Warrior General Games

Originally posted by Caseos
Originally posted by Daddy Warbucks

Originally posted by Caseos


I still think it's about strenght of schedule. When you guys start playing teams that are more evenly matched that you don't just simply blow out of the water you'll move up the power rankings. It shows no power at all when you crush teams that are made up of players that are an average of 4-5 levels below you.

Yet it shows power to be one of those teams for some reason.

What shows power is when you beat teams that are more evenly matched with you. Your Pigs lost to the only two teams that you have played that were anywhere near a good matchup. Then you crushed every team that you've played that you had a HUGE statistical advantage over.

Up to this point my Bullfrogs have beat all of the teams that were, relatively speaking, close match ups and have only lost to the teams that had HUGE statistical advantages.

As Killer says, it's going to even out over the course of the season and everyone will take their rightfull place in the hierarchy, but early on in the season, what we have seen in the power rankings has been largely about the matchups.



That's nice in theory but power rankings don't entail entry of every player's level, that would be ludicrous.
It has to do with schedule, I know. But statistically speaking alone, if team A beats team B, team B should be lower on the rankings.

Team A beating team B absolutelty does not make them a statistically better team. To prove the point about statics, in the 2007 NFL season, the Miami Dolphins were 1 and 15 while the Baltimore Ravens were 5 and 11. The Ravens were clearly the statistically better team and would clearly have been above the Dolphins on any analyists power rankings that season. But Miami beat Baltimore for their only win that season.

By your reasoning, since the Dolphins beat the Ravens, the fish should have ranked higher than the birds.

Team A beats team B and that puts A above B on the STANDINGS. If you look at the Western conference standings I think you'll see the results that you want, at least you'll see results that are simpler to understand. Your Guinea Pigs are right where they should be in the standings based solely on wins and losses.

But power rankings are a purely mathematical calculation that takes into account any number of factors that the person who wrote the script decided were important.

At the end of the season the only statistics that will matter at all will be the standings that were a result of wins and losses, and the power rankings will be useless information that everyone will forget about.

There are clearly differences in the teams in this leage. Our conference is made up of teams in at least 2 but possibly 3 separate "tiers". Your team is clearly in the top tier, and my team is in the second tier. We have yet to lose to a team that is in our teir, and you have. In a head to head matchup your team is without a doubt the better team, but overall, statistically speaking, that might not neccessarily be the case. And that is what this power ranking script is showing at this point and time in the season with the limited number of data points that it has available to make calculations with.

My degrees are in physics and engineering, not statistics, but I know enough to know that they can be very complicated and difficult to understand.

The unfortunate part is that the projections aren't listed.
These numbers aren't any more valid than the current league standings, they're simply an alternate list which pulls from separate factors. Considering that the league standings are those which will be relevant at the end of the season, the only thing these standings provide are a consolation for teams that likely won't be at the top in the end. It doesn't really make sense to me.

I've seen projections before from these PR. Based on how much each team scores on those it faces, and what other teams have scored on them, and how little they've been scored on etc, they create a projection that assumes where each team will be at the end of the season. Imo those are more interesting.

Originally posted by Daddy Warbucks

I've seen projections before from these PR. Based on how much each team scores on those it faces, and what other teams have scored on them, and how little they've been scored on etc, they create a projection that assumes where each team will be at the end of the season. Imo those are more interesting.

Feel free to post those.

I wouldn't trust any projection about final standings from this game. It is still WAY too early in a beta game that has balance and AI issues yet to be worked out. The amount of points you score and the number of points scored against you is irrelavent for a number of reasons. First and formost being pure chance the random number generator may have gone seriously in your favor during any given game. Second is that the number of points you score in a game is just as much about the matchup (if not moreso) than it is about your teams awesomeness. If your teams overall ranking is 45 and you play a team with an overall ranking of 23 it's meaningless that you scored 146 points on them. That's like the New England Patriots playing the Harrison High School Hornets.

Now if two teams with overall rankings of 45 play each other and you still score 146 points on them and they score 7, that says somehting about your ability and your game planning.

Until Bort makes the leagues evenly matched (withing 3-5 points at season start) there are going to be discussions like this one every season, because the teams are just so unevenly matched as to make standings, and power rankings, and ultimately statistics totally useless for determining realistic outcomes.

The Guinea Pigs are a really good team, and my money is on them to win the conference, if not the league this season. Don't put any stock in what you find in power rankings, they are just fun numbers that are good for starting a buch of smack talking.