Other scores

Geezaldinho wrote:BYU dug themselves a hole again.

They are down 0-1 to Long Beach State in the 14th minute.

And the are down  0-1 at Final Time to suffer a loss.  Although it would help the Pilots for BYU to have won, this does not break my heart.

Me neither. it moved their RPI from just ahead of us at 18,  to 27, one step above LBS... still high enough to be a valuable opponent, but in their rightful place if you are a UP fan.

Still a bit of a shocker, though.

Stanford up on SCU in the second minute.

ESPNU , If you want to look.

1-1 at the half both defenses are lucky the damage isn't worse.

2-2 at end of regulation time. OT just started.

does anybody think that was a foul that got the winning free kick?

Terrible call. The PK was legit, but the foul leading to the game-winning goal was not even close to a foul. SCU got robbed.

In the most interesting game of the day, however, Oklahoma State beat North Texas in OT. 52 fouls. 9 yellows (including 3 team yellows) and two reds (including a team red).

In an interesting result from Friday where S. Miss over a lowly SWAC team Alcorn St.   They won 10-1 but were offsides something like 16 times.

That last foul was a really bad call. On the PK penalty, silly and unnecessary foul.

Stanford

Does anybody think that was a foul on the free kick?

"As she took the kick, Walker-Hartshorn sprinted out of the way and Uhl essentially boxed out the four-player Santa Clara wall, leaning against them to create a gap that Verloo spun the ball through. Tostanoski, possibly screened, did not dive as the ball edged inside the post, giving Stanford the victory, with 2:02 to spare"

Boxed out?

I asked your first question to open the discussion.

But for the restart, It is perfectly legitimate to screen other players, as long as you don't commit a foul. All players are entitled to their position. You can push on another player as long as you don't displace their position, in which case, they are pushing back.

Santa Clara fans need to let this one go. I  have seen Claims for everything from blocking  to handling by Verloo because she repositioned the ball after the whistle but before the kick. There was even a hand ball claim on the play before the foul ( a similar handling occurred in our game on Stanford's goal-- Rosen, I think)

None of these claims seem legitimate and whatever shoving might have taken place seem within accepted norms in modern soccer. Santa Clara did a little shoving themselves.

As to the foul that led to the kick, I find that hard to justify.

The PK foul seemed legitimate, though I will add that a very similar call was not made when Micaela Capelle was shoved to the ground from behind  in the box in the SDSU game this weekend. She was even dribbling towards the center of the box.

You get some calls and don't get others. Maybe Jerry needs to work more on free kick defense. Defenders on the post, for example, would have saved both free kick goals.

Boston College 1-0 Wake Forest, about the 70th.

Score holds, BC beats WF.

In Holy Cross news, ND is up 5-0 on Maryland.

Another interesting one: Denver scores in the 88th to win 1-0 at BYU.

UO v. WSU 0-0

We got a nice win against a pretty good team last weekend, and it drove our RPI rank down considerably. nc-soccer.com has those nifty future projection graphs, and it shows that winning from here on won't do a lot for our RPI. Most of the teams currently ahead of us have rosier projections. So there's no way for us to end up in the top 16, despite our good record.

I think RPI depends way too much on two things that don't reflect a team's own performance. One of them they're stuck with: their conference. And the other is a crapshoot: whether the non-conference teams you schedule do as well this year as they did in the past. Virginia and Marquette were both top-25 teams last year, but getting stomped by Virginia helped Santa Clara's RPI more than beating Marquette helped ours. Most of our other non-conf opponents—Central MI, Missouri, SDState, even UO—have worse records this year than they did over recent years. That's what's sinking us.

To me, the 25-50-25 weighting of the three elements in RPI is just wrong, intuitively. (As well as looking like it may have been chosen specifically to favor a certain conference and a certain region). (Those are the correct numbers, right?) 40-30-30 looks more plausible. I mean, a better fit with how a rational person would evaluate a team's record. What do you say, UPSF? Or anyone?

