We all know that we need to recruit “A players,” or as Mark Zuckerberg calls them in the case of Facebook, “100X programmers.” It has become a well-worn cliche that great employees drive exponentially more value than do merely good ones. But despite the massive delta between As and Bs, Silicon Valley seems content to ignore its own advice.

MIT Sloan School Center for Digital Business research fellow Michael Schrange wrote a series of Harvard Business Review (HBR) articles titled “Why Zuckerberg was Right” and “The Coming Collapse of Average” confirming the existence of such exponentially effective stars and quantifying the impact their presence (or absence) can have on an organization.

Everyone from Schrange, to best-selling “Good to Great” author Jim Collins, to the VCs that run Silicon Valley knows that people are the key to building great organizations. So why does the Valley operate as if the opposite were true?

The startup ecosystem spreads too much money across too many mediocre companies, thereby diluting the scant A player talent pool. VCs are then loathe to kill deals that aren’t working, propping up the losers and further restricting talent from re-entering the marketplace. Companies, for their part, raise as much money as possible as quick as possible and hire rapidly, seemingly ignoring the need to identify elite talent.

Some will argue that the shortage of A players dictates the hiring of Bs and Cs. They’ll say that time is the most precious resource, and that they cannot afford to wait to find winners. This is foolish, given the downstream effects of having mediocre employees deliver mediocre work – be it ineloquent code, uninspired strategy, or, worst of all, poor hiring decisions.

Schrange writes:

What 100X teams and talents have, is either the innate or acquired ability to make their skill sets and sensibilities multiply in effectiveness rather than merely add up…Great coders [as one example] don’t just write great code, they grasp the essence of problems in ways that makes their code architectures clean, accessible and maintainable. The code not only works, other coders — perhaps less talented — can safely and creatively build on it, with it and for it. Their code costs less to support and evolve even as it invites and facilitates new value creation.

In other words, employing mediocre talent is actually counter-productive. The principle extends beyond coding to every role within an organization. Thinking about it in terms of Pareto’s 80/20 principle, if 20 percent of a team are A players and do 80 percent of the work, then for the rank-and-file to keep pace with a 10 percent improvement by the A players, the B and C players need to improve by 40 percent. Not likely to happen.

The conclusion then is that hiring B players, and worse C and D players, should be avoided at all costs. Silicon Valley would be better served to contract in the breadth of the companies it creates, as a means of increasing the “depth,” or quality of its output.

“There’s arguably never been a worse time to [employ a] mediocre, average, or typical employee,” writes Schrage. “Mediocrity is a cost to be managed and a burden to be borne, not a potential to cultivate or a resource worthy of serious investment.”