When one notices the same pattern repeating over and over again, it is worthwhile to change the conditions of the experiment to avoid repeating the same problem over and over again. I have at times commented on the curious nature of traffic on Thursday mornings, which often runs to two and even three hours in length for my commute , for reasons that are entirely incomprehensible personally. It is not a surprise, for example, that Monday and especially Friday would often be lighter in terms of traffic than the rest of the week, but why Thursday would be a more difficult day to drive than Tuesday is not something I know. I am definitely interested in hearing what reasons could be found, and am somewhat disinclined to speculate on the matter, seeing it as someone unprofitable to guess the intentions of hundreds of thousands of people and at least tens of thousands of drivers who are in my way, and in whose way I am as well. In the absence of being able to understand the reasons for the difference in days, one has to deal with.
In dealing with that reality, an experiment of sorts is in order, specifically, an experiment in terms of how early one has to leave to make the trip in a timely fashion. For the purposes of this discussion, a timely fashion would be less than an hour, since if there is going to be a sacrifice of sleep, and there will be for this to work, there will need to be some sort of payoff in better traffic. Given the patterns of traffic, it appears likely that there is a divergent pattern, where traffic traveling through certain choke points (the Interstate Bridge, Vista Ridge Tunnel, etc.) goes freely until a certain threshold is reached, at which point the slowdown would spread from those points onward. It is pretty clear that something like that happens given that the same spots are consistently slow when one travels, where there are no accidents or disabled cars in the way. Of course, figuring out exactly what happened is the job of transportation officials seeking to determine how to upgrade areas or provide alternate routes, and is beyond the scope of any such experimentation as I can undertake.
Such an experiment appears to be important for several reasons. For one, the specific problem is not a transient one; it is not going away any time soon. It has shown a consistent, if puzzling, pattern, and insanity is trying the same thing over and over again hoping to get a different response. Given the fixed nature of many of the constraints–the time I would need to be at work as a boundary, the fact that the starting and ending destinations are not subject to change for the forseeable future, and the fact that there are only a few limited routes to get where I am going, it appears obvious that the easiest factor to change is the departure time, as that is the only variable that is completely within my own control–the low fixed number of routes because of the scarcity of bridges over the Columbia River makes choosing routes of very limited utility in saving time, after all, since whether one drives down I-5 or courses a way through downtown, either way one ends up at the same bridge.
Of course, that means something fairly obvious, and that is that in order to leave at an early hour that I am going to have to get to bed earlier. There are some nights, of course, where it is going to be extremely difficult to get to bed early, seeing that the time I get home on Tuesday nights is likely to be late for the next month, at least. Aside from that, though, it will be worth seeing if avoiding the masses of people will allow me to have a more enjoyable commute experience, at least on the way to work, given the fact that at times the best one can do in a particular situation is cope with the reality that is, rather than actively change it to the way one would prefer it. As much as I would prefer to change reality, often coping is what is possible to do, so in that light, I suppose I have to resign myself to trying to go to bed early so that I am at least moderately alert for a forest animal when I wake up early the next morning. I suppose that ought to start tonight as well.
 See, for example: