Day trading vs swing trading: what you can test, and what you can only read about
Day trading and swing trading — or, however you search it, swing trading versus day trading — are the two styles most new traders are told to choose between, usually with a confident story about which one is “better” — for beginners, for busy people, for the risk-averse. This page does not sell you that story. It does something narrower and more useful: it separates the two by what the evidence can actually show, because one of them you can put to an honest test on real data before you trust it, and the other you can only read about. The honest value lives in that act of testing the testable side — not in being told which style to pick.
What each one actually is
Day trading means opening and closing a position inside a single session — flat by the close, nothing held overnight. Holds run from seconds to hours, the decisions react to intraday price and volume, and it is frequently done on margin or leverage; in Europe much retail day trading happens through CFDs, a leveraged derivative. Scalping — dozens or hundreds of tiny trades a day — is day trading taken to its fastest extreme.
Swing trading means holding a position for roughly a few days to a few weeks, carrying it overnight and over weekends — which brings gap risk, the price jumping while the market is shut. It acts on daily or multi-day signals rather than the tick-by-tick tape, and it places far fewer trades, so each one pays spread and commission far less often, though each is exposed to those gaps. Hold the same idea for months to years and it stops being swing trading and becomes position trading — longer again; some of the slower rules in our own census sit on that position end.
The honest split: horizon and testability, not merit
The honest difference between them here is not which one pays — it is which one can be put to a test. Our engine runs on daily bars, so it can mechanically re-run the kind of closed-bar rules that much of mechanical swing trading is built from, on real price history, with costs charged and no hindsight. It cannot touch anything intraday, so day trading is simply out of its reach. And the by-eye half of BOTH styles — the hand-drawn levels, the patterns read in the moment — is beyond any backtest, ours or anyone's. So one side of this comparison can be checked before it is trusted, and the other can only be read about. That testable asymmetry, not a winner, is the real difference — a fact about horizons and data, not a verdict that the testable style is the better or safer one.
The side we can put to a test
Here is the one place our own data can speak, and only to the swing-horizon side. The famous mechanical, long-only, daily-bar rules we test hold for a few days at the fast, swing end out to many months at the slow end — the 200-day-MA regime, time-series momentum, the 50/200 cross and Heikin-Ashi all hold for months — and the census records no holding period, so this is not a measurement of swing traders and we did not record a holding period for any rule. On the famous US stocks in our frozen, open census, 721 of 839 judged backtests (86%) lagged simply holding the same asset, net of costs. A real minority still beat it, and finding one of those is worth everything — but two cautions ride with that minority: a rule can lag buy & hold and still have made money, and a beat on a survivor-only, long-only set of famous names is flattered by hindsight and by how many variants one could have tried. It is a measured and bounded result — not a verdict on swing trading as a style.
From our frozen, open census — as of September 25, 2026 · the frozen release · DOI 10.5281/zenodo.22974116. See the whole glass on the honest scoreboard.
The side we can only read about
On the other side, the day-trading research — external evidence, not our data, because our daily-bar engine cannot test intraday at all — points a sobering way: in the largest studies of real brokerage accounts, most day traders finished behind, and fewer than one in a hundred were reliably ahead of fees year after year, which is a different and smaller claim than the roughly one in five who finish any single year ahead. A skilled few genuinely clear the bar — which is exactly why the question deserves an honest look rather than a folklore number. These two figures do not compare, as the next section explains. The full day-trading evidence, named and dated →
These two numbers do not compare
A warning before you set those two pictures side by side: they do not compare. One counts real human day traders who finished ahead of fees; the other counts mechanical rules that beat simply holding the same asset. Different people, a different yardstick, a different horizon — and a rule can have made money and still lagged buy & hold, just as a losing rule can still have beaten it. There is no ratio to take between the two, and anyone who quotes you one has invented it.
So — which one? The honest read
So — day trading or swing trading? The honest read is that the question has no single answer we can hand you, and the two cannot be scored against each other here. What we can say is a matter of scope, not preference: the mechanical, daily-bar side of swing trading can be checked on real history before a cent is risked, and most of the rules we checked that way still lagged simply holding — while the intraday side cannot be checked by us at all, and the external studies of it are sobering, with a skilled few clearing the bar. Neither of those is a nudge toward one style or away from the other. It is a cost and a probability on each side, laid out so you can weigh them yourself.
One side of this comparison you can actually test, right now, for free: the daily-bar, mechanical rules a swing-horizon trader leans on — re-run on famous stocks and crypto, costs charged on every trade, no hindsight, measured against simply holding. The intraday side we cannot test for you, and we will not pretend to.
Limits
Two different kinds of evidence sit on this page and should not be blended. The day-trading figures are external research — specific markets and periods (Taiwan equities, Brazilian index futures, US household accounts, European CFDs), each an honest but partial window, none the last word — and our daily-bar engine cannot test intraday at all. The one first-party figure is survivor-only (famous US names that lived to be famous), long-only, a run of mechanical rules whose horizons span the swing range out to months rather than a cohort of swing traders, one roughly five-year window, no out-of-sample split, and not significance-tested — a bounded result, held in its lane.
Sources
- Our own data (the one first-party figure): The Honest Backtest Census (frozen release, September 25, 2026; DOI 10.5281/zenodo.22974116) — daily-bar, long-only mechanical rules on famous US stocks; the swing-horizon side only.
- The day-trading figures are external research, named and dated in full on Can you make money day trading? — the largest real-account studies (Barber, Lee, Liu & Odean, Taiwan; Chague, De-Losso & Giovannetti, Brazil; Barber & Odean on US household turnover) and the European regulator ESMA on leveraged CFDs. Summarised and linked here, never re-stated in place of the source.
This page compares two trading styles by what the evidence can show and is educational, not advice, and not a recommendation to trade, to avoid trading, or to prefer one style over the other. Past performance is not a reliable indicator of future results. This is an educational, analytical tool — not investment advice, and not a recommendation to buy or sell anything. You make all decisions and execute on your own broker.