[malor]

I assume you mean the red arrows on the monthly development reports. If yes, just think of them as estimates. You scout is making evaluations constantly. Behind the scenes, OOTP is doing it's random development calculations. Some months, you win, some months you lose. Over time, the more you spend and the better coaches you have, the more likely the better players will emerge and be promoted through the minors and eventually make the majors. This is not a one-year process, it is a 3-5 year journey for a few players, with most failing and topping off to be minor league roster fillers for the best of the rest. The worst should be cut each year to make room for a new year of draftees, international complex promotions and maybe a few free agent signees you think may have a chance or fill a need for your organization.

If you sweat each and every up/down for each player, your going to have one continuous headache. Just place your players at the minor league level that best matches their current scouted potential, stats, age, pro-experience and let them play a few months. Check the stats once a month for your minor league teams, move the players that are crushing their level up, move those that are really struggling down. The others stay put to fill out the rosters. At some point, after doing this enough times, you will be able to spot the better players sooner and promote them sooner.

While it may be possible, I don't think you will ever know right away you have a future star. The game simply does not allow you to know positively how a player will develop unless you cheat and use the editor. Even than, the randomness of the development engine may change that. I've had many 1st round busts and many more other players in the later rounds develop to at least last for a few years in the majors as a bench player. All the rest of the draft classes last about 3 years and are released.

This is the life of a baseball player and the GM that manages the organization. It is what makes OOTP the best baseball simulation that is available.

I hope this helps. The only thing you are doing wrong is looking to closely at numbers that really don't mean much until you have enough history to see the developing patterns.