Months ago I read somewhere that people overestimate "often" and "likely" so this minor post is about how to use and evaluate these terms when discussing the odds of events (like dying of the Wuhan virus).
"Impossible":
So assume we have 100 events and out of the 100 none of the outcomes meets our criteria, whatever they are. This is a case of "impossible". This is the "can't happen" scenario.
"Rare":
Same 100 events but this time out of the 100 we have 1 event that meets our criteria. This is called a rare case. Sometimes referred to as an 'outlier". These are events that though not impossible, happen so infrequently that they are not generally something you expect to happen at any given time. Yes, you'd be surprised to see it BUT it's not something you concern yourself with.
"Sometimes":
Same 100 events but this time the criterion event happens 10 times. This is sometimes. It's not impossible, it's not really rare but you should be aware that it will come up from time to time. Still events not meeting the criterion will happen far more often.
"Often":
Same 100 events but this time 40 criteria fitting events occur. This is often. It cannot be ignored. Still there are more times than not that this event does NOT happen. You'd be a fool though if you treated it like a "rare" or "sometime" event.
"50-50"
Obviously this is break even. criteria event happens half the time.
"More likely than not":
Here we cross the 50% threshold. In these cases the criteria event happens most of the time vs. other outcomes. You should *expect* this outcome when confronted with the situation. Note that even though this event is expected the incidences of non-criteria events is in "often" territory and you should keep that in mind.
"Near certainty":
In this case, the criteria event is in the 90 out of 100 category. This could also be considered "within the margin of error". In such a case you would put your money on this event "all day, every day". Sometimes the non-criteria event will appear.
"100%"
Obviously this happens all the time. There is no "rare" event. No other possibility exists. The word "definitely" can be used here since "definitive" is 100%. This is why when discussing things like future COVID deaths, making declaratory/definitive statements is a sign that the speaker doesn't understand statistics.
"Give or take...":
This is a lay term for "variance". So if we look at the "rare" case we would say: "10 times out of a 100 give or take..." this is recognizing that things rarely fall into neat patterns. It won't be 10x every time, but over the long haul the events will converge on that.
Did you know that the engine temperature dial/indicator in your car doesn't actually show the temperature of your engine? Nope. The temperature of the coolant in your engine constantly varies. If it showed you in real time, the needle would be bouncing all over the place. Instead the computer averages out the temps and puts the needle where that average falls.
So keep these things in mind when watching news reports of the pandemic. Do the word they use match the data? Are they overstating the issue? Understating the issue?
