This topic has been up for a few weeks now, and it has taken me a while to answer because I wanted to put together a thoughtful reply.
I’ll start out with the caveat that my post below applies to kayaks. I do row my Goat Island Skiff a bit, but not enough that I can provide any meaningful data.
It probably goes without saying, but the best way to compare boats is to paddle them, ideally side by side in the conditions that they will be used. Paddler power is half of the speed equation, and power is significantly impacted by a number of factors (comfort, stability, stroke dynamics, etc) which can only be determined when you put your butt in a boat and paddle it for a while. This is a lesson that I recently learned the hard way when I bought a commercial race boat without paddling it. Although it is by far the fastest hull that I own, it becomes a medieval torture device after about 30 minutes and speed drops precipitously.
In the homebuilt world, it is sometimes very difficult to get demo paddles. In that case, probably the best way to compare speeds is to use Kaper drag data if you can get it. CLC does not publish this data but it was included in all the reviews done by the old Sea Kayaker Magazine. Luckily, I bought copies of their reviews for both the Ch17 and S&G Petrel Play before they went under. Some designers, like Nick Schade, Bjorn Thomassen and Dan Cauette (Clearstream Custom Watercraft) provide drag data on their designs.
If you can’t demo the boats and can’t get the drag data, you are left with using speed data reported by other paddlers. Paddler power is half of the speed equation, so it is important to understand what kind of a “motor” was used to generate the speed data. A skilled, fit, thin, young paddler may report speeds in his Wood Duck that are greater than a portly mature paddler’s speed in a Pax 18, although the latter boat is undoubtably faster. For me, the greatest value in this kind of data is to create speed differentials between boats. If a paddler reports that he is 0.2 MPH faster in boat A than boat B, it is likely that I will see similar results.
One last item to discuss is the use of GPS for speed measurement. I suspect that most of the numbers reported above are from GPS receivers. The bottom line is that GPS computed speeds for short sprint distances (<30-60 seconds) are meaningless. For this reason, I will not be providing “sprint” speeds. More info provided below.
With all of that out of the way, here is the data for our kayaks, including two commercial boats. My motor is 58 years old, 5’7”tall and weighs 152#. I would probably be classified as an advanced paddler and am in excellent aerobic condition. Race pace data is 8-12 miles, heart rate ~152 and an 80 SPM cadence with Epic Small Mid-Wing paddle. All day pace is heart rate ~125 with a 70 SPM cadence and smaller Jantex Rio Small Plus Wing paddle. Data was recorded with Garmin Fenix 5x and downloaded to Garmin Connect:
S&G Petrel Play (14’ x 23”) – Race 5.2 mph (11:30 min/mile) – All Day ~4.8 mph (12:30 min/mile)
90% BTD Frej (15’ x 19”) – This is my wife’s boat, so I paddle it very little. I have one 5-mile paddle averaging 5.4 mph (11:05 mile/min). This boat is very tight for me so I can’t get my legs into the stroke and get very little rotation. I would call the pace midway between race and all day. This is a very sleek boat and is quite fast for its length. With a higher deck, I don’t doubt that I could paddle it faster than either the 16LT or 17LT.
Ch16LT (15’6” x 23”) – Race 5.4 mph (11:06 min/mile). All Day 5.2mph (11:30mile/min).
Ch17LT (16’10” x 23”) – Race 5.5 mph (11:00 min/mile). All Day 5.2 mph (11:30 mile/min).
Shearwater Double (18.5’ x 27”) – I don’t have a lot of precise speed info because nearly all its use is recreational and leisurely. My flea-weight (5’2”, 115#) wife and I did one 10k recreational race and averaged 5.7 mph (10:30 min/mile) in horrible conditions (30 kt headwind for the first half). We could have shaved several minutes off our time, but we throttled way back at the half-way point. We were far ahead of the nearest boat and I had paddled a 20k race the day before, so we took it easy. During relaxed paddling, my wife and I paddle in the low 5’s and when she stops paddling, I move the boat in the upper 4s. I have paddled it with a guy friend who is a decent paddler and we could maintain 6 mph but not much faster.
Wahoo FSK (18.5’ x 20.5”) – Race 6 mph (9:55 min/mile). All Day 5.7 mph (10:30 mile/min). Note that race pace is with raised Nelo rotating seat which helps with rotation. All day pace is with a comfortable foam seat. I have paddled this boat in many races, and find it competitive with the Epic 18x, V8 and Stellar 18R. It is a bit slower than the Epic V8 Pro. This was my first race boat and I have over 1200 miles in it.
Mystery (20’ x 20”) – Race 6.2 mph (9:40 min/mile). All Day 5.8 mph (10:20 min/mile). Note that race pace is with a raised Nelo rotating seat which helps with rotation. All day pace is with a comfortable foam seat. I have paddled this boat in three races, and find it faster than the Epic 18x, V8 and Stellar 18R, and competitive with the V10 Sport. It is slower than the Epic V10 and Stellar SEI. I have 170 miles in this boat and plan to race it in the Chattajack 31 in ten days.
