Training Session Analysis (build 43)
J70 Grand Slam 1 Race 4 — Saturday, May 9, 2026
Data Analytics
GPX file: speed and course aren't recorded in this format, so they're worked out from how far the boat moved between fixes (over a 6-second window, which is what makes the numbers steady enough to be useful). Expect them to be close but not identical to a dedicated instrument's. A GPX carries position and time only — there's no compass or heel sensor behind it, so heading is course over ground and heel isn't available. There's no start line or start gun either, which is what the Training Session Analysis is for.

Race window

This file has no race timer, so the whole recording is being analysed — 03:02 PM to 04:09 PM. Set a finish time below to cut off the sail home.

Manually cut the race off at a finish time — anything logged after it is ignored everywhere below (stats, charts, everything).

Race replays

The whole race, the start, each leg and each mark rounding, played back on a wind-up map. Pick one and press play — the boat is drawn at its real heading, at real elapsed race time. Individual tacks and gybes have their own overlaid traces further down, which say more about a turn than watching it does.

0.00 m269.47 m538.94 m808.41 m1077.88 m1347.35 m1616.82 m1886.29 m2155.76 mTWD122°−3°
0s
0.00 kts  ·  355°  ·  126° TWA  ·  0.00 kts VMG upwind
Ladder linesturning about the next mark, averaged over 10s

Wind

Manually entered

Across 15 tacks, your settled headings average 175° and 69° — 106° apart — which puts the wind at 122° (high confidence). That agrees with the direction in use, so nothing needs changing.

Which wind to analyse against

A boat sailing a settled straight line is a wind vane: its heading sits half its own tacking angle to one side of the wind (half its gybing angle either side of dead downwind). Turns, and the seconds either side of them, are left out, as is the minute around every mark rounding. Live is the default, because a wind that moved is better described by a wind that moves — switch to fixed to use one direction throughout.

Reading the wind off your own track over 52 min, averaging 124° and swinging 25° across the race. One boat is one wind vane, so treat this as a good guide rather than an instrument — the fleet tools average the same reading across every boat in the race.

The wind through the race

Wind direction read off your own straight-line sailing, minute by minute. The dashed line is the race average (124°); the axis is labelled in real bearings. Total swing across the race: 25°.

114°119°124°129°134°03:04 PM03:17 PM03:30 PM03:43 PM03:56 PM

Ticks along the bottom are your tacks and gybes.

Track

Scroll to zoom · drag to pan · hover for details
Open this area in OpenStreetMap ↗

Boat speed

Heading & manoeuvres

Polar — speed (SOG) by wind angle

Upwind leg comparison

Windward legs between tacks, compared on boat speed and heel angle.

Leg Time Tack Duration Distance Avg speed (SOG) Max speed (SOG) Avg speed (STW) Max speed (STW) VMG Avg heel Max heel Avg heading Wind angle

Upwind leg by leg breakdown

Each upwind leg mark to mark (not tack to tack), one column per leg in race order.

Downwind leg by leg breakdown

Each downwind leg mark to mark (not tack to tack), one column per leg in race order.

Tack comparison (±30s)

Manoeuvre instantDarker = further made good

Every tack, aligned and rotated so "up" is toward the wind — overlaid to compare distance made good, ranked best to worst. Row numbers match the "#" column in the table below. Tacks missing the full before/after window are excluded.

-50m0m50m-50m0m50mtoward wind ↑#2 · new tack Port — made good +48.61 m (rank 1)#2 — made good +48.61 m (rank 1)1#7 · new tack Starboard — made good +46.26 m (rank 2)#7 — made good +46.26 m (rank 2)2#6 · new tack Port — made good +45.44 m (rank 3)#6 — made good +45.44 m (rank 3)3#21 · new tack Port — made good +42.39 m (rank 4)#21 — made good +42.39 m (rank 4)4#17 · new tack Port — made good +42.09 m (rank 5)#17 — made good +42.09 m (rank 5)5#18 · new tack Starboard — made good +37.96 m (rank 6)#18 — made good +37.96 m (rank 6)6#19 · new tack Port — made good +37.85 m (rank 7)#19 — made good +37.85 m (rank 7)7#8 · new tack Port — made good +34.40 m (rank 8)#8 — made good +34.40 m (rank 8)8#9 · new tack Starboard — made good +33.23 m (rank 9)#9 — made good +33.23 m (rank 9)9#10 · new tack Port — made good +32.10 m (rank 10)#10 — made good +32.10 m (rank 10)10The manoeuvre itself (t = 0)
Rank # Time New tack Made good Lost
1#203:06 PMPort+48.61 m3.0 m
2#703:12 PMStarboard+46.26 m3.8 m
3#603:09 PMPort+45.44 m1.0 m
4#2103:54 PMPort+42.39 m3.6 m
5#1703:44 PMPort+42.09 m0.0 m
6#1803:44 PMStarboard+37.96 m5.8 m
7#1903:48 PMPort+37.85 m1.3 m
8#803:20 PMPort+34.40 m3.5 m
9#903:24 PMStarboard+33.23 m0.7 m
10#1003:28 PMPort+32.10 m6.5 m

Speed through each tack

Pick the tacks you want to compare — click to add one, click again to drop it. 0s is the tack itself. Each line takes the colour of its own chip; the "#" is the same tack number as the table above. Speeds are smoothed with a 4-second average so the shape of each turn reads through the GPS noise; the figures in the table are from the raw track.

