Originally posted by DaRockShow
Originally posted by Bolts2thaShip
Originally posted by Bolts2thaShip
But nonetheless, a head-to-head victory, combined with--you know--not losing by 101... you'd think that would be sufficient statistical evidence to show they are better. It wouldn't matter who the other victory came against if the CCSM has gotten blown out in every game, and one of the losses being to the same team (SoYo) and the other loss being to a team that is not doing too bad, either(Warsaw). Seems like a good rough estimate of talent, but it just shows there are still obvious flaws in the system.
SoS may be big-time over-weighted. Because it really doesn't matter who you are playing if you are laying down and dying for them. Maybe a smaller modifier based on the MoV or something? Or maybe even a bonus multiplier to SoS based on a closer MoV? Not too sure what the best approach is, but I am sure the situation has come up before this one. Perhaps the original author is working on new tweaks.
But your are missing the fact that whether you lose by 40 or 240, it really doesnt make a difference if the deficit is more than 35. You are putting to much weight on a 100 point game. Math is math and numbers dont lie, over the course of the season if CCSM continues to lose do you think they will still be in the same position they are in now? Dont take one weeks results to heart, if we finish the season and they end up 0-16 and in the top 15, then there is a problem with the numbers. The way this is set up, that is not going to happen. SoS will work itself out more and more as more games are played, especially since we all play the same teams until the last week of the season. Not sure why guys arent seeing this.
Actually, I'm not at all missing that part... thus the whole last 25% of the post mentioning the MoV adjustments that might help prevent stupid things like this from happening.
Originally posted by Bolts2thaShip
Originally posted by Bolts2thaShip
But nonetheless, a head-to-head victory, combined with--you know--not losing by 101... you'd think that would be sufficient statistical evidence to show they are better. It wouldn't matter who the other victory came against if the CCSM has gotten blown out in every game, and one of the losses being to the same team (SoYo) and the other loss being to a team that is not doing too bad, either(Warsaw). Seems like a good rough estimate of talent, but it just shows there are still obvious flaws in the system.
SoS may be big-time over-weighted. Because it really doesn't matter who you are playing if you are laying down and dying for them. Maybe a smaller modifier based on the MoV or something? Or maybe even a bonus multiplier to SoS based on a closer MoV? Not too sure what the best approach is, but I am sure the situation has come up before this one. Perhaps the original author is working on new tweaks.
But your are missing the fact that whether you lose by 40 or 240, it really doesnt make a difference if the deficit is more than 35. You are putting to much weight on a 100 point game. Math is math and numbers dont lie, over the course of the season if CCSM continues to lose do you think they will still be in the same position they are in now? Dont take one weeks results to heart, if we finish the season and they end up 0-16 and in the top 15, then there is a problem with the numbers. The way this is set up, that is not going to happen. SoS will work itself out more and more as more games are played, especially since we all play the same teams until the last week of the season. Not sure why guys arent seeing this.
Actually, I'm not at all missing that part... thus the whole last 25% of the post mentioning the MoV adjustments that might help prevent stupid things like this from happening.
Last edited Jun 20, 2008 16:44:10
