KG Erwin -> RE: How to Explain the Ratings System's Subtleties? (6/28/2006 3:32:31 AM)
|
Al, I DID mean that, but I get what you're saying, too. Yes. Having the engine automatically generate ratings based on player input could allow for each new season to be recreated without the need for a Lahman update. Yes -- that would be great. Now, going back to importing historical players -- I've tried both historical rosters and draft rosters, and set up templates for both. I think what you're saying is, each time I import Jackie Robinson from 1947 his ratings should always be the same. Is that what you're saying? I disagree, but it depends on the scale you use -- I kinda like the 1-100 scale. However, each time one imports the 1947 players, the ratings differ each time. I DO like this aspect -- Pure Sim isn't like the movie "Groundhog Day". Each time you repeat a given season, the ratings change. However, ALL players for a given season are rated relevant to each other, NOT some universal constant. That's the thing that's throws many players off. This is where it starts getting deep. If one throws together a league of all-time geat players, the ratings will start getting really skewed. I frankly don't think the game was meant for that. The original intent, in using all fictional players, was an exercise in judging talent and tailoring teams to your style of play. That still holds true, BUT, with the import of the Lahman database these criteria could be applied to actual players. This is how the big picture really got confused. In one 1946 draft, Red Schoendienst might get a speed rating of 85. In another, it might be 76. He's still a good runner, but a random element is applied to determine how good he was in that particular year against every other runner in the league. The tendency is still there, but his exact stats are based on the given number for that year. This model WILL smetimes give ahistorical numbers for individual players, but overall team BAs and ERAs will retain a close resemblance to history.
|
|
|
|