All of the replies given so far are really good, though for some, I think more applicable to advanced big-water crossings than simple paddling on average lakes and rivers. I infer something in your description of the situation that seems to suggest that the heart of the problem might be your tendency to “zone out”, and I’m not sure what the best solution to that might be. If that’s what the problem is, though, I’m VERY familiar with it. For years I had a regular paddling partner who simply did not focus on technique or staying on course. She mostly tended to veer left. If she was on my right, invariably she’d come crowding right alongside until I needed to get out of the way somehow. If she was on my left, she’d veer off unrestrained in the wrong direction while remaining completely unaware, and in situations where safety and keeping in close contact was an issue, she’d sometimes even complain that I was getting too far away when in fact she was just plain going in the wrong direction! Some people just aren’t wired to focus on details and there’s probably nothing that can be done to substantially change that. However, if any of this seems to match what’s going on with you, all I can suggest is that you learn to recognize that this is how your brain works and use that as motivation to try to remain more connected to all your visual cues.
You also mentioned that paddling toward a landmark sometimes doesn’t work, and specifically you mentioned that this might happen when the weather changes. I’m not sure how to interpret that. If paddling toward a landmark doesn’t work in calm conditions, all I can guess is that the topic of the preceding paragraph applies. If something such as windy conditions ruins the reliability of your use of distant “aiming points”, that’s perfectly normal, and I have some suggestions.
When paddling in windy conditions, take every opportunity to use a PAIR of landmarks to keep track of your course. This won’t always be possible, but watch for these opportunities and use them. For example, if you want to aim at that house on the distant shore, keep track of how that house lines up with some landmark behind it, such as an identifiable spot on a hill in the background. If you are aiming at a tree and all that can be seen are trees, pick another tree behind it. They don’t have to “visually match” on your sight line, just keep them aligned via an imaginary vertical line that connects them. Just be aware that the closer the pair of objects are to each other, the less sensitive this method will be for it’s intended purpose (see below). If you happen to be paddling toward a peninsula, this is a perfect chance to practice. Paddle toward the point of the peninsula while noting a distance landmark on the far shore behind it which aligns with the tip of the peninsula. In any situation like these, if there’s a crosswind, you will quickly be able to detect how your actual course becomes diagonally “off” from where the boat is actually pointed because the pair of landmarks will not both stay lined up with each other. Make adjustments to your heading to bring the two objects back in alignment, and then keep adjusting so that there is no “drift” of the two objects relative to each other. In other words, when your corrected heading is proper, the two objects remain lined up with each other even though your boat is not pointed directly at them. The fact that the two objects stay lined up with each other tells you that your actual course is directly toward them. But that’s not the end of my recommendations about that.
When you are paddling in situations where you can line up two distant objects along your intended path, not only can you navigate that way, but you can become accustomed to how much you need to correct your heading to account for various kinds of crosswinds. I use canoes and rowboats, which have “a lot more boat” above the waterline than a kayak, and usually, “less boat” below the waterline, which of course magnifies the tendency for being pushed sideways so that my actual course becomes substantially “off” from the direction the boat is pointed. This will be less of a problem with a kayak, but I’m just pointing out that this issue of drifting off course is a very real thing in my boating world. One thing I do is try to pay attention to these visual cues using paired landmarks that tell me how much I’m straying off of the line matching my intended path, even when it doesn’t really matter. Anytime you can line up two objects that way, it’s just an opportunity to practice. Make it a point to do this even using landmarks that are bad for long-term use but temporarily available (like two trees, which you can’t keep paddling toward for more than a short while, or the alignment of a particular location on a boat dock with a distant mooring buoy), and just pay attention to those two landmarks for the short time that you are able to use them. In my own case, as a result of this kind of practice, if I’m paddling or rowing across a big lake in strong wind and all I have is a single landmark on a distant shore to aim at (nothing in the background to help me identify sideways drift), I already have a pretty good idea how much I need to “aim off course” in order to “stay on course”.
And here’s a final hint about correcting for crosswinds. If you don’t have a pair of landmarks on the sight line that matches your intended path, and you are not absolutely sure how much to correct for the crosswind, make the correction bigger than you think it needs to be. Make it a LOT bigger. There are two reasons. First, it’s more likely that you will underestimate the effect of the wind than to over-estimate, so overcompensating from the get-go simply makes sense. Second, if you really do overcompensate and your actual course ends up taking you in a direction that is toward the upwind side of your intended path, it’s far easier to gradually re-adjust in the downwind direction than to do the opposite. Yes, if you over-correct for the wind along most of your route, you will be wasting a bit of energy the whole way, but you will partly make up for that when re-correcting in the downwind direction, and overall it’s not as bad as having to aim even more sharply into the wind to re-correct when you are closer to the endpoint of your crossing, which will be the case if you under-correct for most of the way. Usually, of course, it’s not that big of a deal to screw up your estimate for accounting for crosswind when going across a lake, but becoming more aware of how to detect and correct for sideways drift can be kind of fun, besides being a really useful skill sometimes.
