My recollection is WIS RPI uses a slightly different formula than the real RPI.

Quote: Originally posted by aporter on 1/12/2010

Quote: Originally posted by thewizard2 on 1/12/2010It's there in black and white.
0-5 vs. RPI Top 25 does not equal an RPI of 6 and shouldn't in any mathematical formula.

It's actually good strategy by bellini and certainly don't fault him for it. He was smart and took advantage of how the RPI system works.

I never said wins should count for more, but as mentioned numerous times before, the RPI system is far from perfect.

You shouldn't use the Top 25 from WIS to make your point. Although he is 0-5 against the Top 25, he could of had wins against teams that were more deserving of being in the Top 25. Beside the RPI formula already accounts for the fact that he lost those games. He wins 2 or 3 and he may be sitting with the #1 RPI

You're talking about the WIS RPI Top 25, correct? Not the regular Top 25 polls?

"could of had wins against teams that were more deserving"

It would appear by this quote you're agreeing that there is an issue. If they were more deserving, they would be in the RPI Top 25.

Correct, the RPI formula calculated those losses into the RPI system, but didn't provide the correct output.

This post could not be converted. To view the original post's thread, click here.

Quote: Originally Posted By thewizard2 on 1/12/2010

Quote: Originally posted by aporter on 1/12/2010

Quote: Originally posted by thewizard2 on 1/12/2010It's there in black and white.
0-5 vs. RPI Top 25 does not equal an RPI of 6 and shouldn't in any mathematical formula.

It's actually good strategy by bellini and certainly don't fault him for it. He was smart and took advantage of how the RPI system works.

I never said wins should count for more, but as mentioned numerous times before, the RPI system is far from perfect.

You shouldn't use the Top 25 from WIS to make your point. Although he is 0-5 against the Top 25, he could of had wins against teams that were more deserving of being in the Top 25. Beside the RPI formula already accounts for the fact that he lost those games. He wins 2 or 3 and he may be sitting with the #1 RPI

You're talking about the WIS RPI Top 25, correct? Not the regular Top 25 polls?

"could of had wins against teams that were more deserving"

It would appear by this quote you're agreeing that there is an issue. If they were more deserving, they would be in the RPI Top 25.

Correct, the RPI formula calculated those losses into the RPI system, but didn't provide the correct output.

What is the correct output? Are you calculating the RPI for every team and finding that the math is done incorrectly? As it has been pointed out, RPI is simply a formula. If you would like to see a change in the amount of emphasis placed on RPI, or an adjustment of what inputs need to be added to make it a more complete tool, then you need to bring that up.

The biggest problem of your argument is that you are attacking the rating that you are then using to support your argument.

This post could not be converted. To view the original post's thread, click here.

This post could not be converted. To view the original post's thread, click here.

When you look closer, its really plain to see why DC University has an RPI of 7. They're 16-6, played 22 games....17 of those were on the road where they're 11-6 (6-4 against that non-con...not bad)...they're 5-0 at home. Road wins and losses are essentially worth more than double home wins and losses (1.4 to 0.6...I'll get to this more later) and he manipulated his non-con RPI by scheduling 10 AWAY games against pimp competition.

When the RPI introduced its home/road weighting somewhere between 2003-2005 (it was 2004 I believe), I was ******. They were inherently making road games more worthwhile WIN and LOSS than home games by more than a 2 to 1 margin (1.4 to 0.6...its only a tenth away from having road wins count 3 to 1 v. home wins). It wrongly skews the RPI to favor/reward teams that played a lot of away games, not necessarily those that have played and done well against tough schedules. The fact that the RPI folks went 1.4 and 0.6 is ridiculous...they should have at most went 1.25 and 0.75, but if they were intent on weighting home and road, it probably should have been 1.2 to 0.8, 1.1 to 0.9, or 1.05 to 0.95, because the way it is currently, it inherently rewards road warriors simply because they play on the road. I know WIS is looking to be realistic here, but maybe the best thing to do is use the RPI WITHOUT THE WEIGHTS, like it used to be...then you don't see these guys rushing to schedule 10 AWAY non-cons, etc.

But like I said, given the RPI format and DC University's schedule, its completely understandable why they have a 7 RPI. EDIT: Their RPI adjusted W-L is 18.4 and 3.6 that's a 5 game difference from 16-6

Quote: Originally posted by colonels19 on 1/12/2010When you look closer, its really plain to see why DC University has an RPI of 7.  They're 16-6, played 22 games....17 of those were on the road where they're 11-6 (6-4 against that non-con...not bad)...they're 5-0 at home.  Road wins and losses are essentially worth more than double home wins and losses (1.4 to 0.6...I'll get to this more later) and he manipulated his non-con RPI by scheduling 10 AWAY games against pimp competition.When the RPI introduced its home/road weighting somewhere between 2003-2005 (it was 2004 I believe), I was ******.  They were inherently making road games more worthwhile WIN and LOSS than home games by more than a 2 to 1 margin (1.4 to 0.6...its only a tenth away from having road wins count 3 to 1 v. home wins).  It wrongly skews the RPI to favor/reward teams that played a lot of away games, not necessarily those that have played and done well against tough schedules.  The fact that the RPI folks went 1.4 and 0.6 is ridiculous...they should have at most went 1.25 and 0.75, but if they were intent on weighting home and road, it probably should have been 1.2 to 0.8, 1.1 to 0.9, or 1.05 to 0.95, because the way it is currently, it inherently rewards road warriors simply because they play on the road.  I know WIS is looking to be realistic here, but maybe the best thing to do is use the RPI WITHOUT THE WEIGHTS, like it used to be...then you don't see these guys rushing to schedule 10 AWAY non-cons, etc.But like I said, given the RPI format and DC University's schedule, its completely understandable why they have a 7 RPI. EDIT: Their RPI adjusted W-L is 18.4 and 3.6 that's a 5 game difference from 16-6

You miss the point of why they did this! They did it because teams like Duke, UNC, Wisconsin, BYU, area all unbeatable at home. It was a push to show who could win on the road and at neutral location in the tournament.