Stellar SEI Surfski (20’ x 18.1”) – Race 6.3 mph (9:30 min/mile). Personal best on my standard eight-mile training course is 6.4 mph (10:24 min/mile). All Day is probably 6.0 mph (10 min/mile) but the hard surfski bucket is hard to sit in over three hours. This is my currently my preferred race boat, and I have numerous races and over 850 miles in it. It has shown itself competitive with the Epic V10, but slower than a well paddled elite surfski. You can tell when an elite ski is not well paddled, because the paddler will be swimming next to it. Those are easy to pass.
Stellar Rapid-S (20’3” x 16.3”) – This is the boat that I bought without a demo paddle and have always regretted it. At Race pace, I can paddle this boat 6.5 mph (9:15 min/mile) but only for 2-3 miles when using the standard ICF K1 style hard seat. After that, my legs fall asleep and my speed drops significantly. I replaced the hard seat with a foam seat which greatly improved comfort but restricts rotation a bit. With the foam seat, I can paddle 10-12 miles in relative comfort, but my speed is basically the same as the SEI above. This is not a boat that I will paddle more than a couple of hours. Due to the stability of this boat, I am unable to generate as much power as I can in the other boats. I also must be very careful where I paddle it because wave/wakes are very perilous, and it is impossible to wet re-board this boat. I have paddled the Rapid-S 375 miles but hope to replace it with either a Stellar SES or SEA soon. Both of those boats are traditional skis which can easily be re-boarded in deep water.
GPS SPEED DISCUSSION
A bit of a warning here, GPS is a subject where I have significant technical knowledge as a result of my work. IMHO, the number one error of consumer level GPS users is to overestimate the speed accuracy of GPS. GPS calculates speed by determining the distance travelled between fixes and dividing by time. At surski speeds (~7 mph), you cover approximately 10 feet per second, so to get a good 2 sec speed reading, the unit must be able to measure 20' distances accurately. The problem is, GPS is not nearly that accurate. According to the owner of the GPS system (US Government), consumer grade non-augmented GPS receivers have an accuracy of about +/- 16' (the spec accuracy is generally 30'). This means that when a GPS tries to give you a two second speed reading, it will measure the distance as 20' +/- 16', which results in a speed reading of 7 +/- 5.5 mph. GPS manufacturers recognize this limitation so virtually all add speed smoothing algorithms to their processors so that the user does not see that much scatter. These algorithms do a good job when operating at a steady speed but can produce some wild results with rapid speed or course changes.
I have lots of observations to back up the theory. Over the last twelve years, I have recorded and downloaded nearly every one of my workouts (run, paddle, bike, row, hike). When I go back and review the downloads, I find that approximately 20% show a “maximum speed” that is physically impossible for me. Another 40% of the events show maximum speeds that although physically possible, certainly exceed my effort that day. Regardless of what my GPS says, I can’t run as fast as Usain Bolt.
Very high end (Aviation and Professional Marine) GPS units are much better WRT speed for two reasons. First, they all employ augmentation (WAAS here in North America) which improves the average accuracy down to <1 meter. Secondly, they operate at a higher frequency. Most consumer level GPS receivers operate at 1 Hz, where high end units operate at 5 or 10 Hz. The operating frequency of a GPS receiver is more like the clock speed of a computer. A 1 Hz GPS calculates position once per second where a 10 Hz unit does it 10 times per second. (Do not confuse this with the refresh rate of the display.) So, if we redo the geeky math above for 7 mph with a high-end GPS, a two second distance measurement would be 20 +/- 6’ which equates to 7 +/- 2 mph.
So, what do GPS manufacturers say about the speed accuracy of their units? The answer is very little. Reviewing the owner’s manuals of the twelve GPS devices that I currently have, only two make any mention of the accuracy of their displayed speed. One is an older (non-augmented) driving GPS and it says that its speed should be “within one MPH at highway speeds.” The other is a newer augmented (WAAS) hiking GPS with a twelve-satellite receiver, which lists its speed accuracy as 0.22 MPH steady state. The implication of the latter spec is that the speed accuracy would be (much) lower in a dynamic environment involving acceleration/deceleration/turns.
The bottom line from this discussion is that the speed calculated by your GPS has little meaning in dynamic environments like surfing down a wave, sailing in a puff of wind or during turns. Once stabilized for 30+ seconds on a constant course at a constant speed, the speed displayed should stabilize and be reasonably accurate. This assumes of course, that the GPS is placed in a position that it has a good unobstructed hemispherical view of the sky. GPS signals are weak when they reach the earth from the satellite orbiting at 12550 miles, so it does not take much of an obstruction to block the signals from some satellites and significantly degrade your GPS accuracy.