Atheist Round Winners [Turing 2013]

You may want to keep this year’s answer key and list of entries in the atheist round handy while you read the results.

Atheists earned their reputations as skeptics in this round, as their scores were much lower than the Christians gave in their round.  By the crudest metric — percent of entries with more than a 50% pass (Very Likely Atheist + Likely Atheist) — only six entries out of eleven hit the mark among the atheists.  For the Christians, only three entries didn’t hit this benchmark.  And, after all, you’d expect a good number of entries to read as what they’re written as.

So the atheists were more choosy, but whom did they choose?

The entrant with the highest percentage of atheists rating him ‘Very Likely Atheist’ was the very Catholic Gilbert of The Last Conformer (entry 10, who netted 42% of atheist readers, and 74% rating him as more likely atheist than Christian.  Personally, I’m wondering how many votes his ‘System 1/System 2‘ reference netted him.  Brendan Hodge of DarwinCatholic (entry 1) was the next runner up, with 36% of atheists judging him as very likely atheist and 70% rating him likely or very likely atheist.

And then the real atheists all appeared in a block.  The next five placements went to all five atheists, with Chana Messinger (The Merely Real) taking the lead as the real atheist most likely to be recognized as such (entry 5, 28% very likely atheist).  Right behind her was Chris Hallquist of The Uncredible Hallq (entry 2) with 27% of atheists giving him the most plausible rating.

After these two, very likely atheist ratings slipped below 25% (i.e. lower than if people were voting at random).  But all the rest of the Christian entries were trapped at the bottom, as you can see below.

And the Miss Intellectual Congeniality Award goes to Chana Messinger, who was far and away the contestant whom non-atheist voters most wanted to have a chatty coffee with.

In fact, she scored a hat trick: Most Atheist Atheist, Miss Intellectual Congeniality (among non-atheists) and Miss Intellectual Congeniality (among atheists).  Fifty percent of atheists judges wanted to sit down and chat with her, and, if she wasn’t available, their alternate partners were Gilbert (47%) and Brendan (44%).

Christian Round Winners [Turing 2013]
[Turing 2013] Atheist Entry #11
[Turing 2013] Answer Key
[Turing 2013] Christian Round Entries Index
About Leah Libresco

Leah Anthony Libresco graduated from Yale in 2011. She works as a statistician for a school in Washington D.C. by day, and by night writes for Patheos about theology, philosophy, and math at She was received into the Catholic Church in November 2012."

  • butterfly5906

    Can you break down the results by the religious identification of the voters? Did the atheist voters do a better job of finding the real atheists than the Christian voters?

    • LeahLibresco

      This graph only includes atheist voters. I can contrast them with Christians for this round, but I can’t in the previous round, as the number of atheists voting was too small.

      • unfated

        Can you give us a total for how many Atheists and Christians each voted? Did that number change between the Christian and Atheist rounds?

        I was honestly getting a little fatigued at the end, and am not completely sure I voted on every post in the second round. If you repeat the test again – which I hope you do – perhaps cap it at 8 to 10 total entries.

      • Brendan Hodge

        Yeah, the size of the sample we’re looking at for both Christian and Atheists results would be nice to know. (And of course, you’d be disappointed in us if we didn’t ask.)

  • Martha O’Keeffe

    Congratulations, Gilbert! Should I now be denouncing you to Rome for heterodoxy? :-)

  • unfated

    The thing that turned me away on Gilbert was his mention that he shouldn’t use “scientific paper” as a bonus question. That seemed to me like the equivalent of someone talking about “the weird stuff that nerds like” and it just seemed to forced to be real.

    Also, I’m not entirely sure I marked anyone as a very likely anything – perhaps just one or two out of all of the entries. In retrospect after seeing how Leah presented the results, I think I should’ve committed more fully to both some of the Christian and Atheist entries. I wonder if more of the atheist voters felt noncommittal about choosing “very likely” candidates in the same way.

    • Slow Learner

      Yes – most of my answers on both rounds were “Likely” rather than “Very Likely”.

    • Alexander Stanislav

      I also did not put almost any “Very Likely’s” except for the ones I thought were fakes.

    • Christian H

      Well, I’m Christian and I rated almost no one as “very likely” Christian or atheist–perhaps two or three people across both rounds. So it’s not just atheists who do that.

    • Gilbert

      Across both rounds, I voted “very likely Christian” five times and “very likely atheist” twice.
      So my variant hypothesis: Atheists actually gave less evidence either way.

      My guess would be this is because most of the signal was in the optional question, which most atheists didn’t answer. In fact that would have been a fairly good heuristic by itself, unless I miscounted only 1.5 atheists (Guy and Joy) answered the optional questions and only 2 Christians (Elliot and Beth) didn’t.

      Since each side is presumably better in detecting their own tells, this means atheists actually had less data to go on and were rightly more skeptical.

      The lesson for next time would be not to have optional questions, but then that might not have been possible, because making the third question compulsory might have lead to recruiting problems and removing it entirely would have made the test really boring.

      [Edit: I actually did miscount. The correction I made doesn't affect the main point.]

  • Slow Learner

    Wowzers – only just about 50% of atheists agreed I am an atheist, and that’s combining Very Likely and Likely. (My entry was A9).
    I’m a little disappointed that I am apparently better at lying than I am at telling the truth!

    • Martha O’Keeffe

      I don’t know if it’s any consolation that this non-atheist pegged you as an atheist, with good arguments, especially since I too was fooled by Gilbert :-)

  • Tom

    I only answered “Likely x” for anything, but then this was my first round and I’m a terminal bet hedger.

  • avalpert

    Looking at the results from both rounds and the commentary throughout I’m pretty sure this process says more about the voters and the nature of how people evaluate a transparent suggestion that someone may be lying than it does the actual writers and their ability to project the other sides views.

  • Brendan Hodge

    I too tended towards Likely rather than Very Likely: on the Christian round I had one “Very Likely Atheist” vote (C1 – Ozy Frantz) and one Very Likely Christian vote (C7 – Elliot). On the Atheist round, I had two Very Likely Christian votes (A7 – Elliot, who I thought I recognized as Elliot, and A8 – Ink) but no Very Likely Atheist votes.

    Having looked at this kind of survey data at work, I always know that I should just go ahead and be definite so as to produce better data, but I can never quite conquer my fear of being wrong so I mostly register mild opinions.

  • stanz2reason

    Maybe 8 entries (4 Atheist + 4 Christian) would be better as I’ll also definitely admit a fatigue of sorts after reading through 2 dozen essays.

    In order to keep people from hedging their bets as someone noted below, you might also require (not sure how… maybe suggest) people spread their votes (2 x ‘very likely’ Christian, 2 x ‘likely’ Christian, 2 x ‘very likely’ Atheist, 2 x ‘likely Atheist) and having 8 entries allows that to happen evenly.

    Maybe releasing all the Christian ones & all the Atheist ones at once might be good too. I’m curious how the # of votes each entry got corresponded to when it was released. I guess the raw data will offer the best suggestions.

  • Brendan Hodge