gil, consider yourself owed!

i dont know about anyone else,  but the CC posting improvement was, in my opinion, one of the greatest improvements in the history of HD,  and quite possibly in the whole history of humankind.  i am forever in your debt.


i wanna hear more about this banning and the crying and the pissedoffedness.    that sounds awesome.

and i want to hear more about free scouting and other bugs.    how does that stuff screw you over?  wouldnt it give you a big edge if you keep it to yourself?

i wonder what i would do if i found a bug that gave me an edge by allowing me to scout for free or get free visits or whatever.
hmmm...id like to think id tell seble.   but it probably would depend on whether i was in any worlds with kelby.

RPI is not and should not be the end-all, be-all. It is a good measure, but no more. And it is exactly the same way in real life. 21, 30, 30 RPI have all been left out before.

Posted by oldave on 2/3/2014 3:40:00 PM (view original):
gil, consider yourself owed!

i dont know about anyone else,  but the CC posting improvement was, in my opinion, one of the greatest improvements in the history of HD,  and quite possibly in the whole history of humankind.  i am forever in your debt.


i wanna hear more about this banning and the crying and the pissedoffedness.    that sounds awesome.

and i want to hear more about free scouting and other bugs.    how does that stuff screw you over?  wouldnt it give you a big edge if you keep it to yourself?

i wonder what i would do if i found a bug that gave me an edge by allowing me to scout for free or get free visits or whatever.
hmmm...id like to think id tell seble.   but it probably would depend on whether i was in any worlds with kelby.

trust me, you don't want to hear about it, and nothing about it was awesome. i wasn't really banned, someone complained that i had two teams within 1000 miles, in different divisions (as i had for years). even though i picked up both before you couldn't have teams within 1000 miles in different divisions, i was forced to drop one of either SIUE or south carolina in tark, which was the program i was most into at the time (SIUE obviously being hard to drop for legacy and conference loving reasons). so i refused to drop one and just dropped all my teams everywhere. but in the month before while i had to decide i went on sort of a mega rant, probably posting about 50 pages of dissenting opinions on the forums and in tickets, over the course of a month. in retrospect, i wish i hadn't. it didn't change anything and just crapped up the forums, and wasted a lot of seble's time. and there are still tons of coaches with multiple IDs, who don't use obvious IDs like "gillispie" and "gillispie1". but luckily the guys who report these things aren't aware there are so many veterans who still have multiple teams, so mostly, the incident did not affect the majority of coaches with multiple teams. most of them were smart and just shut up, but i always was open about having multiple teams, on principle, so i got dinged for it. emy willfully moved one or two of his teams to comply, and still the idiots came out and called him a cheater for not moving them earlier :/ but the important thing is the idiots who think the connection between coaches with multiple teams per world and success is driven by cheating, not the passion of those coaches, now can feel better and can go on trying to find other reasons for their lack of success, blissfully ignorant of the fact that there are still tons of guys with multiple teams per wor--- err, i mean, happily knowing that all the coaches now only have 1 team per world.

yeah the free scouting trips was a bug that would have been advantageous if i used it, but it seemed unfair to do so. you could also get certain efforts eaten if you did them at precisely the right time, that was the part about it screwing you. what ****** me off is i made the effort to tell seble about that free scouting trip info bug and then a month later all these people get all ****** off because i have 2 teams in a world, despite the fact that i always had 2-3 teams and tark, and never made any attempt to hide it. to them, it was cheating, because "no coach would have 2 teams and NOT share FSS data with themselves". seems to me the guys who can't conceive of a coach not cheating given the opportunity, are the guys we should really be looking at :O

I am hoping to do a more serious study, but it appears to me that the "Away" game winning percentage (the %, not the number of wins) seems to have equal weight with the RPI in the Projections report. This appears to contribute to so many teams with poor RPI, (maybe they've played really weak opponents), being rated more favorably than teams with better RPI but worse away winning %. The seeding doesn't appear to take wins against Top 25, 100, etc. into consideration since that all sort of shows up in RPI. It appears that teams (like myself) who schedule more away games against good human opponents (trying to get the SOS and RPI as low as possible) are being penalized relative to those who just win the games they play away. Anyone else looking at this?

Exactly backwards. RPI doesn't take those things into account. RPI is a simple formula that, aside from amassed records and the home away multiplier doesn't care WHO you play in any given game.

You ARE penalized for playing good human teams ...and losing.
In noncon that you scheduled you were 3-5 against top 100, 1-4 against top fifty.
Overall you are 7-7 against top 100, 1-4 against top fifty.

Yeah, compare them to my Redlands team:

7-6 vs top 100
4-4 vs top 50

That sort of performance is what gives me an RPI of 40 but a Projection of 25.

To be the man, you got to BEAT the man....sometimes....

Posted by arssanguinus on 2/7/2014 8:31:00 AM (view original):
Exactly backwards. RPI doesn't take those things into account. RPI is a simple formula that, aside from amassed records and the home away multiplier doesn't care WHO you play in any given game.

You ARE penalized for playing good human teams ...and losing.
In noncon that you scheduled you were 3-5 against top 100, 1-4 against top fifty.
Overall you are 7-7 against top 100, 1-4 against top fifty.

I didn't mean that the formula literally cares WHO you play. I meant amassed record, both of your opponent and opponents' opponents' records. Those show up in RPI. But giving the seeding formula equal weight to simply "away winning percentage" negates the advantage you can have in scheduling tough RPI opponents away.

Posted by Trentonjoe on 2/7/2014 8:45:00 AM (view original):
Yeah, compare them to my Redlands team:

7-6 vs top 100
4-4 vs top 50

That sort of performance is what gives me an RPI of 40 but a Projection of 25.

To be the man, you got to BEAT the man....sometimes....

What I'm saying is that the Projection doesn't care about top 100 or top 50. That indirectly shows up in your RPI (the tougher your SOS, the higher your RPI ceteris paribus). You can't exploit the RPI game the system as much as you used to be able to. I think the engine looks at "away winning percentage" without caring who it's against.

For example, compare me (#51) with the team projected above me:
Rank  RPI SOS Overall Home Away
50    74  267   24-2    11-2     13-0
51   37    82     19-7    8-0       11-7

I played a much harder schedule and have a much better RPI. But #50's 13-0 away offsets that, it's a huge boost. Could be something else is weighted higher, like overall record. But eyeballing it, it seems much more dependent on Away Winning %.

What is offset is that you didn't have a good record against the teams that made up your better RPI/SOS.

The projection report explicitly DOES include components against the top 50 and top 100 RPI. Its not even a question.

Back from the update from when the projection report was introduced, this is what goes into the projection report.

•Improved logic for tournament selection/seeding. The new logic will go through each game and score it based on outcome (win/loss), opponent rank, opponent RPI, margin, and location (home/away/neutral).

Posted by a_in_the_b on 2/7/2014 11:37:00 AM (view original):
Back from the update from when the projection report was introduced, this is what goes into the projection report.

•Improved logic for tournament selection/seeding. The new logic will go through each game and score it based on outcome (win/loss), opponent rank, opponent RPI, margin, and location (home/away/neutral).

Thanks for posting that. The team above me has two wins against the top-100, with one loss better than any of mine. I guess that helps make up the difference.