It’s inevitable that successive classes of startup accelerators will be compared, and so, with this week’s demo day for the second class of Flashpoint, Georgia Tech’s startup accelerator, I’ve heard countless comparisons to the first class, which I was a part of. So how do you compare different classes from a startup accelerator?

Comparing startup accelerator classes can be as misleading as comparing children. Comparing a class of 9 month old startups to a class of 3 month old startups is as absurd as comparing 9 month old and 3 month old babies. You may think you remember what those 9 month olds were like, but without photos, video, and careful note taking, the comparisons are likely to be wrong. Further, Survivorship Bias means we’re likely to recall the successful survivors of each accelerator class while forgetting the failures.

So what empirical criteria exist to evaluate startup accelerator classes? I would suggest that initial quality of companies entering a class, viability of graduating companies after 6-12 months, and funding outcomes are all plausible metrics for evaluating startup accelerator classes.

Incoming Company Reputation – How hot are the companies even before they start the accelerator? This can be an indicator of the reputation of the accelerator. Any new accelerator will do well to cherry-pick a bit on their first class and pick some known commodities. Success takes time. Picking companies closer to success helps the early reputation of the accelerator. Our first class had local brand-names in Pindrop Security, CodeGuard, RideCell, and SportsCrunch, who had all raised real-deal rounds. The new class’ only known company was We & Co, with some funding in the bank, and a west coast swagger that confused some locals.

Founder Pedigree – I’ll naively examine this from a perspective of previous startup exits. The first Flashpoint class had Pindrop’s Paul Judge as Chairman and a couple of founders with 8-figure exits in their past. This second class has been impressive. 3 companies have founders who have built $100M+ companies, and at least one other company has founders with an 8-figure exit. Flashpoint attracted more proven founders in its’ 2nd class, which is a good thing for everyone involved. Flashpoint attracted more proven founders in its’ 2nd class, which is a good thing for everyone involved.

Survival – Long term viability is not comparable for a newly graduated accelerator class. Few will mention it, but 3 companies from Flashpoint’s 1st class have effectively folded (moving on from an unviable business is a good thing).

Fundability & Exits – 7 companies from the 1st class have more funding behind them (), and 6 more companies are still active without new funding. 6 months from now, we’ll see how this fresh class has done, even though the thriving survivors from the first class will be even more advanced by then, and presumably some of the new grads will have moved on.

Atlanta can be a tough town for young startups. Few graduates from either Flashpoint class have their entire business sorted out. If they did, few of them would be asking for investor money. If you compare 6 month graduates to fresh graduates, the older class looks better. Comparing incoming momentum, my Flashpoint class had a certain shimmer, but the fresh crop has a nice track record. Over the long term, we’ll all help each other succeed.

Some people were harsh on the new graduates. The hard truth for Atlanta critics is that people who have seen Flashpoint demo days AND demo days of other cities have said that our startups are much BETTER both in pitch and in company quality than other demo days (with the notable possible exceptions of TechStars, 500 Startups, and YCombinator). I’m encouraged by the current Flashpoint class teams that are carving out their funding rounds with messages refined in the fire of Flashpoint. I’m discouraged by teams that aren’t seeing as much traction. Most of these teams are chasing big markets with plausible plans hatched from real conversations with real customers, and the willingness to admit when their assumptions are wrong and adjust course. They’re good startups working hard to become good companies. It may take more cycles of creating companies, selling, and reinvesting, but Atlanta should be finding ways for these companies to thrive.

A pivot is a change in strategy without a change in vision. You cannot have a pivot without vision (that’s just wandering around).

— Eric Ries (@ericries) July 6, 2012

For web startups, the “pivot” has become a nearly universally applauded badge of honor. There is much to be said for having the boldness to admit you’re wrong about some aspects of your startup and adjust course. Far too many startups have failed by following the same failing strategy into the ground. I’ll use the term “pivot” loosely, even allowing for a change in vision, but a pivot into an area where a team has no vision is a recipe for failure.

The pivot-madness has grown so strong that the pivot-backlash has begun. The Harvard Business Review slams a YCombinator founder for “Too Many Pivots, Too Little Passion“, but they’ve read the situation exactly wrong. They present the story of Brandon Ballinger, who vacillated between 5 different ideas before settling in on one for YCombinator’s demo day. Startup founders ought to be people of passion – perhaps even people of many diverse passions. If a YCombinator founder explores several markets they’re interested in before identifying one that shows some traction, that’s not a bad thing. In fact, it’s far BETTER than picking one of their passions, chasing it for 12 weeks, and hoping for the best. Having multiple passions doesn’t indicate a lack of passion.

In my time at Georgia Tech’s Flashpoint Startup Accelerator, we saw a number of pivots in our class. We had a company pivot from helping people buy cars to resolving family conflicts, from digital publishing to corporate newsletters, from deals for nerds to deals for churches, and many more. All pivots were rooted in Customer Development, and showed some degree of market traction. In the long run, the success of these teams seems to have been more rooted in passionate the team was about their validated business than about the degree of traction.

My company, Badgy, was pressured heavily and repeatedly to pivot. My passion for Badgy comes from previous work in social networks and social games. While many mentors assumed my strongest passion was for social apps, I knew when they pressured me toward building social analytics or social conversation management solutions, that just wasn’t my passion. My passion is ultimately in a fascination with human behavior and motivation, with finding ways for people to find unexpected joy in activities they haven’t seen or done before. We pivoted to broaden our channels. We pivoted our value proposition from “engagement” to increasing distribution of social content, and eventually to driving sales – all of this still perfectly aligned with my passion around human behavior.

Much ado has been made of YCombinator’s invitation to accept companies with no idea. While this may seem insane at first, and it has been defended as a bet on the team, I believe it’s deeper than that. I believe they are looking for talented teams with things they are passionate about. Vision comes from passion, and I believe that every talented team with things they are passionate about can find a way to build those passions into a high growth startup if they are willing to throw away their assumptions and find the path to connect that passion to a market.

A current Flashpoint company I know, We&Co, is passionate about the service industry, restaurant servers, bartenders, retail workers, etc, and they’ve masterfully experimented in that space to find a growth market. Another killer Atlanta startup, Mowgli Games, is passionate about democratizing the creation of creative content. Although this started as a social game to create music that would monetize their users, they’ve moved beyond initial assumptions and have transcended the mundane label of “social game”.

So if you’re thinking about a “pivot”, take a step back. Figure out what pieces of your startup you are passionate about. “The whole thing” is not an option. “I’m passionate about doing a startup” is the worst option – startup-fever is not a passion, it’s just going along with the crowd. In basketball, once you stop dribbling, you have to keep one foot planted and you can move the other foot around to find an angle at a shot or a pass. If you don’t have any feet planted, you’re just spinning in a circle and looking silly. That’s what pivoting without passion looks like. It looks pointless, and it looks ineffective, because it is. First, figure out what you’re passionate about, then iterate around that passion to discover a business.