I know that you messaged Jack, but it wasn't aimed at him, I thought it was fair to reply, especially as you made some interesting points and come from a different perspective.
In an ideal world, we would of course like to lock down everything earlier. We do lock down the ME from adding features earlier in the development cycle, and then it is about balancing the engine and fixing bugs and issues. If we were to lock down the engine even earlier, there would be even less features and improvements, and on a yearly iteration of our game, our development cycle is quite small anyways, so losing 3-6 months of development, iterations and improvements would mean there would be little difference between one product to the next.
In terms of tooling and automation, we automate the engine continuously, we continually balance the simulation and this is a constant process of making sure the Match Engine is as balanced as it can be against real life stats, though automation and stats can only get you so far and the eye test is vital. We spend time improving our tooling and automation each cycle.
In terms of our development practices, I've been the producer on the Match Team for 5 years now and we've changed and evolved as a team during that time, our development practices have improved during that time and we review them continually, especially after every cycle to see how we can improve the way we work to get the best product for our fans and do the best within the time frame we have.
To highlight Jack's point, we are striving for the perfect simulation of football, that is what we are aiming for and well, there is always things we can improve and we can always make it better.