- My Astros's Brad Lidge 5th blown save... (18 inning game)
- Royals's Luke Hudson 11 runs in 1st inning...
- 11 earned runs charged to Orlando Hernandez...
- Same game Reyes hits 3 homers, in a loss...
I've seen some crazy things in PureSim and chalked it up to it just being 'wonky' in some respects. After witnessing the above I'll complain no more. My experiences in PureSim are in-line with MLB I guess
That's actually one of the trickiest parts of writing a sports simulation: its easier to get the 'average' correct, and much harder to have the correct distribution curve, especially when its non-regular.
Batting average, for example, shouldn't be a bell curve centered on the average batting average; there should be a billion guys capable of hitting .160, many fewer capable of hitting .200, many fewer capable of hitting .250, .280 hitters need to be somewhat rare, with .320 hitters being extremely rare - but still acheiveable. That's not a distribution that its easy to model.
Further, in any given situation, you need to have the 'on any given Sunday' feel, so that bad teams have a chance, excellent players can blow it, etc; but you can't make it so random that the human player doesn't have any actual impact on the game.
I've wondered about this, too. I've had some horrendous things happen, sometimes to me, sometimes to the AI manager.
Flukes DO occur, but, as rpommier noted, real-life stuff can beggar credulity on those "once in a blue moon" occasions.
