Winner Route Summary - Newport to Bermuda Race
Rev: v1.4 · Updated: 2026-05-29
Annotated tactical summary of route choices by race year, cross-referenced with the Routing Tactic Catalog and the W-* weather analogs.
TL;DR
- The default is near-rhumb + low-maneuver (T-RH + T-CL); deviate only on multi-source-confirmed current. Clean execution won in all six years.
- Light-air years (2014, 2018, 2024) deliver the strongest Stream (up to 5 kt NE core) — the Stream becomes the equalizer, not boat speed.
- The west corridor is a trap most years. RTOFS-only signals lured boats west in 2016 (~5 h slower than rhumb); only 2022 rewarded the east limb (59.4 h median, fastest in the set) — and only with SST + altimetry confirmation.
- Each year carries a Lupo Comparable (41–44 ft racer-cruiser); weight those choices at least equally with the fleet winner where they differ.
How to read: find your pre-race analog year, then check what the winner did, what the comparable-size boat did, and what the trap was. Annotated YB-track maps sit at the bottom of each section — gold = corrected winner's actual track.
Corridor medians by year
Median elapsed (hours) for the strict 40–45 ft cohort binned by Gulf Stream crossing corridor, from YB tracker tracks 2012–2024 (scripts/analysis/stream_corridor_analysis.py). n in parens.
| Year | Cohort | EAST | RHUMB | WEST | Read |
|---|---|---|---|---|---|
| 2024 | 64 | 86.4 (1) | 67.7 (52) | 67.2 (3) | Rhumb year; east/west samples too thin to read |
| 2022 | 67 | 59.5 (29) | 60.6 (30) | 75.1 (3) | East limb fastest in the whole set; ~15.7 h west penalty |
| 2018 | 54 | 95.4 (2) | 93.5 (39) | 97.6 (7) | Light-air park; west ~4 h slower than rhumb |
| 2016 | 67 | 77.3 (1) | 78.7 (21) | 83.7 (23) | West ~5 h slower — the RTOFS-only CCR trap |
| 2014* | 70 | — (0) | 124.0 (28) | 124.0 (37) | Slowest year; west ≈ rhumb |
| 2012 | 71 | — (0) | 87.8 (47) | 84.9 (6) | West 3 h faster but n=6 — not decisive |
*2014 is the held-out seed/check year. Bold = the corridor the corrected winner used.
Read these within-year only. Medians compare corridors inside one wind regime; they are not cross-year speed-comparable. EAST is statistically meaningful only in 2022 (n=29) — every other year's EAST cell is n=1–2.
2024 - W-PF / W-BH / W-RC: Classic Reaching Race, Strong Current Year
Line honors: Pyewacket (Volvo 70) — 59h 18m
Corrected winner: Carina (Rives Potts) — 64h 12m corrected; 5th NB win
SDL runner-up: Hound (Dan Litchfield) — 64h 25m including 30-min OCS penalty
J/122: Ceilidh finished; Alliance sank after striking submerged object
Confidence: MEDIUM
Synoptic: Front north of the start produced a slow, fluky first day. Once front cleared: 14–24 kts SW reaching. Gulf Stream up to 5 kts NE — one of the stronger current years.
Gulf Stream: Strong NE current; southerly meander creating NE-flowing eastern limb — NOT a CCR. Critical challenge: exit before being set too far east. "We needed to figure out when to get off the ride."
Carina (corrected winner): Navigator Chris Lewis (also won 2022) credited early departure from the RI coast — cleared the frontal light patch, exploited the meander, exited at the right latitude before being set east. Low maneuver count, clean reaching. Tactics: T-PF (start), T-ME (meander exploit + exit), T-CL.
What lost: Overstaying the meander and being set too far east.
