I've been inquiring about lighting safety in general, and the more I read the less certain I am about anything.
A: While boating incidents are a big fraction of lightning deaths, and deaths of fishers are over twice as great as of beach bums, there's a confounding factor: Fishermen spend more time on the water, and are not dismayed by cloudy weather.
B: The actual ground strike density is fairly low. Alberta Forestry has a lightning detection system that is about 90% efficient at detecting ground strikes, and plotting them to an error of about 1/2 km. Peak strike densities run about 4 per km2 per year.
C: Strike density is not what you would expect. The foothills an hour's drive from the mountains have over twice the strike density as the mountains themselves.
D: Height isn't as big a win as you think. All other things being equal the protective cone is only about 30 degrees wide: That is, a 100 foot high tree will have few strikes within 50 feet of the base. A thousand foot high cliff will have few strikes within 500 feet of the base. In mountains, average slopes aren't that steep. (I'm trying to get data from Alberta's detection network to see if there is strike clustering on ridges compared to passes.)
E: In terms of risk the U.S. typically has 15-20 lightning deaths a year. For comparison some 7000 people a year die slipping on the soap in their bath. *
(As a commenter points out, most people bathe more frequently than they experience thunderstorms. Absolute risk: chance of an event happening. Relative risk: chance of an event happening with other constraints. The absolute risk of bathing is much lower than the absolute risk of thunderstorms. The risk per bath or per storm shows bathing safer. If you compared risk per bath vs per audible strike, it gets murky.
F: The shoreline may not be safer. Given the partial attraction to taller objects, the fringe of trees around a lake may get more strikes than the lake itself. So you want an even spread of lower trees a few hundred meters away from a stand of taller trees. This discussion on reddit's askscience https://www.reddit.com/r/askscience/comments/1mm794/when_lightning_hits_a_large_body_of_water_how_far/ suggests that strikes have to be very close (20-50 feet) to have a dangerous effect. You may be safer paddling near the shore rather than going onto the shore.
G: Consider the following scenario: You're out for a day paddle, and are about an hour away from your cottage. Storm comes up, lightning crashes, and icy rain is pouring out of the sky. You go to shore, but you are still wet, and rain keeps you wet, and the wind's evaporation cools you more, and now not moving and not generating heat. You cool off. You start to shiver. So the question then is: Does being on shore reduce the risk of harm by lightning more than it increases the risk of serious hypothermia? In the outdoors we often have to make this decision of relative risk of alternative actions.
H: I have been caught by thunderstorms canoeing, and have gotten off the lake post haste. Associated with these storms are often very strong winds. I consider the wind to be far more dangerous than the lightning.
This does not mean ignore thunderstorms, but overall I consider the risk (Probability of death or serious harm) to be inflated. Does not mean that you shouldn't avoid the hazard when reasonable to do so. LOTS of things in outdoor pursuits have an element of danger. The risks however are as much from the weather in the rest of the storm as the lighting -- wind can capsize your canoe, drive you onto a lee shore, or blow you far from shore to where the water is really rough. Rain can be blinding. The combination cold enough to induce hypothermia.
One time on Reindeer Lake in Saskatchewan, we saw a squall line coming at us. We made for shore and beached in dead calm, unloaded the canoes, and started to set up camp. We brought the canoes up on shore. 10 minutes after we landed we had 3 foot breakers pounding the rocks. Our camp was chosen in a hurry, and was a ridge of rock about 50 feet wide, and 8 feet off the water. It was a spit sticking out then paralleling the lake. The wind picked up one of canoes and carried over the spit into the water behind. This was a 26 foot voyageur canoe weighing some 280 lbs.
Overall if the choice was to go out in stormy weather or sit inside waiting for your next heart attack, I suspect that sitting inside is the higher risk activity.