely predict what you’ll actually do, because they mean something different at every shop.

A “quant trader” might write code most of the week. A “quant researcher” inside a pod can be a cog feeding someone else’s book. A “quant developer” at one firm builds the alpha; at another, they only plumb it. You can be “in HFT” and never touch latency. Two firms down the street from each other hand the same title to people whose Tuesdays look nothing alike. And whether you can switch lanes later isn’t fixed: at a place like Jane Street they keep the lines blurry on purpose, at a siloed pod you can sit in one for years. The firm decides that, not the title.

So stop sorting seats by what’s printed on the business card. What actually defines the job is five things.

What actually defines a seat: five dimensions

Whatever your title, your real day comes down to where you land on these five:

1. Asset class. Equities, rates, FX, credit, commodities, with derivatives layered on any of them.
2. Frequency. Microseconds. Or minutes and days. Or months. Your holding period isn’t a matter of taste; it falls straight out of how your money gets made.
3. Alpha source. Where the returns actually come from. Being faster. A statistical signal. Pricing options. Reading a market or a domain. Building the engine. Executing someone else’s flow.
4. Pipeline stage. The line that turns data into positions runs data, then signal, then blending, then portfolio construction, then execution.
5. Org type. A flat prop shop. A multi-strat pod. A systematic fund. A bank desk. A commodity merchant. A pod is really an economics wrapper (your own sub-book, a tight drawdown limit, a cut of what it makes), so it sets how you get paid, not what you do all day.

Two seats can sit very differently on that pipeline. Some span the whole stack: at a small shop, one person takes raw data all the way to live positions. Others are one stage of it: at a giant platform you might own nothing but the signal step, or only execution. Neither is better, they’re just different days. And nothing forces these to be clean boxes; the same person can cover several asset classes, two frequencies, or two alpha sources at once.

To make it concrete, picture three people, all called quants. One quotes futures all day at a prop shop: stay fast, stay flat, don’t get run over. One hunts signals across thousands of stocks at a systematic fund: find something real, prove it isn’t just noise, watch it decay. One prices exotic options for clients on a bank desk: build the model, mark the risk, never carry a bet of their own. Same word on the card, three jobs with almost nothing in common.

Role, by the way, runs across all five and is the cheapest axis to change, which is exactly why fixating on it is a trap. People spend their whole decision budget on the one thing they can move and shrug off the four they mostly can’t.

From five dimensions to six families

Five dimensions is a lot to juggle, and laid flat the seats don’t sort into anything clean. But one of the five does most of the work: where the money actually comes from, the alpha source. It drives both your day and how far you can move later, because it’s the skill you spend years building. The other four are dials you set around it.

So sort every seat by its alpha source and the mess resolves into clumps. Inside a clump you can slide along the other four axes, change the asset, the frequency, the firm, because the skill carries. Across clumps you mostly can’t, since there’s no shared skill to bring. That is your mobility in one sentence: how many seats share your alpha source, or a piece of it that transfers. Do this across the whole industry and you land on roughly six clumps. Those are the families, and each comes with a very different set of pros and cons.

The six families

I built an interactive version of this; it’s right below the list. They run roughly from “alpha lives in your head” down to “alpha lives in the firm’s hardware.”

The map

The six families, on a portability spectrum

From “alpha lives in your head” (most portable) to “alpha lives in the firm’s hardware” (least). Tap a family to see example seats and the firms that hire for it. Lower portability is a tradeoff, not a verdict: it usually means scarcer, more durable, and better paid.

more portable

more firm-owned

What sounds like you: Good at math & stats Physics / hard-science modeler Good at coding & building Into machine learning & AI I follow markets & news Love puzzles, poker & games Fast, competitive, want a scoreboard Patient, deep researcher Careful & risk-aware Good with people & business Want to run my own book Show all

● a seat, colored by family ─ easy move (shared skill) ──→ harder move (on tap) tap a seat to light up where it can go

Tap any seat to see its day-to-day, the moves that open up from it, and the firms that hire for it.

