Scheduling...

[billyjack]

This kind of feeds off the Ratings System thread, but I don't want to hijack that, and I think this could be a good discussion.

So, I'm looking at various schools' schedules.
Two in particular stood out as being odd... Pitt and Louisville.

PITT is 7-1, and OOC this year has played/will play:
Gonzaga (canceled).
Detroit Mercy, Pitt won by +16.
Cornell +44.
Kent State +9.
PURDUE, LOST BY -13.
Duquesne +21.
Central Arkansas +53.
Eastern Washington +33.
Morehead State +10.
upcoming:
Davidson.
Western Carolina.
Maryland-Eastern Shore.
Pitt has an RPI of 19 and a SOS of 23.
Pitt scheduled two tough games, losing vs Purdue, and vs Gonzaga (not Pitt's fault that it was canceled in Japan).
Pitt's best win so far is either Duquesne or Kent State. Davidson upcoming is solid. Pitt along with Cincinnati I think I remember were known for weak OOC schedules... around the time Pitt lost to Wagner and Cincinnati lost to Presbyterian...

LOUISVILLE is also 7-1, and OOC this year has played/ will play:
Samford +41.
Hartford +35.
North Florida +38.
St Francis of NY +44.
Saint Louis +20
MICHIGAN STATE, LOST BY -4.
Grand Canyon +48.
Eastern Michigan +33.
upcoming:
Kennesaw State.
Western Kentucky.
UMKC.
Utah Valley.
KENTUCKY.
Louisville has an RPI of 13, and a SOS of 58.
Louisville scheduled 2 tough games, losing vs Michigan State, and probably losing vs Kentucky.
Other than that, Pitino is doing his Savannah State thing, mercilessly beating the crap out of Grand Canyon and Samford.
Louisville's best win so far is either North Florida or Saint Louis.

Ratings systems that reward margin of victory must love Pitt and Louisville, though somehow their RPI's and SOS's are great even with such cupcake schedules.

Not sure what I want to do in this thread, just wanted to get this down on paper, and give props to the Big East as a whole for having respectable schedules. I think only Marquette scheduled more than a couple of weak teams, but even still MU has played Wisconsin, Iowa, Arizona State and LSU.

Louisville was ranked last week, and I assume they will climb a little today as well. We really have no idea how good or bad they are. In the past week, they've played one game, their 33 point blowout of Eastern Michigan (RPI 65 somehow with wins over Vermont and Omaha)... that followed up their 48 point win vs Grand Canyon (RPI 191 somehow with 4 wins vs teams with 311+ RPI plus Hampton and Southern)... hey, Pitino can't help if his opponents are so bad, right... is he gaming the system with his scheduling...?

[anXUfan]

I'd like to think these weak schedules will come back to haunt come Selection Sunday.

[ArmyVet]

I'd like to think these weak schedules will come back to haunt come Selection Sunday.

They should but what (I think) billyjack is pointing out is that the SOS is actually quite good. No idea how but it is.

[XUFan09]

Having a top 10 team on your schedule, especially on the road, heavily influences your SOS, particularly with only a third of the schedule played so far. In Pitt's case, a number of those weak opponents are also just bad or mediocre rather than God-awful, which helps. In Louisville's case, losing by only 3 at Michigan State is impressive.

The most important thing to keep in mind is that it's still early, so there are going to be a lot of fluctuations. The various algorithms are based on the assumption of a much larger sample size than we have so far. I made this point in the other thread too.

Also, I've noticed in basketball that some teams are just really good at blowing out weak opponents and where they get tested is down the road when they face a solid team. It will balance out in the end.

[NovaBall]

Here is probably why:

RPI is 25% your weighted win%, 25% your opp win%, and 50% you opp opp win%.

So sometimes the lower conference teams that play tough OOC schedules look like RPI gold prior to conference play. Teams like Western Carolina and Incarnate World have awesome opp opp win%. But this will settle down once conference play starts.

Where those schedules will hurt a team like Pitt will be in the nitty gritty report where they are sitting with double digit wins against RPI 200+ teams.

This is why I chuckle when people like Stever get all jazzed up about the RPI prior to January. They clearly don't realize that it means nothing for teams prior to some sample of conference play being involved.

[stever20]

Here is probably why:

RPI is 25% your weighted win%, 25% your opp win%, and 50% you opp opp win%.

So sometimes the lower conference teams that play tough OOC schedules look like RPI gold prior to conference play. Teams like Western Carolina and Incarnate World have awesome opp opp win%. But this will settle down once conference play starts.

Where those schedules will hurt a team like Pitt will be in the nitty gritty report where they are sitting with double digit wins against RPI 200+ teams.

This is why I chuckle when people like Stever get all jazzed up about the RPI prior to January. They clearly don't realize that it means nothing for teams prior to some sample of conference play being involved.

It's actually opposite. RPI is 50% your opp win %, and 25 % your opp opp win %. So if you play like a Kent St and they wind up 25-6, that's a HUGE help for Pitt.

Pitt won't have double digit wins against RPI 200+ teams. They may actually wind up with only like 2 or 3.

I think to Billyjacks question- I think part of it is that SOS also counts the game lost. So for Pitt, that's a 11-0 team added. Michigan St undefeated as well. The loss counts just as much in the SOS as all of the other wins. Also for both Pitt and Louisville- the game they lost was on the road. So only .6 of a loss in calculating the w/l record.

[XUFan09]

At this point in the season, RPI Forecast has become pretty good. It projects wins and losses based off Sagarin's ratings, so there's a margin of error, but it's more accurate than current RPI/SOS. Pittsburgh has a projected noncon SOS of 109, which includes future opponents like Davidson. Louisville has a projected noncon SOS of 88, which includes future opponents like Kentucky.

The major MOV metrics like Sagarin and Kenpom still include their preseason weight, though it's not drastic now. Those weighty will be removed in the next couple of weeks now that there will be enough game data to stand on its own.