At the risk of being boringly repetitive, I wonder why anyone trusts performance calculations, for an intersection departure
where the margins are tight.
There are simply too many unknown variables to make any such calculation more than an approximation at best; I mentioned some earlier in an unnecessarily aggressive post. They include the following (with apologies to those who have pointed some out);
The internediate slope from the intersection, perhaps the lowest point of the runway. The max permitted intermediate slope can be well above the overall published net slope.
The true aircraft weight. A loadsheet ramp weight has an accuracy of, let's say, +/- 1 tonne, perhaps more depending on your aircraft type, if we are being brutally honest. The performance data provide for this, but is the allowance enough?
The runway surface. Has it got the desired friction? Was there a recent shower the crew don't know about?
The actual head-wind component during the roll, if it's a bit variable and gusty.
The actual state of the brakes, bearing in mind that the aircraft will probably be stopping with asymetric power. How close to the next check are they?
The actual distances available from the intersection. OK, ATC can usually provide, but didn't someone mention a ruler on the chart?
Can aircraft always be stopped as their Flight Manuals and performance graphs/tables say they can? (Certainly not one I was in at Beirut, on one fairly alarming occasion.)
And of course, the stopway may not be quite what it's cracked up to be if a heavy aircraft hits it. Not every airport is perfect, but the AIP or equivalent won't say that..
And so on. None is likely to cause problems by itself, but supposing the holes line up one unlucky day?