What I am wondering is if we could more generally alter a smaller range of player attributes and still get reasonably accurate performances ingame. Taking the idea even further, I was wondering if we could work out some sort of Excel formula which would modify a small range of player attributes to adjust their ingame performances to a level similar to their real life performance. Although this might not be as accurate as individual scouting would be, it would be more workable in practice and would hopefully make some players perform more accurately ingame. I don't know whether this is possible but I definitely think it is worth investigating.
The first step in this project is to simulate a season using the Lidas 1.3 08/09 rosters and compare the NHL players' season stats with their real life 08/09 stats. To get an accurate picture, the season will need to be simulated more than once. I think three times is a minimum but five times is more desirable.
To keep things consistent, I will have to create a human GM for each team. This will ensure that players are not traded, etc. Then I will set the game in vacation mode to quickly simulate one season. At the end of the season I will save the player stats to text files and import these into Excel.
As there is a limit of 16 human GMs in one EHM game, I am going to split the simulations into East and West Conferences. So I will set up five saved games where all of the East teams are human GMs in vacation mode and then I will set up a further five saved games where all of the West teams are human.
Setting up and simulating ten lots of one season is going to be rather time consuming for one person. Therefore I was wondering if there were any users who would be kind enough to simulate one season for me. I will set up the saved game for you so all you would have to do is put the game in vacation mode and then using the "print" option in EHM, export the player stats for 15 NHL teams (and their AHL affiliate teams. I won't bother with ECHL affiliates as it means another league to simulate and another set of stats to print) at the end of the first regular season. This will allow me to continue working on the site design in the meantime. Then, when I have completed the site design I can get stuck into this possible project.
So, to summarise:
- I will run five saved games for each Conference (i.e. ten saved games in total).
- For each saved game, a human GM will be assigned to all 15 teams for the particular Conference.
- The game will be set to vacation mode to run until 16 April 2007 ingame.
- Using the "print" option, stats for each of the 15 Conference teams and their AHL affiliates will be exported.
Please let me know if you're willing to help.
