In the old days (prior to FM05(?)), season predictions were dynamic; they changed throughout the course of the season. Teams were differentiated, ranging from title chasers to relegation battlers. I liked this approach because it allowed teams, performing above initial expectations, to approach players that would normally be out of reach.
Enter the new system: A season prediction is published around the start of the transfer window. An example: Your team is promoted and predicted to finish dead last next year. You do well in the transfer market and at the start of the season your squad is totally transformed. Yet the prediction persists - with some annoying consequences. Cue the January transfer window. Your team has defied everyone and is currently in the top half in the league – relegation is virtually impossible. Yet everyone still perceives you as a relegation favorite. Why? You can't approach potential players even though your finances should accommodate their demands. You see teams of similar reputation below you snatching away your targets at discounted rates. You even have a hard time renewing current contracts because your players think relegation is a foregone conclusion. You soldier on and finish the season way above expectations, yet most players are still off-limits until next season's predictions have been published. Why?
I would propose a new system consisting of two indicators: A static pre-season prediction, like we currently have, combined with a monthly dynamic prediction, using the old system, ranging your team from relegation battler to title chaser. The dynamic prediction would determine your ability to approach targets and to renegotiate current contracts. It shouldn't be difficult to implement and it would make everything a little less cumbersome. I know the AI can distinguish between over- and under-achievers in the league but why does this information not spill over into your ability to approach players?