You joke, but this is very similar to the Secretary Problem's optimal solution [0] [1]:

> The optimal cutoff tends to n/e as n increases, and the best applicant is selected with probability 1/e.

So, you deliberately reject a first cohort of applicants, and then pick the first person you encounter that's better than the best one you rejected. Obviously, in the joke that's not possible as they are neglecting to interview them, so I think the sibling comments are correct about it being a skewering of the hiring process, but it's neat that there's such a close parallel.

0: https://en.wikipedia.org/wiki/Secretary_problem 1: https://news.ycombinator.com/item?id=15567307

And recruiters and managers wonder why applicants spam hundreds of cookie cutter resumes and cover letters instead of painstakingly crafting each one. Or why they get annoyed at 'job portal' applications where we have to retype our resume into the various little text boxes the web application provides.

Its just a numbers game if you don't have a relationship with someone who can get you a new job.

> refuse to hire unlucky people

Not getting hired there seems to be the 'lucky' move. Arbitrary hiring guarantees, over time, some pretty terrible characters that a proper hiring process would filter out. If you're pissing away 50% of your potential and, say, there's a 20% chance of getting a 'good' person, now there's a 10% chance of that. Over time, a place with such hiring standards would most likely build out a corporate culture that's subpar. The rare 'good' hires would probably take off much earlier than need be because there's so many more 'bad' hires. Eventually the 'bad' hires rule the roost and then people act surprised at gross incompetence as the norm in many companies, shameless dishonesty, and various forms of harassment being normalized.

I worked at a place that didn't take hiring very seriously and this is exactly what ended up happening.

Except that isn't how statistics work. If you have 100 applications, 20% of which are good, and drop 50% of them, you get 50 applications, but you still have the 20% good application/bad application rate.

Of course, you still get 50% less total good applications, but ratios != total.

If there are 10 'good' hires in the initial pile, then cutting out 5 arbitrarily lowers your chance of getting any one of those 10 by half instantly.

This is what I was trying to express. At the end of the day there are only so many 'good' applicants.

Sure, but usually you don't want to hire _all_ of the good applicants - the reason you get bad hires is the false positive rate, not because there aren't enough total good hires in the pile.

It's a joke from the British version of the Office.

Poe's Law I suppose.

Perhaps, don't take it too literally. I suspect that comment must have been out of desperation, a satirical comment meant to criticize their woefully inadequate hiring process.

From what I've seen in academia, this is a surprisingly adequate process. I've seen them do phone interviews, Skype interviews, divinations, psychometrical testings, on-sites - and then none of their unicorns is right colour. The commitee declares a failed search and tries again the following semester.

You'd think that many, if not most of the people in the applicant pile could do an adequate job, perhaps with some training, considering what the job market looks like in the life sciences. Instead they play precious and then they have to make do with some last-minute lecturer, and that at times plays out really bad.

ha. except perhaps they're the unlucky ones because they chucked out a great engineers resume and the engineer is actually the lucky one ;)

or, "the lucky engineers had their resumes tossed in the trash so they didn't have to waist their time at that company"

that's what i was implying

Related: http://braythwayt.com/posterous/2014/10/04/i-dont-hire-unluc...

And HN discussion: https://news.ycombinator.com/item?id=3804134

Have you ever seen a pile of one thousand resumes that you have to go through?

I will give you a tip. You can't read them all. Don't even bother trying.

If you get a thousand resumes for a few jobs, then reading/analyzing/grading resumes are useless indicator of talent. You have to resort to some other process.