Compare:
01234567kts-30s-20s-10s0s+10s+20s+30s#2 — SOG (rank 1 by distance made good)
#2SOGsmoothed over 4s

Gybe classification

A turn counts as a gybe when it's downwind of head-to-wind and clears this heading-change minimum (tacks always need >45°). Lower it if real gybes are missed, raise it if wobble gets misclassified. Affects every gybe analysis below.

Gybe comparison (±30s)

Manoeuvre instantDarker = further made good

Same as the tack comparison, but for gybes, ranked by distance made good downwind. Has its own before/after window.

-50m0m50m-100m-50m0m50m100mtoward wind ↑#23 · new tack Port — made good +87.69 m (rank 1)#23 — made good +87.69 m (rank 1)1#15 · new tack Starboard — made good +79.62 m (rank 2)#15 — made good +79.62 m (rank 2)2The manoeuvre itself (t = 0)
Rank # Time New tack Made good Lost
1#2304:00 PMPort+87.69 m1.0 m
2#1503:41 PMStarboard+79.62 m3.2 m

Speed through each gybe

Pick the gybes you want to compare — click to add one, click again to drop it. 0s is the gybe itself. Each line takes the colour of its own chip; the "#" is the same gybe number as the table above. Speeds are smoothed with a 4-second average so the shape of each turn reads through the GPS noise; the figures in the table are from the raw track.

Compare:
012345678kts-30s-20s-10s0s+10s+20s+30s#23 — SOG (rank 1 by distance made good)
#23SOGsmoothed over 4s

Top mark rounding comparison (±30s)

Manoeuvre instantDarker = further made good

Every top-mark rounding, overlaid and ranked by distance made good downwind afterward — helps spot a slow or wide rounding. Has its own (typically wider) before/after window.

-50m0m50m0mtoward wind ↑#6 — made good +35.78 m (rank 1)#6 — made good +35.78 m (rank 1)1#2 — made good +28.44 m (rank 2)#2 — made good +28.44 m (rank 2)2#4 — made good +22.45 m (rank 3)#4 — made good +22.45 m (rank 3)3The manoeuvre itself (t = 0)
Rank # Time Made good Lost
1#603:58 PM+35.78 m53.5 m
2#203:07 PM+28.44 m36.8 m
3#403:32 PM+22.45 m51.0 m

Speed through each top-mark rounding

Pick the roundings you want to compare — click to add one, click again to drop it. 0s is the rounding itself. Each line takes the colour of its own chip; the "#" is the same number as the table above. Speeds are smoothed with a 4-second average so the shape of each rounding reads through the GPS noise; the figures in the table are from the raw track.

Compare:
01234567kts-30s-20s-10s0s+10s+20s+30s#6 — SOG (rank 1 by distance made good)
#6SOGsmoothed over 4s

Bottom mark rounding comparison (±30s)

Manoeuvre instantDarker = further made good

Every bottom-mark rounding, overlaid and ranked by distance made good upwind afterward. Has its own before/after window.

-60m-40m-20m0m20m0m20m40mtoward wind ↑#5 — made good +28.91 m (rank 1)#5 — made good +28.91 m (rank 1)1#3 — made good +23.91 m (rank 2)#3 — made good +23.91 m (rank 2)2#1 — made good +23.54 m (rank 3)#1 — made good +23.54 m (rank 3)3The manoeuvre itself (t = 0)
Rank # Time Made good Lost
1#503:43 PM+28.91 m6.9 m
2#303:08 PM+23.91 m0.0 m
3#103:03 PM+23.54 m14.8 m

Speed through each bottom-mark rounding

Pick the roundings you want to compare — click to add one, click again to drop it. 0s is the rounding itself. Each line takes the colour of its own chip; the "#" is the same number as the table above. Speeds are smoothed with a 4-second average so the shape of each rounding reads through the GPS noise; the figures in the table are from the raw track.

Compare:
012345kts-30s-20s-10s0s+10s+20s+30s#5 — SOG (rank 1 by distance made good)
#5SOGsmoothed over 4s

Tacks & gybes

# Time Type New tack Heading Speed (SOG, before → min) Recovery

Sail Smarter Race Analytics is an independent project. Not affiliated with, endorsed by, or connected to Vakaros or Velocitek — those names, and the VKX and VTK file formats, are referred to only to describe compatibility. All trade marks are the property of their respective owners.