With the control simulations finished, the next step in this project was to create replacement players and performing season simulations with these players instead of the starter. Since the default back-up players in the video game vary in ability, I ultimately decided that creating a universal back up player for each position was the best way to obtain most accurate results as possible.
In total, I created five replacement players: a quarterback, running back, left tackle, linebacker, and cornerback. After noting that the default rating for a computer generated player is 65/99, I examined the statistics of the back-up players in each position and increased the skill level in the statistics that were related to their position by 10 points. Although these are replacement players and should not be nearly as skilled as the starting player, their positional skill should emulate the ability of a second string NFL player. For example, I left the all ratings for the replacement quarterback at the default score, but increased his ability in statistics specific to excelling in the quarterback position such as ‘Short Throw Accuracy’ and ‘Throwing on the Run’ to 75/99.
Each replacement player had a different set of skills that was altered to increase their positional ability, but each replacement player maintained an overall ability level of 63/99.
Next: MWAR Testing
Previous: Collecting ‘Control’ Data