Sorry I'm Late

No, you're not. If you were actually sorry, you wouldn't be late next time. And you will be.

I hate that phrase, "sorry I'm late." I have to go to so many meetings. And I can tell you in advance who is going to be late.

So, as part of the service offered here at The End, five (as they say) "True Facts" about people and time:

Five True Facts About People and Time

1. The busier the person, the less likely they are to be late. Busy people are more likely to be good time managers. This may be because they have to manage a scarce resource, and so develop institutional rules for governing their day. For example, when I was junior. I never kept a calendar, and just relied on "memory" for meetings. What that means is that I was usually late, and often just missed stuff completely. By comparison, now I am almost always on time. But I have an assistant and an electronic calendar, and I live by her rules and its dictates. There is a more sinister interpretation, however. Competent people adjust; if you are capable of getting better at your job, you will be given progressively more and more responsibility. And that includes managing your own time and respecting that of others. Incompetent people like to BELIEVE they are busy, because it gives them an excuse for sucking. And, they are always late. (A caveat: TRULY busy people, like Presidents or Provosts, are often late, usually because their previous meeting started late, probably because some shmoe came in and said, "Oh, sorry I'm late." GRRRRRR.)

2. The closer the person's office is to the meeting room, the more likely they are to be late to the meeting. To be fair, this is understandable, and perhaps unavoidable. But pay attention, and you'll see it's true. If someone is coming from another city, the importance of the meeting is elevated in their mind. For someone who has to drive, likewise. Even if you just have to walk to another building, you are more likely to be on time. But for the person in the building, and especially for someone right across the hall...well, they are usually late. They may check in once, notice that several people aren't there yet, and then go make a phone call. Then, when in fact they are the last one to come in, five minutes late, they say (you guessed it), "Oh, sorry I'm late. I was just making a phone call." As if that mattered.

3. Small meetings can have convergence problems, just like maximum likelihood estimations. Someone shows up, puts down their briefcase and papers, sees no one else in the room, and goes to get some coffee. Because they are three minutes early, they stop to chat to a colleague or a staff person. They end up being five minutes late, but their briefcase sits there in mute proxy, evidence they were in fact early but the meeting failed to converge. I have seen meetings start fifteen minutes late, when one at a time person after person says, "Oh, Smith isn't here yet. I'm going to get a soda. Anyone else want something?" and off they go. Smith comes back, and Jones decides to run get a book she left in her office. "We can't start 'til Mbuto gets back; I'll just be gone a minute." (Honestly, I am guilty of this, all the time. Mea culpa).

4. If you are always late, it is not an accident. Random things happen. Being late, if it were random, would probably be negatively autocorrelated, an (AR[-1]) process. That is, if I am late once because I got caught behind a school bus, I will be early next time because I found it embarrassing to be late. But late people are ALWAYS late, so it is more like an idiosyncratic autoregressive time series, (AR_i[1]). If you find you are always late, there is a feasible solution: LEAVE EARLIER. If you choose not to do this, you are deciding to insult your colleagues in a show of passive-aggressive vanity. "Oh, sorry I'm late. You wouldn't BELIEVE what happened this time." Actually, that's right. I DON'T believe it. You are just a pinhead. Chronically late people always time it so that if they catch every light and find a parking space in front of the door of building, they would be 30 seconds late. But "something goes wrong" (surprise), and it is the fault of the thing, not the person's own egregious solipsism.

5. Lateness is ingrained, as a social convention. When first I got to Duke, and became Department Chair, I tried to start our meetings on time. There had evolved a convention that meetings and talks started seven minutes late. It was a translation of origin problem: 1:07 was actually the origin, 1:00, for a talk advertised to start at 1 pm. I pointed out that this was arbitrary, and therefore could be changed. Just move the intercept 0, instead of 7 minutes, and we'll start on time. For a while, I would introduce the speaker at 1 pm, to an empty room, and people would come in late, whispering "Oh, sorry I'm late!" Then, a few people would show up on time, but I would wait until more showed up so that we didn't (1) embarrass the speaker by introducing her to an empty room, or (2) interrupt her with late-arriving "sorry"-whisperers. I gave up, within a month. Now, people come in at X:07 for an X o'clock talk, and there are always several who come in late. "Oh, I'm sorry..." AAAARGGGGHHHHH.

BONUS: There is an easy test for finding out if someone is chronically late. You don't need experience to find out, don't need meetings. Just look at their watch. If they set their watch more than five minutes "fast" (i.e., ahead), they are a late-nik. I find it remarkable that someone would set their watch, ON PURPOSE, to a time different than the correct time. (I set mine according to the NIST Clock, to the nearest second, once a week). As "Doc Potter" says:

Well, first off, set your clock to the CORRECT time - always. Setting it ahead is a trap. It actually encourages you to be late because you look at the time and then tell yourself that your really have more time, so you slow down and procrastinate even more.

The "set watch ahead" trick would only work if it were done WITHOUT YOUR KNOWLEDGE. Just be on time....

UPDATE: An astute reader (anonymous by choice; probably some late-ster) points to this piece on the claim that "Punctuality is Inefficient." Well, that's obviously wrong. He wants to argue that punctuality is, in fact, Pareto optimal. But then he wants to claim that being late / being late (for two people) is the Nash Equilibrium. Then, he bails on that, admitting that in fact the game has no equilibrium. Nonetheless, the good Andrew Chamberlain does make a good point: at its base, the problem is that I dislike MY waiting much more than I feel bad about YOUR waiting. You feel the same, only in reverse. But, of course, manners and conventions are ABOUT solving this sort of problem.

UPDATE II: MR's Alex Tabarrok summarizes some interesting research. An interesting Nash Equilibrium argument. But hard to buy, since it can just unravel. If I expect you to arrive 10 minutes late, I show up 12 minutes late, and so on. Not at all clear that there are just two strategies, show up on time or show up 10 minutes late. But if there are just two strategies, then it is true that there are two equilibria (though only one is subgame perfect).