I am a huge Wisconsin fan and I must say if anyone comes to Wisconsin they are probably going down, Duke and Purdue both whom were undefeated at the time lost AT Wisconsin. They play those two games on the road and its going to be pretty tough for them to squeak it out.

Away games are worth more because they are harder to win and if you can win at someones place you are quite a bit better than them on that day. Home games at the top schools swing the game 5-10 points easy!

Now if you want WIS to implement a 6 game road schedule max go for that, that is something that we could agree on to curb the full away game schedule.

Schroedess raises the good point that the reason away games are weighted so much is that WIS made away games so much harder to win. Especially with A HCAs so easy to get these days, a road win really is worth much more than at home.

So here's the thing: a team, in the end, is only as good as the teams it beats (and loses to). If your plan is to "beat" the system by scheduling 10 road games, then you better schedule teams you're far, far better than, because if you only look for teams you're just a bit better than, you're going 6-4 or 7-3, which will get you about the same thing as going 10-0 against those teams at home. Or, you could play 5 at home, 5 on the road, maybe go 8-2. Or schedule 10 cupcakes on the road and go 10-0: your RPI in all these scenarios will end up in the same range. The point is, while the RPI isn't perfect, it is giving you, within its limited ability, what you've earned. If everybody chooses their strategy and schedules appropriately -- which is what happens -- then it's just time to show up on the court and do your best. Just like in RL.

There has to be some way to weigh what a team is worth; and record, opponents' strength, and home/away are the three things that measure outcomes, and therefore should be what matter. How you use this tool, as in RL, is up to you.

Jeff I was actually referring to Real Life. Wis does make it hard at D1 I have heard to win on the road, but in real life it is about impossible at the top 50 schools in the nation. You will get games here and there but the vast majority of schools win at home most of the time and only sometimes on the road.

Also in 2005 is when the weighting was added.

"In 2005, the RPI underwent a major revision. The 25-50-25 formula remained but the values of home wins and losses, and road wins and losses, were changed to the current weightings. Also, the committee dropped the bonuses and penalties of Factor IV.

Throughout the years, home winning percentages in Division I men’s basketball has always been at 66 or 67 percent. Men’s committee chair Bob Bowlsby explained the reason for the won-lost weightings in 2005 as, 'The committee adopted a formula that more accurately reflects the historical data regarding a team’s performance at home.'"

I was just pointing out how well you'd equated what happens in RL to HD :-)

Ah I got ya. Although I do agree with some that teams HCA should take longer to accumulate (5 seasons or more to move one full grade in my mind) and reduce (5 seasons or more to move one full grade), also there could be a soft cap to teams in smaller conferences that can only seat like 10 students lol.

Pack to the OP's main concern. . . .

this team completely deserves their RPI. there's no flaw. . .

Road games:
They have losses vs 4, 7, 8, 12, 19, and 122.
Thay have wins vs 32, 46, 49, 62, 76, 90, 92, 107, 117, 188, 213, and 232

Home games:
They have 0 losses
They have wins vs 62, 107, 122, 188, 232


i dont know what you're getting at with the "flaw" remark?

they have a 5 RPI, so they should beat every team with an RPI of 6 or worse? no

This team has one bad loss all season. The other five losses were against five of the best programs in knight (all on the road). They've beaten seven playoff caliber teams on the road. I just dont see the argument of RPI being flawed.

RPI works great. and it works better and better as we get more data. so deeper into the season it gets even more accurate.

RPI isnt bad at all, i think i finished 110 this season rpi wise. SOS is like 71.

Quote: Originally Posted By schroedess26 on 1/12/2010Jeff I was actually referring to Real Life. Wis does make it hard at D1 I have heard to win on the road, but in real life it is about impossible at the top 50 schools in the nation. You will get games here and there but the vast majority of schools win at home most of the time and only sometimes on the road.

Also in 2005 is when the weighting was added.

"In 2005, the RPI underwent a major revision. The 25-50-25 formula remained but the values of home wins and losses, and road wins and losses, were changed to the current weightings. Also, the committee dropped the bonuses and penalties of Factor IV.

Throughout the years, home winning percentages in Division I men’s basketball has always been at 66 or 67 percent. Men’s committee chair Bob Bowlsby explained the reason for the won-lost weightings in 2005 as, 'The committee adopted a formula that more accurately reflects the historical data regarding a team’s performance at home.'

They went too far, like I said if you're going to do this, do at most a 1.25 to 0.75 ratio, not a 1.4 to 0.6 which almost a 3 to 1 ratio and the weights seem arbitrary as hell. My numbers are arbitrary as well, but seem to make more sense. 1.2 to 0.8 1.1 to 0.9 an 1.05 to 0.95 would work better IMO than the current weights...