Hope this isn't too late in the thread or too pedantic.
There are two different things being discussed here - life expectancy and probability of an event (in an RPG).
Life expectency is the average age of death in the population computed at a given life point. Generally, it's reported from birth when used casually. However, it can and is also a calculation that is expected to change throughout the lifespan. A person's expected lifespan lifespan actually has a tendency to increase throughout the span ....as the person passes various risky events. (example below for the actually interested).
What the initial post was noting was more like the risk factor - "what is the probability that a person will be dead by or at a certain specific age"; and, in a game setting, it is easily answered - as he did. Applying it to a population model, or even more, a real world model, and adding in all the issues of changing risk as time progresses, and variable hazard issues, doesn't work. The only real time dependent covariate in the traveller survival function is something involving Greed of player(G) and Coolness of Character (C), and Bitchingness of skills thus far acquired(B), as measured in standard arbitrary scales of 0 to 1.0.
So the lifespan (expected) need to be modified by the function: "Rolling and starting play before you die of old age"(R) (the inverse of which is: "having to take another aging test just so you can get another level of fusion weapons ") :which can be estimated as something like Rprob= G/((C*B)^1/2), which would suggest that the lifespan estimate (L) should be modified by 1/Rprob creating L'; the immediate risk at any point in a travellers career is obviously the derivative of L' at time t; proof is left to the reader.....
HERE IS THE EXAMPLE FOR THE TRULY BORED
Heart failure is a good example. Say, a person has a life expectency of 60 years at birth - that means that th average age of death in that group is 60 years old. At one year, his LE may well be up to 70 - assuming the first year of life has a high mortality. Then, at age 18, it may be up to 80 - assuming that all the various childhood risks are survived...it can also be noted as having an estimated lifespan of whatever is left - in this case 62 years (80-18), Now at thirty, his LE hasn't changed much, say 82 ; his est Lifespan is down, obviously (you live years faster than you gain them, ain't that a bummer). Now a common event (say) is a heart failure incident at about age 40 -the avearge age of heart failure in his population being say, 41. Life expectency for a person with heart failure is 1 year. So, if he has a diagnosis of Mitral Regurgitation at age 41, his estimated life span is now 1 year. His life expectancy is 42. (these numbers are all hypothetical and made up, BTW, I don't have life tables in front of me). However, at age 42, he now has a lifespan estimate of 5 years, and an expectency of 46 - a large proportion of patients will die during their first incident ("I'm coming, Martha") or very shortly thereafter. Those that survive the event, generaly survive much longer.
However, it turns out he is a Scout, so in fact at age 18 his life expectancy is Zero terms, as sure enough, during his first term, he gets blasted naked and on fire into the icy void of space as his ship collides with an alien artifact and explodes while being chased by pirates as the jumpdrive malfunctions.