Post by WayoshiM on Jul 25, 2011 23:43:43 GMT -5
Hi everyone,
On the old Sony boards 11 months ago, I introduced WOFTracker, an application I've been working on with Java (it's nothing professionally polished or anything). It is able to take actions of a WOF game as input and output a text file with a precise recap of the game.
It wasn't ever meant to be a replacement for the Recap Team, although I did use it as a template for my own recaps - the main goal was eventually for the program to also construct statistics for a WOF episode, save these statistics, and be able to combine this data with any other number of episodes (think: compare week to week average prize puzzle amounts to see the difference a theme makes, look at the average R1 winnings from the whole season, etc. etc... things with the same depth as those tracked by hand on Scorpz's well-known Price site would be automatically done. Brilliant!
After my first year of college pretty much stifled work (I made text recap files from it for awhile, but it does take a bit, and no real point once the major bugs were removed), I'm back at it again, and am now getting into the statistic portion. I've set up the structures pretty well to record the statistics, now I need to figure out exactly what I want to record. One thing I ran into tonight is I originally had a way to separate all wedges hit by player (red, yellow, blue) in a maingame. But then I realized... how useful would that be? Even if you treat every player uniquely in combined game analysis, what the "average" player hits in a game (across all rounds, as the Wheel changes) is a murky picture. Separating all wedges hit by round is very clear.
I'd like to open this thread up as I go through figuring out what will eventually be analyzed on WOF episodes (hopefully for S29 in some partial, if not complete, form) to any suggestions on what to include. I may have already thought of it (frequency of letters called, in puzzles, winning averages, spin strength averages, what happens to the cardboard, among other things), but I'm certain to have missed some things, and I figure everyone should get a chance to give input in this aspect.
On the old Sony boards 11 months ago, I introduced WOFTracker, an application I've been working on with Java (it's nothing professionally polished or anything). It is able to take actions of a WOF game as input and output a text file with a precise recap of the game.
It wasn't ever meant to be a replacement for the Recap Team, although I did use it as a template for my own recaps - the main goal was eventually for the program to also construct statistics for a WOF episode, save these statistics, and be able to combine this data with any other number of episodes (think: compare week to week average prize puzzle amounts to see the difference a theme makes, look at the average R1 winnings from the whole season, etc. etc... things with the same depth as those tracked by hand on Scorpz's well-known Price site would be automatically done. Brilliant!
After my first year of college pretty much stifled work (I made text recap files from it for awhile, but it does take a bit, and no real point once the major bugs were removed), I'm back at it again, and am now getting into the statistic portion. I've set up the structures pretty well to record the statistics, now I need to figure out exactly what I want to record. One thing I ran into tonight is I originally had a way to separate all wedges hit by player (red, yellow, blue) in a maingame. But then I realized... how useful would that be? Even if you treat every player uniquely in combined game analysis, what the "average" player hits in a game (across all rounds, as the Wheel changes) is a murky picture. Separating all wedges hit by round is very clear.
I'd like to open this thread up as I go through figuring out what will eventually be analyzed on WOF episodes (hopefully for S29 in some partial, if not complete, form) to any suggestions on what to include. I may have already thought of it (frequency of letters called, in puzzles, winning averages, spin strength averages, what happens to the cardboard, among other things), but I'm certain to have missed some things, and I figure everyone should get a chance to give input in this aspect.