Am I just lucky?

I get surprised at how often I bump into singular times when I check my phone. (By "singular" I mean number configurations that respond to some simple pattern.)

Suppose we agree to consider singular those times with the following patterns (in a 24-h time display): AB:00, AB:BA and AB:AB (A and B may be equal). Are the odds in favor or against?

Estimate the probability of unintentionally hitting at least one of these singular times when you check your phone during the day.


Details and assumptions

  • Consider only the period from 10:00 to 23:59, both included (840 minutes).
  • Suppose you check the time some 30 times at random during that period.
  • Every minute in the period has the same chances of being picked.
0.3 \approx 0.3 0.7 \approx 0.7 0.9 \approx 0.9 0.5 \approx 0.5 0.1 \approx 0.1

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

1 solution

Gabriel Chacón
Dec 2, 2018

The purpose of the problem is to show that many coincidences are not that rare when you take a close look at them.

First, let's count how many "singular" times we have in the period. Take the hour 15, for example. There are 3 singular times: 15:00, 15:15 and 15:51. Let's suppose this also happens for every hour in the 14-hour period (3 x 14 = 42). There are exceptions we have to deduct, though – hours that have only 2 of these 3 possible configurations: 11 and 22 (permutating A and B makes no changes) and 16 through 19 (where B > 5 B>5 , there is 19:00 and 19:19 but not 19:91). The total of singular times is then 3 × 14 6 = 36 3\times 14-6=\textbf{36} .

One way of estimating the desired probability is to think of the problem as equivalent to picking 30 different times at random from a set of 840. The probability of having at least one singular time among the 30 picked is the complementary probability of not having any:

p = 1 ( 840 36 30 ) ( 840 30 ) 0.74 p=1-\dfrac{{840-36}\choose 30}{840\choose 30}\approx \boxed{0.74}

Another way is to see that there is a total of 36 {36} favorable cases out of 840 {840} possible ones. The probability of encountering any of those times when we check the phone once is P = 36 840 = 3 70 P=\frac{36}{840}=\frac{3}{70} . The probability of encountering at least one of such times in 30 checks (supposing times can be picked more than once) is:

p = 1 ( 1 P ) 30 0.73 p=1-(1-P)^{30} \approx \boxed{0.73}

This probability is surprisingly high indeed! The odds were clearly in my favor after all.


  • Notice the second value is very similar to the first. Allowing the times to be picked more than once to make the events independent does not make much of a difference since 30 840 30 \ll 840 .

I have still a lot to learn about probability...

There is (about) 3/4 of picking at least one "singular" time configuration in a daily schedule of consulting the time! Well, 1 or 2 in 30 are not so many...

Congrats for such an ingenious problem and the clarity in exposing the solution!

Félix Pérez Haoñie - 2 years, 3 months ago

Log in to reply

Thank you, Felix!

Gabriel Chacón - 2 years, 3 months ago

Log in to reply

You are welcome. I'll try some others from you but your level seems too high for me.

Félix Pérez Haoñie - 2 years, 3 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...