Skip to main content

The Swing Trader's Edge (Overnight Drift, Institutional Accumulation)

Updated Jun 23, 2026 at 8:47pm

Research Draft Medium 818 words

A swing trader holds positions for days to weeks, which only makes sense if multi-day holds capture return that a same-day round-trip would miss. Two phenomena are often cited as the structural rationale for this. The first is overnight drift — the well-documented finding that, historically, a large share of equity returns has accrued while the market is closed rather than during the trading session, so being positioned overnight (which day traders by definition are not) has paid off. The second is institutional accumulation — the reality that large players cannot enter or exit a position in one print, so they build it over days or weeks, leaving a multi-day footprint in price and volume that a swing trader aims to ride. Both are genuine, peer-reviewed observations. Neither is a guaranteed, mechanical edge, and the rest of this note is as much about the caveats as the rationale.

Overnight drift

The core empirical fact: across long historical samples, U.S. equity index returns have accrued disproportionately overnight (close-to-open) rather than intraday (open-to-close). Cooper, Cliff & Gulen ("Return Differences between Trading and Non-Trading Hours: Like Night and Day," SSRN/2008) documented that the equity premium over their sample was essentially earned during non-trading hours, with intraday returns near zero or negative, and showed the pattern holds for individual stocks, indices, and index futures across exchanges. Lou, Polk & Skouras ("A Tug of War: Overnight Versus Intraday Expected Returns," Journal of Financial Economics, 2019) decomposed many factor strategies into overnight and intraday components and found returns to different clienteles systematically segregate by period — a "tug of war" in which overnight and intraday premia are often opposite in sign.

Honesty matters here. This is primarily an index- and clientele-level effect, not a clean per-stock edge any swing trader can simply harvest. It is time-varying: practitioner analyses (e.g., Elm Wealth's review) show the raw drift weakened materially after the 2008–2015 papers circulated, consistent with partial arbitrage. And it is largely uninvestable at scale once realistic transaction costs and market impact are subtracted — the gross long/short numbers do not survive net of costs for large size. The useful takeaway is directional, not a strategy: multi-day holds have historically participated in a return component that closes daily, which is part of why a holding-period edge can exist — but it is a tailwind that varies by regime and security, never a guarantee.

Institutional accumulation

The second rationale is mechanical rather than calendar-based. A fund that wants a large position cannot buy it in one order without moving the price against itself, so it splits the parent order (a "metaorder") into many child orders executed over hours, days, or weeks. The market-microstructure literature describes the resulting price impact with the square-root law — impact scales roughly with the square root of order size relative to volume — and documents that this execution is deliberately spread out and disguised to minimize footprint (see the metaorder / price-impact literature, e.g., work surveyed in arXiv microstructure papers).

For a swing trader this matters because a genuinely large accumulation is not instantaneous: it tends to leave a persistent, multi-day signature — rising price on sustained, above-average volume, shallow pullbacks that get bought, relative strength versus peers. The swing trader's thesis is to detect that footprint early and hold alongside it until the accumulation (and the drift it creates) is complete. That is the legitimate kernel behind "follow the smart money." It is also frequently over-claimed: most volume is not a single coordinated institution, footprints are inferred and noisy, and algos specifically work to make accumulation look like ordinary flow.

The honest caveat

Treat both of these as sources of potential edge and rationale, not edges you can switch on. Overnight drift is real in the historical record but index-level, regime-dependent, decaying, and not investable net of costs at scale. Institutional accumulation genuinely leaves multi-day footprints, but those footprints are inferred, easily imagined where none exists, and actively obscured. They explain why a multi-day holding period can be rewarded; they do not tell you which stock, when, or how much. Any real swing strategy still rests on risk management, position sizing, and confirmation — these phenomena justify the time horizon, they do not replace the work.

Sources