Often, but not because they "attract" price. The New York session took the overnight high on 63% of days and the previous day's high on 56%. Yet a level with no name at the same distance from the open was reached just as often (difference 0.0 ± 1.5 percentage points): what matters is distance, not "liquidity". And fading the overnight sweep, after costs, averaged $4.09 per trade (±$7.88): no proven edge.
We measured 1,813 Micro Nasdaq sessions (2019-2026), New York time. "Takes the high" means the 9:30 am-4:00 pm session trades above it at some point, including when it opens above it.
How often New York takes each high and low
| Reference | Takes the high | Takes the low | Both | Already broken at 9:30 |
|---|---|---|---|---|
| Asia (20:00-24:00) | 78% | 71% | 48% | 70% |
| London (2:00-5:00 am) | 76% | 71% | 47% | 55% |
| Full overnight (6:00 pm-9:30 am) | 63% | 56% | 24% | 0% |
| Previous day (9:30 am-4:00 pm) | 56% | 44% | 9% | 41% |
The Asian range is small: on 70% of days it's already broken before New York opens, so "taking the Asian high" usually happens overnight. The high is taken more often than the low because the Nasdaq rose over these years.
Do these levels attract price?
If the market "hunted" the liquidity above highs and below lows, it would reach those levels more often than any other level at the same distance. We compared each level with its mirror: the same distance from the open, on the other side, where nothing special sits.
| Distance from the open | Cases | Reaches the level | Reaches the mirror level |
|---|---|---|---|
| 0× – 0.1× | 1580 | 92% | 92% |
| 0.1× – 0.25× | 2367 | 76% | 76% |
| 0.25× – 0.5× | 2229 | 56% | 59% |
| 0.5× – 1× | 1657 | 34% | 32% |
| over 1× | 537 | 15% | 12% |
Distance as a multiple of the median daily range of the previous 20 sessions. All four references pooled, 8,370 cases, only days that open inside the range.
There's no difference. A level within a tenth of a normal day's range gets reached 92% of the time, and one more than a whole day away 15%, whether it's the London high or an arbitrary price. The table works as a probability chart for your targets.
Fading the sweep, with money
The popular setup: the session breaks the overnight high (or low), and within 30 minutes a 1-minute bar closes back inside. Sell (or buy) at the next bar's open with 1 MNQ, stop 1 tick beyond the extreme, exit at 4:00 pm. Commission $1.20 and 1 tick of slippage in and out.
- 1,481 trades: $4.09 on average (±$7.88 of 95% margin of error). Zero is well inside the margin.
- Lost 87% of the time. The stop is tight (about $44 a trade) and few trades run to the close.
- Negative in 3 of 8 years. A few big days carry the average.
What it's good for
- Targets by distance, not by name: use the table above to estimate how likely your target is to be reached today.
- Stops beyond obvious highs and lows get hit as often as any other price at that distance: size them for normal movement (see how much each hour moves).
- If you trade sweeps, log them and measure your own results with enough trades (how many?).
Educational content based on our own experience. Not investment advice. Firm rules change: check them on each firm's website. This article contains no affiliate links.