Lupo Comparable (~41–44 ft racer-cruiser): - Hound (Dan Litchfield) — SDL5 class winner, 64h 25m corrected with a 30-min OCS penalty; faster than Carina before the penalty. Route: early departure, cleared the frontal patch, clean reaching. Tactics: T-PF, T-ME, T-CL. [COLLECT — confirm Hound LOA/class in 41–44 ft bracket.] - Ceilidh (J/122) — finished. [COLLECT — corrected time / position.] - Alliance (J/122) — sank after striking a submerged object at 0300; no performance data.
Safety note: Alliance's upper rudder bearing ripped out, causing catastrophic flooding; all crew rescued by Ceilidh. Brief Lupo's crew on flooding response and bilge-pump location before departure.
Lesson for Lupo: Start position is race-defining. A comparable-size boat (Hound) executing cleanly led the fleet — but a 30-min OCS converted a likely overall win into a near-miss. Do not OCS. Get south of the frontal patch early, exploit the current core, and pre-plan the meander exit latitude.
2022 - W-PF / W-SQ: Post-Frontal Squalls, Forming Cold-Core Ring
Line honors: Argo (MOD70)
Corrected winner: Illusion (Cal40, Stan Honey) — ~51h corrected
J/122 result: Zig Zag (J/122, Andrew Clark) — 2nd overall SDL
Confidence: MEDIUM
Synoptic: Cold front crossed the fleet during the first night. Post-frontal NW 18–24 kts. A forming CCR (oxbow meander) sat near the rhumb crossing latitude.
Gulf Stream: CCR partially formed, not fully closed. RTOFS showed confused current; SST confirmed the feature. Boats east of the western limb were fine; boats that dove in lost significantly. The east corridor median was 59.5 h — the fastest in the whole set — at a ~15.7 h penalty for going west (see corridor medians).
Illusion (corrected winner): Stan Honey navigated into a Stream eddy, held it ~7 h for maximum favorable current, then exited at the right position. Post-frontal NW heavy-air reaching is the boat's strength; top speed 22 kt. Tactics: T-PF, T-ME (eddy exploit), T-CL.
What lost: Entering the CCR western limb (2–3 kt adverse in the ring interior); late kite douse ahead of squalls.
Lupo Comparable (~41–44 ft racer-cruiser): - Zig Zag (J/122, Andrew Clark) — 2nd overall SDL, immediately behind Illusion. Rhumb-line approach, clean post-frontal NW reaching, no eddy detour. Tactics: T-PF, T-RH, T-CL.
Lesson for Lupo: Post-frontal NW is the money leg — kite up as soon as the front clears. Zig Zag's rhumb + clean-crossing plan nearly beat Illusion's 7-h eddy hold and is far more replicable — prioritize fast reaching over a complex current detour. Identify the CCR before crossing and stay off the western limb.
2018 - W-LA / W-PL / W-BH: Parking Lot Year
Line honors: Rambler 88
Corrected winner: Grundoon (Columbia 50, 112:12 corrected)
Best J-boat: J/121 Apollo — ORC Class 14 winner (~62:21 corrected)
Confidence: MEDIUM
Synoptic: High pressure stall mid-race created a parking lot south of the Stream. Fleet waited hours in near-zero wind at 30–33°N. Boats that threaded a narrow pressure corridor or timed the crossing to avoid the park got dramatically better elapsed times.
Gulf Stream: Up to 5 kt NE in the core — exploitation decisive because wind was so light. West corridor ran ~4 h slower than rhumb (97.6 vs 93.5 h; see corridor medians).
Grundoon (corrected winner): Columbia 50 is a light-air specialist in ideal conditions — sailed carefully through the high-pressure zone and exploited the current aggressively. Tactics: T-CF, T-PF (light), T-CL.
What lost: Stuck in the park at 30–33°N; forcing sail changes searching for nonexistent wind. Large boats that expected to motor away in pressure got caught in dead air.
