|
Amaroq -> RE: player rating help (6/2/2006 9:38:10 PM)
|
That's actually a great question, and one without a 'quick' answer.
In the 'typical' game in this genre, a player gets a rating that has a concrete relationship to some probability. So, let's say that you and I were writing a baseball game, we might give a hitter a number, "HR", that is the 'percent chance that, if he gets a hit, it becomes a home run'. When we're doing the calculation of an at-bat, we'd have code that looked like this: Determine if its a hit. If it is a hit, roll a random number from 0-100. If that random number is less than the "HR" number for the batter, its a home run! Our game would be childishly simplistic - we haven't considered the pitcher, the ballpark, the altitude, the weather, the wind, the outfielder leaping at the wall - but that's the basic concept.
Many games in the genre surface some of those 'ratings' in their direct form, and then have some others that are 'hidden ratings'.
Games in the genre typically 'scale' those ratings, so we might change it from a percentage, to a number from 0 to 100 where any hitter with more than "25%" (40 HR on 160 hits, for example) is shown as '100' - the player wouldn't be able to tell the difference between McGwire at his peak and Bonds at his, they're both at the 'top of the scale'.
The better games in the genre will combine numbers - for example, we might keep 'HR', 'chance of a double', and 'chance of a deep fly ball' as three separate numbers in the engine, and blend them into a number which we display to the user as 'Power'.
Our competitor might 'read' the data on HR, doubles, and sacrifice flies, and use those three to generate a 'Power' attribute, which is then used to determine a 'drive which might become one of the three', and use other variables (ballpark, outfielder, weather) to determine what that long drive turns out to be. Both are legitimate approaches.
PureSim does one of those two, and without reading the code I can't tell which one it does.
However, it also adds an entertaining twist on the 'scaling' function: it uses a kind of T-scale, such that *typically*, a player rated at '50' is close to the average player in the universe - in other words, if you're playing a 35-man association, 'good enough for the big leagues', and if you're playing a 60-man association, 'just about between AA and AAA'.
...
'Potential' can mean different things in different games of the genre, also. Some games use it to represent 'the best this player could possibly be at his peak'. Other games use it to represent 'the chance that this player will improve', or 'how much this player will improve at any given time.'
PureSim falls into the latter category, with the 'green arrow up' icon indicating a positive chance that the player will improve, and the 'potential' number indicating how much he's likely to improve on a given year. See my post over in The Locker for more information on how to get the most out of your farm system and developing young players
|
|
|
|
