Willie the Wimp
You are correct - 50 is not a big number. But 50 out of 3000 pilots means over 1.5% less flying was possible(since all crews are either hour limited or days off limited) - and 1.5% of $100Billion is $1.5Billion.
We still have the same planes - just flying them 1.5% less - so 1.5% less revenue. Fixed costs remain the same - variable costs amount to about 40% - so that means losing those 50 extra pilots cost the airline a billion in 'opportunity cost'. That is actually lost revenue minus reduced variable costs. Unfortunately we can only reduce the variable costs by the marginal(spot) price of fuel not the hedged rate as the hedged stuff gets used first.
Now if the training machine is running at max chat - and let's just pretend it is for say 2-3 years - that is 2-3 years in a row that losing that 50 pilots CANNOT be recovered. So now we are up to $2.7Billion in cost(3 years).
If we lose 50 extra every year for 3 years - well, you can see the costs mounting up. For the year 2 loss of 50 pilots add another $1.8B, for year 3 add 0.9B. So for just that 3 years, a loss of $5.4B - for 'only' 50 extra pilots leaving per year!
So I guess we are like a fleet of a hundred spare Spitfires during the battle of Britain only missing 1.5% of their parts - just 100 propellers! Apart from that, they are good to go. Accountants see a 98.5% equipped fleet of fighters - the real world sees 100 grounded aircraft!