A Haunting Actuarial Analysis – Part II

By Rick Gorvett
Posted on

Part I was posted last week and can be read here: https://www.casstudentcentral.org/a-haunting-actuarial-analysis-part-i/

Are you back?  Good.  Here is a summary of the aggregate survey results, including comments often originating from some of my wonderful CAS Staff colleagues, from most risky to least risky monsters.

(1) Dracula / Vampires 

DraculaF.jpgBased on average survey ranking, the most risky Halloween monster is Dracula / Vampires.  While no one selected Dracula as the most risky, almost all respondents ranked him (it?) second!  Concerns about Dracula, from a risk management perspective, centered primarily around severity:  if it bites you, that’s pretty much all she wrote, and you join the can’t-get-enough-of-that-O-positive club.  Interestingly, the fact that Dracula is only active at night was used by respondents on both sides of the frequency issue, as an indication that the risk might be both lower and higher:  active “only at night” versus “every night.”  One person suggested that, from the standpoint of exposures, there’s only one Dracula – but don’t the bitten then also become vampires?  Also, it was pointed out that the original Bram Stoker version could shapeshift, and was master of those he “recruited,” and thus had the potential to be a particularly grueling risk.  These last few comments demonstrate the importance of correctly ascertaining and defining the essential characteristics of your risk, in order to properly quantify and assess the underlying exposure, as well as the frequency and severity potential.

(2) Zombies 

ZombieF.jpgHere I must embarrassedly provide a caveat.  When I first sent this survey to my colleagues, I had included only the first seven monsters – I had somehow forgotten zombies!  Several recipients of the survey pointed out this dreadful lacuna (which I contend was caused by the insidiousness of zombies themselves, and their effect on my brain!), but by then it was too late to add zombies to the survey.  Nevertheless, a few respondents themselves chose to add zombies to their rankings and, although the sample size is much smaller than for the other classes, zombies did achieve the second-highest risk ranking, very close behind Dracula / Vampires.

Risk characteristics of zombies (at least for fast-moving ones) include their swarm mentality, the difficulty in “killing” them, and their relentlessness.

(3) The Wolfman 

WolfmanF.jpgThis one surprised me a bit – clearly, I’m no horror expert.  I thought that, for this dude, exposure would be low because he’s only a problem on nights with a full moon.  But others felt that frequency per unit of exposure (on those infrequent nights, he has an insatiable appetite) and severity (if you happen to be in the same room with this guy, you’re probably dead) would be significant, even if the exposure itself is only about once per month.

(4) Witches / Wizards

WizardF.jpgComing in fourth on our list of greatest risk-hits are witches and wizards.  A problem with them is quantifying the exposure level:  you never quite know who is and isn’t a witch (who would have suspected Selena Gomez?!).  And, like most of the high-risk-ranking monsters, the potential severity is enormous – one thought, or nose-wiggle, or finger-snap could take out the whole town.  On the good side, with respect to frequency of damage, unless they’re provoked, they’re generally not harmful.  So “just leave them alone” would be sound risk management advice!

(5) The Mummy

MummyF.jpgThere was some disagreement among respondents regarding The Mummy, and so here’s another example of the need to precisely identify, characterize, and define your risks.  Some felt he should be low on the list (perhaps you can simply outrun him).  Others were concerned that he’s the one on the list that truly seems to want to kill everyone, and that he himself can be hard to kill, requiring magic or an ancient Egyptian curse.  (So I guess we can’t just nuke him?)

(6) The Grim Reaper            

GrimF.jpgNow this is a really interesting one, and it has a very important lesson for actuaries and risk analysts.  The responses and rankings on The Grim Reaper had by far the highest standard deviation of any of the eight monster types – i.e., there was more disagreement in the rankings.  In fact, roughly the same number of people ranked TGR most risky, as ranked it least risky.  I think what has happened here is a common issue:  causality versus correlation.  Let’s face it:  if TGR shows up, you’re pretty much toast.  But is TGR the cause of death, or just the messenger (and thus the link between TGR is one of correlation, rather than cause)?  My interpretation (and again, we need to define and characterize our risks accurately!) is that death is caused by other factors, and TGR just shows up at the appropriate time.  In that sense, TGR would not be risky per se – but would be a rather reliable indicator that death is imminent.

(7) Ghosts                  

GhostF.jpgGhosts were generally low-rated, on the basis that they may be unable to do physical damage (although see Ghostbusters – “I think he can hear you, Ray”), and they tend to stick to their own stomping grounds.  But there’s certainly potential for emotional distress and mental damage.  Plus, again, how do we know the extent of the exposure?

(8) Frankenstein’s Monster 

Widely agreed to be at or toward the bottom of the risk list, Frankenstein’s Monster was interpreted by survey responders as a largely misunderstood and potentially friendly bloke (unless you cross him).  FMonsterF.jpgNot the brightest bulb on the tree, he can be both outrun and out-thought.  (But what’s up with that haircut??)

Many thanks to my friends and colleagues on the CAS Staff that responded to the survey!  Also, if you are interested in reading more on monster risk-rankings (as well as much other interesting and useful postings and information), please see the MyPath Blog, proprietors of which I thank for the initial idea that led to this Student Central post.

Happy Halloween!

Mark Your Calendar

Search CAS Directory