Sometimes statistics are intuitive. Somethings they are not. Most of us make inherent assumptions because we like simple models, direct ratios, obvious correlations and easily understood comparisons. That’s not a put-down, for most of the history of the human race simple models were perfectly adequate. We are all hard-wired to see things that way because it works. For those of us that never expect to get anywhere near the speed of light, Newtonian physics works just fine. And simple statistical comparisons work in statistics a lot of the time too, but sometimes they don’t. Take this example:
The TSA guy who put this together is providing current numbers but they have provided same day of the week last year comparisons in an effort to give a perspective of what is going on. And for the purposes of butts in seats the numbers are informative and helpful for getting the trend but it is a little deceptive for most people in telling the rate of change in the trend. The reason for this is the variability of the numbers in the left column are not independent. That sort of hides both how bad this got and how quickly things are improving.
The non intuitive part of this is that most (not all, but most) airline trips are not independent trips, they are part of a trip pair. What that means is that at the nadir of these numbers when they were in the 90,000 range, a few weeks ago, only perhaps HALF of the pax being counted were INITIATING a trip voluntarily, the others were simply trying to get home from a trip they had previously taken. So back then the actual number of people doing voluntary flying was only maybe 45,000, the rest merely trying to get home.
The same, of course, holds true for the figures on the right. If a year and a half ago 2.5 million people were going through the turnstiles, only 1.25 million or so were INITIATING their trip, as many, or very nearly so, were simply coming home from a trip they had already initiated. EVERY number in each column strongly influences the numbers in the column above it and is strongly influenced by the numbers below it. And in a steady state, that really isn’t all that deceptive, but when the numbers are changing it gives the appearance of distorting the trend rate. The reality is that we went from a nadir of perhaps 45,000 people a day initiating trips and our target to getting back to normal is only about 1.25 million people a day initiating trips. The rest of the traffic we get quasi-automatically from people coming home.
So the bad news is that what we are really looking for - the number of people voluntarily initiating travel - actually got much worse than it at first appeared. The good news is it’s coming back faster than it appears. What we REALLY need is for 1.25 million people a day to initiate travel. Right now that number is probably about 100,000. But it’s going up quicker than it appears and - as long as the increases are sustained - don’t have nearly as far to go as it first appears.
Not blowing smoke, not saying it’s all going to be rainbows and unicorns, but it probably isn’t as bad as it seems to many people.