• 1 Statistical-signal research a signal you found Most doors
• 2 Volatility and options pricing pricing convex risk Decent
• 3 Macro and domain reads a curve, an issuer, a domain High within, low across
• 4 Engineering and infrastructure build the engine Good
• 5 Microstructure and speed colocation, queue, execution Locked-in
• 6 Agency and execution the firm’s mandate and flow Capped

So which one is yours

This is the part most people get no guidance on at all. Don’t optimize for keeping every door open. Optimize for fit. No family wins in the abstract. It depends entirely on who you are. Four honest questions narrow it down.

What energizes you, and at what feedback speed? If a fast reflexive game recharges you and you want a scoreboard at the end of every day, look at speed or options-MM. If you can sit with a problem for six months with no readout and still care, look at signal research or factor. The test I keep coming back to is this. Imagine half a year goes by and you get nothing back. Does the problem still tug at you, or do you quietly check out?

Your real strength is the next question, and it’s the one I personally got wrong, so don’t flinch on it. Technical strength points you toward signal research, vol pricing, engineering, or speed. A strength that’s a read on a market, an industry, a situation points you toward macro, domain, or event-driven. Teaching a math person finance is easier than teaching a finance person math, so lean into the strength that’s hardest to acquire later. Pick the family whose scarce input is the thing you’re already good at. Ignore whichever one has the shiniest name on the door.

How much variance can you actually hold? Be honest. There’s pod intensity, where a 5% drawdown can halve your book and a touch more can end the seat. There’s prop stability. There’s bank security. If a market you had no hand in could end your job, and that prospect would wreck you, the pod isn’t your family yet.

And what are you really maximizing? Money now, learning, optionality, autonomy, depth, lifestyle. They pull against each other. Optionality is just one item on that list, and it should only break a tie when fit is genuinely too close to call. If you already know a niche fits you, take the niche. Seniority is going to specialize you regardless, so you might as well pick the specialty rather than drift into one.

One caveat, and it’s a big one

Everything above is the seat-level map, and it matters. It doesn’t decide your experience by itself, though. Here’s roughly how it shakes out in practice. About half of how good the job feels day to day traces back to your direct manager and your team, whether you’ve landed somewhere that teaches and shares or somewhere that leaves you to sink. The firm and its brand, the platform, the door it opens next, that’s maybe another quarter. What’s left is the specific seat and its asset-class beta, the part this whole map describes.

So the map can tell you which kind of seat suits your strength and where it might lead. What it can’t tell you is whether one particular desk is a good place to spend three years of your life. For that you have to talk to the people on the team, get a read on the manager, and intern first if there’s any conceivable way to pull it off. A summer in the seat teaches you more about the actual day-to-day than any framework ever could.

What I’d tell my younger self

I can’t claw back the time I spent in a seat I’d have read very differently with this in hand. Nothing was wrong with the domain itself. It just wasn’t right for me, and the reason had been sitting out in the open the whole time. The seat rewarded a strength I didn’t have, and the strength I did have had nowhere to go inside it.

So I drew up the version of this I wish someone had handed me on day one, which is the map above. Drop yourself onto it and trace where your seat actually leads. If there’s one thing to take from all of it, it’s the order to think in. The dimension you can’t undo later comes first. Whether the work fits your strength comes somewhere right after. The title is way down near the bottom, which is a little embarrassing to admit, because back then it was about the only thing I looked at.

Common questions

Is it better to be a quant trader or a quant researcher?
The title is a weak predictor of the actual job. A “quant trader” can spend most of the week writing code, and a “quant researcher” in a pod can be a cog feeding someone else’s book. Sort seats by the work, not the title: where they sit across the five dimensions above.

Which quant career path has the most mobility?
Statistical-signal research. The signal you build is yours and crosses assets, and the toolkit is generic, so almost every other research seat shares an intersection with it. The tradeoff is the thinnest moat: it’s the most crowded corner, and the alpha decays as others run the same ideas.

How should I choose between quant career paths?
Optimize for fit, not for keeping doors open. Ask what energizes you and at what feedback speed, what your real strength is (technical modeling versus reading a market or domain), how much variance you can hold, and what you’re actually maximizing. And remember that roughly half of how good a job feels comes from your direct manager and team, not the seat.