USDCAD after BoC: the 30-pip move from Tokyo to New York
On the trading day after a Bank of Canada rate decision, how often does USDCAD move 30 pips from the Tokyo open—and where does it finish at New York’s close?
The finding
A 30-pip move in at least one direction occurred in about nine out of ten sessions. Among sessions that touched that distance, just over half finished at least 30 pips from the opening price in either direction; 43.75% returned inside that distance by the close. A touch was therefore much more common than retaining a move at the close.
During the session
Reference: USDCAD price at 09:00 Tokyo on the next trading day after the decision. Observation ends at 17:00 New York that day, accounting for US daylight saving. This includes all qualifying weekdays, not only Thursdays.
| Level reached | Sessions | Frequency |
|---|---|---|
| At least +30 pips | 91 / 159 | 57.23% |
| At least −30 pips | 95 / 159 | 59.75% |
| Either +30 or −30 pips | 144 / 159 | 90.57% |
| Both +30 and −30 pips | 42 / 159 | 26.42% |
Upside and downside groups overlap. A touch can occur at any time during the window. A rise in USDCAD means USD strengthens against CAD.
What remained at New York’s close?
| Starting condition | Closing outcome | Frequency |
|---|---|---|
| Touched either ±30 | Closed at least 30 pips away, either direction | 56.25% · 81 / 144 |
| Touched either ±30 | Closed inside ±30 pips | 43.75% · 63 / 144 |
| Touched +30 | Closed at or above +30 | 40.66% · 37 / 91 |
| Touched −30 | Closed at or below −30 | 46.32% · 44 / 95 |
Across all 159 sessions, 23.27% closed above +30 (37 sessions), 27.67% below −30 (44), and 49.06% inside ±30 (78). No close was exactly ±30 pips. Irrespective of distance, 48.43% closed higher than the Tokyo open and 51.57% lower.
The average open-to-high distance was 49.1 pips, average open-to-low distance 47.1 pips, and average high-to-low range 96.2 pips. These are price distances, not trading returns.
Session rules, sample and data quality
The research window is September 16, 2006–September 15, 2026. Of 163 requested sessions, 159 had complete hourly coverage: 156 after scheduled decisions and three after emergency decisions. Rate holds are included. Included sessions run from October 18, 2006 to September 3, 2026. The emergency decisions were October 8, 2008, March 13, 2020 and March 27, 2020.
The next weekday after each decision was selected, skipping weekends, and its hourly coverage checked. This differs from the EURUSD/FOMC study’s Thursday-only selection.
Four incomplete sessions were excluded, with no missing prices imputed: June 5, 2009 and September 11, 2009 (20 of 21 hourly candles each); March 2, 2011 and January 21, 2021 (21 of 22 each). Their missing hours persisted on rechecking.
Tokyo open is 09:00 Asia/Tokyo (00:00 UTC). New York close is 17:00 America/New_York: 21:00 UTC in US summer time or 22:00 UTC in winter. The first hourly candle’s open is the reference; the close is the last quote before the exclusive ending boundary. Each included session has all 21 or 22 expected hourly candles. One USDCAD pip is 0.0001.
Source: hourly USDCAD history from the FxPro demo platform. Broker clock conversion assumes London local time plus two hours, using historical timezone rules. This clock convention has not been independently verified for every historical year. Complete hourly timestamps do not prove completeness of every tick; the order of touches within an hourly candle is not reconstructed.
Decision dates were checked against official Bank of Canada information and annual press-release archives. Original records, event mappings, exclusions and calculations are retained for audit and available on request at research@nerdquant.com.
What this study does—and does not—measure
This is descriptive price research. It places no trades and does not model spreads, commissions, slippage, option premiums or executable fills. Frequencies use the denominators shown; historical observations are not forecasts. The sample spans different volatility regimes and does not establish that BoC decisions caused these moves.
Average holding period and average maximum account drawdown: not applicable. There are no simulated positions or account equity curve. The observation window is 21–22 hours; an open-to-low price excursion is not an account drawdown.
Research date: September 16, 2026. Percentages may differ slightly from 100% when summed because of rounding.