Two drinks: what alcohol actually does to your sleep

01Does alcohol help you fall asleep?

Partly — and that's the trap. The classic review of every controlled study on healthy sleepers found alcohol shortens the time to fall asleep and consolidates the first half of the night at essentially any dose [1]. A newer meta-analysis of 27 studies is stricter: the reliable fall-asleep-faster effect showed up only at high doses — roughly 0.85 g/kg, about 5 standard drinks [2].

Both agree on the back half. As the alcohol clears, the second half of the night runs the tab: more waking, lighter sleep, the 3 a.m. stare at the ceiling [1]. The nightcap isn't sleep aid pricing. It's a payday loan.

02How many drinks does it take to hurt REM sleep?

About two. The 2025 meta-analysis found REM sleep disruption beginning at a low dose — 0.5 g/kg or less, which the authors put at roughly 2 standard drinks — and worsening progressively from there [2]. The first REM period arrives later at every dose, and total-night REM falls in most studies at moderate and high doses [1].

REM is the sleep stage most tied to memory and mood consolidation, and it's concentrated in the second half of the night — exactly the half alcohol fragments. That's why 7 hours after wine can feel like 5.

03Why does your watch score tank after drinking?

Because your nervous system spends the night in traffic instead of parked. In 4,098 Finnish adults wearing beat-to-beat heart-rate recorders through normal life, drinking evenings ran higher heart rate, lower heart-rate variability, and less physiological recovery in the first 3 hours of sleep than the same person's dry evenings — dose-dependently, and starting at the lowest dose bracket [3].

Two findings worth sitting with: being fit didn't blunt the effect, and being young made it stronger, not weaker [3]. The gym does not pre-pay the bar tab.

04How long before bed should you stop drinking?

Earlier helps — the liver clears alcohol at a steady rate, so a smaller dose finished sooner means less of it riding into the night. But the fade is slower than the buzz. In one trial, 10 middle-aged men drank 0.55 g/kg — call it 3 drinks — a full 6 hours before bedtime. Breath alcohol was zero in every one of them by lights-out. Sleep still suffered: total sleep and REM fell, and second-half wakefulness doubled [4].

Small study, older men, one dose — but it's the cleanest demonstration that zero on the breath is not zero in the brain. The practical version: the two dials that matter are how much and how often. When you drink matters, and it matters less than either.

05So are two Friday beers a problem?

Here's the honest ledger. Two beers on Friday, done by 8 p.m., is a modest, contained cost: some REM loss, a lower recovery score, maybe a warmer first half of the night [2][3]. Saturday morning with the kids will be fine. What the evidence argues against isn't the ritual — it's the drift: two becoming four, Friday becoming most nights, the drink migrating toward bedtime because it “helps you sleep.” It doesn't. It sedates you, then invoices you at 3 a.m. [1]. Keep the beers if they're worth it to you. Just pay for them with open eyes, not with your Saturday.