Lupo Comparable (~41–44 ft racer-cruiser): - J/121 Apollo — ORC Class 14 winner (~62:21 corrected); best comparable-size boat in the 2018 fleet. Tactics (consistent with the W-LA class win): T-CF, T-CL. [COLLECT — Apollo's Stream crossing / light-air routing from official 2018 results.]
Lesson for Lupo: In W-LA, current exploitation is decisive — Apollo's Class 14 win confirms this applies directly to Lupo's bracket. Find the Stream core and use it; fly the lightest sail early; do not thrash through changes in the park. Lupo (DLR 162) is moderate-light, not a featherweight — the Stream is the equalizer, and patient execution beats raw boat speed.
2016 - W-PF / W-PL / W-CC: Cold-Core Ring Trap
Line honors: Comanche (34:53 — record at the time)
Corrected winner: Warrior Won (Xp44, ~70:40 elapsed)
J/122 result: Alliance retired — water ingress
Confidence: MEDIUM-HIGH
Synoptic: Pre-frontal southerly → post-frontal NW. High pressure built south mid-race. Gulf Stream had a partially developed CCR near the rhumb crossing latitude.
Gulf Stream: The CCR was the defining feature. RTOFS predicted favorable current on the ring's western limb — the prediction was wrong; the ring underdelivered badly. The west corridor ran ~5 h slower than rhumb (83.7 vs 78.7 h; see corridor medians) — a near-even east/west split (n=23 each) where the west half chased an RTOFS-only signal and lost.
Warrior Won (corrected winner): Stayed near rhumb or slightly east of the CCR; did not chase the RTOFS-predicted current that never materialized. Tactics: T-PF, T-RH, T-CL.
What lost: Committing to the CCR western limb on RTOFS alone — RTOFS was 20–40 nm off; boats extended 30–50 nm west for minimal gain.
Lupo Comparable (~41–44 ft racer-cruiser): - Warrior Won (Xp44) — overall corrected winner (~70:40 elapsed). The Xp44 is a direct Lupo comparable by design and performance profile, so the corrected winner IS the comparable — the highest-confidence 2016 data point for Lupo. - Alliance (J/122) — retired (water ingress); no performance data. [COLLECT — other 41–44 ft class results from official 2016 results.]
Critical lesson for Lupo: Do NOT commit to a CCR routing maneuver on RTOFS alone. Warrior Won won by staying near rhumb and refusing RTOFS predictions that were 20–40 nm off. Cross-validate RTOFS with SST and altimetry (AVISO) before trusting it — see the Cold Core Ring guide.
Note: 2016 YB data has a start-timestamp issue; cluster of points near start area before lat drops below 41°N is a recording artifact.
2014 - W-LA / W-PF / W-RC: Light Air, Frontal Zone, Strong Current
Line honors + Gibbs Hill corrected: Shockwave (Reichel/Pugh Mini Maxi)
SDL corrected winner: Actaea (modified Hinckley B40 yawl, Michael Cone) — won by 45-min margin
J/122 result: COLLECT
Confidence: LOW-MEDIUM
Synoptic: Light air dominated — "one of the slowest races in recent history." Pre-race forecasts significantly inaccurate. "Airless sinkholes" plagued the fleet. Frontal zone moved across the course during the race, shutting down wind in its wake.
Gulf Stream: Up to 5 kt NE push in the core. "The trick then became exiting at the right moment" — boats that overstayed were set too far east. This is the slowest year in the set (rhumb and west medians both ~124 h; see corridor medians), with west ≈ rhumb.
Winning boats: Went "a conservative distance to the west" — the right call as the frontal zone shut down wind on the east side — then exploited the current and exited before being set too far east. Tactics: T-PF (west for pressure), T-CF, T-CL.
What lost: Following inaccurate pre-race forecasts into frontal-zone light air; failing to exit the Stream at the right latitude.
Lupo Comparable (~41–44 ft racer-cruiser): [COLLECT — 41–44 ft class results, and J/122/J/121/J/120 finishes, from bermudarace.com official 2014 results.] Until confirmed, use 2018 Apollo (J/121 Class 14) as the proxy.
