If you’re an entrepreneur today looking to start a company, there’s a decent chance you’ve looked at an accelerator program.

The proposition is compelling: funding, access to resources, events, fellowship with other entrepenreurs, and introductions to venture capitalists who can fund your growth.

Given the popularity of accelerator programs, which are now popping up all over the world, Aziz Gilani, a director at VC firm DFJ Mercury, looked at 29 accelerators in North America as part of a study for the Kauffman Fellows program. His goal: evaluate them based on the number of exits they’ve produced.

His findings: put simply, “there were not enough exits to evaluate.” Only two of the 29 accelerators Gilani looked at, YCombinator and TechStars, had any exits of note. What’s more, nearly half (45%) of the accelerator programs Gilani evaluated didn’t produce a single graduate whose startup went on to raise venture capital financing.

Gilani’s study paints a bleak picture of the accelerator world. Despite the fact that these programs have become increasingly popular, and in some cases draw significant numbers of applicants, it appears that most of them are not helping entrepreneurs do any better than they would on their own.

Critics of the Gilani’s research point out that it can take years for exits to materialize, and in any case, exits are a poor metric for success. For many startups today, they say, building a profitable, self-sustaining business that doesn’t need funding or an exit is a real possibility.

So who is right? The truth probably lies somewhere in between.

On one hand, it probably is unfair to claim that startup accelerators are of no real value, even if the vast majority of graduates don’t raise funding and/or haven’t achieved an exit yet. Starting a business is inherently risky, and most new companies don’t succeed. Accelerators, not surprisingly, may not change that, but for founders looking for mentoring, fellowship with other entrepreneurs and access to investors, joining an accelerator program may be an attractive proposition.

On the other hand, given that none of those things guarantees success, and may do nothing to improve a startup’s odds, there’s a legitimate question as to whether entrepreneurs should be so enamored by these programs.

At the end of the day, building a company from the ground up is tough. To succeed, entrepreneurs must do a lot of things right, and there are numerous things they can’t control (like ‘timing’, or more generally, ‘luck’). Accelerators may or may not help entrepreneurs do more things right, but it’s worth considering that more than anything else, they’re vehicles for investors to more efficiently put their capital to work.

In other words, accelerators may do far more for the investor community than they do for entrepreneurs, raising the possibility that founders who really want to accelerate their new businesses would do well to consider whether the supposed intangibles of these programs are worth the price of admission.