Without evidence of benefit, an intervention should not be presumed to be beneficial or safe.

- Rogue Medic

Happy Full Moon Friday the 13th

Technically, the full moon is not until 00:33 – 33 minutes after the end of Friday the 13th, so that may help the superstitious to feel better, since these superstition events are not actually coinciding – pitting twice as many Gods against the superstitious (a double whammy). Or the superstitious may feel worse, because they now have two days in a row of the Gods conspiring against them. The reality is that only their own beliefs conspire against them. it is all in the heads of the believers.

Even when someone does claim to come up with some evidence to support their beliefs, those conclusions are not supported by higher quality research.

In conclusion, Friday the 13th appears to be dangerous for some women. Since Friday falls on the 13th day of the month only twice a year on average, prospects for significant public health gains are limited. However, the risk of death for women who venture into traffic on this unlucky day is higher by 63%, and it should be possible to prevent one-third of the deaths occurring on this particular day. Even then, the absolute gain would remain marginal, since only one death per 5 million person-days could be prevented.[1]


The total number of deaths is small. Drawing that conclusion, based on a small sample size is a problem. In order to be able to come up with larger numbers, to minimize the effects of the small sample size, other researchers looked at the motor vehicle collisions, rather than just fatal motor vehicle collisions. The assumption that the cause of the fatalities was anxiety, produced by superstition among the drivers is projecting a lot onto the drivers – without any evidence to support this supposed cause.

It should not be a surprise that the results of a much larger sample size contradicts the assumptions based on the much smaller sample.

We conclude that, in the Finnish traffic accident statistics for 1989–2002, females have not incurred more injury (or fatal) road traffic accidents on Fridays the 13th than expected, as a driver, bicyclist or pedestrian. We suggest that Näyhä’s contradicting result on fatalities is due to different sampling, non-optimal setting and chance in a fairly small data. However, this does not imply a nonexistent effect on accident risk as no exposure-to-risk data [18] are available. People who are anxious of “Black Friday” may stay home, or at least avoid driving a car. The only relevant data [4], suggesting a small decrease in highway traffic, is rather limited and should be confirmed with more extensive research.[2]


The law of small numbers is an attempt to expose the mistake of extrapolating from small numbers as if the small numbers are representative. Small numbers are misleading. Small numbers are often used to promote ideas that are not supported by adequate numbers – such as the claims that epinephrine improves cardiac arrest outcomes that matter, or that amiodarone improves cardiac arrest outcomes that matter.[3]


[1] Traffic deaths and superstition on Friday the 13th.
Näyhä S.
Am J Psychiatry. 2002 Dec;159(12):2110-1.
PMID: 12450968

[2] Females do not have more injury road accidents on Friday the 13th.
Radun I, Summala H.
BMC Public Health. 2004 Nov 16;4:54.
PMID: 15546493

Free Full Text from PubMed Central.

[3] Chapter 10
The Law of Small Numbers

Thinking, Fast and Slow
Daniel Kahneman
Wikipedia page



  1. And those are real publications, wow. They remind me of fancy Ig Nobels, e.g. this year’s in medicine: Silvano Gallus, for collecting evidence that pizza might protect against illness and death, if the pizza is made and eaten in Italy.

    • The first study is an example of bad science and bad math. (An outlier is a data point that does not fit in with the other data points.)

      1. The author is looking for an outlier in a small sample.

      2. When finding that outlier, the author claims that something about the outlier is causing the outlier to be an outlier.

      3. The hypothesis generated by the data mining is not tested on other data sets, because the author has jumped to a conclusion.

      4. The author’s conclusion is not admitted to be just hypothesis generating, but justified by the author, in spite of the weakness of the data.

      Why this phenomenon exists in women but not in men remains unknown, but perhaps the twice-as-high prevalence of neurotic disorders and anxiety symptoms in women (7) makes them more susceptible to superstition and worsening of driving performance. The interpretation of the finding is complicated in that a portion of the victims were passengers, bicyclists, or pedestrians.

      In an exchange with a critic of the conclusion of the paper, the author does admit that his results are speculative, but continues to try to justify his results.

      Possible flaws, listed in the article, include deaths of passengers, who obviously cannot be part of the causal chain, and it is also difficult to see why drivers (men or women) beset by this superstition would select women as their victims. Dr. Smith presents an additional problem that cannot be solved without a large study linking accidents to subsequent deaths. So far, any explanations must remain speculative.

      Traffic accidents and Friday the 13th.
      Smith DF.
      Am J Psychiatry. 2004 Nov;161(11):2140; author reply 2140. No abstract available.
      PMID: 15514434

      The quote is from the author reply (highlighting is by me.

      If men will have women as passengers more often, than women will have men as passengers, that can significantly affect the data, but that is dismissed as some sort of malice on the part of superstitious men.

      The only thing reasonable is that the conclusion is speculative.

      The larger paper, which debunked this author’s conclusions, made it clear that the speculation does not survive replication.