Lesson for Lupo: Third light-air data point (with 2018 and 2024) — these years deliver exceptional Stream current and the Stream becomes the primary differentiator. Watch for frontal-zone east-of-rhumb wind shutoff; conservative western routing was right here. Pre-race models were wrong in 2014 — in W-LA years, plan for both W-LA and W-BH scenarios.
2012 - W-PF / W-BH: Fast Reaching Race
Line honors: Rambler (90ft, 39:39)
Corrected winner: Carina (~48ft)
Comparable J-boat: J/120 Mireille — ~55:57 corrected
Confidence: MEDIUM
Synoptic: Post-frontal NW–N at 15–25 kts through most of the race. Fast tactical reaching race.
Gulf Stream: Normal position; no significant CCR or WCE. Stream benign — not the decisive factor. The west corridor edged rhumb (84.9 vs 87.8 h; see corridor medians), but n=6 west makes that not decisive.
Carina (corrected winner): Simple, clean execution — ~2 jibes and 2 tacks for the whole race. Stayed near rhumb, did not chase lateral positioning, sailed consistently fast at the polar sweet spot. Tactics: T-RH, T-PF, T-CL.
What lost: Overplaying lateral positioning in a year where rhumb was near-optimal; too many maneuvers chasing marginal pressure.
Lupo Comparable (~41–44 ft racer-cruiser): - J/120 Mireille — ~55:57 corrected; best comparable-size boat in the 2012 fleet, the benchmark for a W-PF reaching race. Stayed competitive through clean execution, not lateral positioning. Tactics (inferred): T-RH, T-PF, T-CL. [COLLECT — Mireille routing detail and corrected gap vs Carina.]
Lesson for Lupo: In a W-PF reaching year, execute simply and fast. When rhumb is working, do not overthink lateral positioning — every extra maneuver chasing marginal pressure costs more than it gains. Carina's ~2-jibe / 2-tack run over 636 nm is the benchmark for low-maneuver, high-execution racing.
Pattern Summary by Tactic
| Tactic | Worked | Failed | Net verdict |
|---|---|---|---|
| T-RH near rhumb | 2012, 2022 (partial), 2024 | 2016 if ring on rhumb | DEFAULT; deviate only on confirmed data |
| T-CF current first | 2014, 2018, 2024 (all 5 kt Stream) | 2016 (CCR detour) | Valid only if current confirmed by SST + altimetry |
| T-ME meander exploit + exit | 2022 (7 h eddy), 2024 (meander) | None in set | High value when confirmed; needs pre-planned exit latitude |
| T-CL conservative low-maneuver | 2012, 2014, 2018, 2022, 2024 | None in set | Preferred default for Lupo |
| T-PF pressure first | 2012, 2014 (west), 2022, 2024 | 2018 (light air) | Strong in frontal years; moderate in light air |
| T-CCR cold-ring exploit | Theoretical | 2016 (decisively) | Requires SST + altimetry; high risk |
Across all six years the winner ran a low-maneuver, near-rhumb route with opportunistic Stream use confirmed by multiple data sources. Light-air years (2014, 2018, 2024) deliver the strongest current — in W-LA, the Stream is the equalizer.
Sources
- Corridor medians & cohort sizes —
scripts/analysis/stream_corridor_analysis.pyover YB tracker tracks 2012–2024 (strict 40–45 ft cohort); cached indata/yb_cache/stream_corridor_stats.json. - W-* analogs & per-year classification — weather_analog_framework.md.
- Tactic codes (T-*) — routing_tactic_catalog.md.
- Gulf Stream features (CCR / WCE / RTOFS) — gulf_stream_framework.md.
- Race results & narratives — bermudarace.com race wraps, North Sails victory list, Sail World, Scuttlebutt, Seas of Solutions.
PRE-RACE RESEARCH — not race-period routing advice. Updated 2026-05-29.