The rosier projections for other teams depend on them winning their games, which teams in some conferences have not so great a chance of doing.
It is starting to look like only 7 ACC teams might be at .500 once they knock each other off. It doesn't matter how good your RPI is if you don't make that mark.

Wakes projection when they were #1 in the RPI looked great.

Don't even look to see where the  BYU projection was and where it is now.

dwm wrote:We got a nice win against a pretty good team last weekend, and it drove our RPI rank down considerably. nc-soccer.com has those nifty future projection graphs, and it shows that winning from here on won't do a lot for our RPI. Most of the teams currently ahead of us have rosier projections. So there's no way for us to end up in the top 16, despite our good record.

I think RPI depends way too much on two things that don't reflect a team's own performance. One of them they're stuck with: their conference. And the other is a crapshoot: whether the non-conference teams you schedule do as well this year as they did in the past. Virginia and Marquette were both top-25 teams last year, but getting stomped by Virginia helped Santa Clara's RPI more than beating Marquette helped ours. Most of our other non-conf opponents—Central MI, Missouri, SDState, even UO—have worse records this year than they did over recent years. That's what's sinking us.

To me, the 25-50-25 weighting of the three elements in RPI is just wrong, intuitively. (As well as looking like it may have been chosen specifically to favor a certain conference and a certain region). (Those are the correct numbers, right?) 40-30-30 looks more plausible. I mean, a better fit with how a rational person would evaluate a team's record. What do you say, UPSF? Or anyone?

Although the 25 (winning record) - 50 (average of opponents' winning records) - 25 (average of opponents' opponents' winning records) looks odd at first glance, it isn't. There is a much greater spread between what individual teams' winning records can be (wins plus 1/2 ties, with the sum divided by games payed) than there is between what the average of opponents' winning records can be and even greater than there is between what the average of opponents' opponents' winning records can be. When you take these different spreads into account, the effective ration is roughly 50% winning record, 40% opponents' winning records, 10% opponents' opponents' winning records -- or, more simply, 50% on our record and 50% on our strength of schedule.

For this and much more, all pretty much up to date, go to the RPI for Division I Women's Soccer website here: https://sites.google.com/site/rpifordivisioniwomenssoccer/

It addresses this exact question on the RPI: Formula page under the heading Computing the RPI, subheading Calculation of RPI, where it provides the exact effective ratios for a couple of years.

Geezaldinho wrote:The rosier projections for other teams depend on them winning their games, which teams in some conferences have not so great a chance of doing.
 It is starting to look like only 7 ACC teams might be at .500 once they knock each other off. It doesn't matter how good your RPI is if you don't make that mark.

Wakes projection when they were #1 in the RPI looked great.

Don't even look to see where the  BYU projection was and where it is now.

As the Geez points out, the very neat "projection" gizmo at nc-soccer assumes that the team wins all the rest of its games. Also, it assumes that the team's opponents' winning percentages will stay as they are now. The team may or may not win the rest of its games, but all of them certainly won't. And, most opponents' winning percentages will change. Now that conference play has started, this means that how teams are likely to do in their conferences is critical. Thus, for example, there's a chance the Marquette will turn out to have been a better opponent than its current record indicates.

In scheduling, teams in the running for the NCAA tournament have to do the best they can both in terms of teams that will be good for their RPIs and teams that, beyond RPI effects, will be good opponents under the other tournament at large selection and seeding criteria. There are data available on the nc-soccer website that coaches can use to help them figure out what are good teams to schedule. Also, the RPI for Division I Women's Soccer has a page, NCAA Tournament: Scheduling Towards the Tournament, that is intended as a coaches' scheduling resource with information on the important factors to consider and on how to use the nc-soccer resource.

Finally, for those really interested in using the nc-soccer resources, there is a manual for using its webpages posted on the RPI for Division I Women's Soccer's RPI: This Year's Reports page. The manual is an attachment at the bottom of that page.

Coaches always have to take some risks in scheduling. The Scheduling Towards the Tournament webpage discusses this. The nc-soccer resources include some that can help coaches in evaluating the risks. Other things that can upset the scheduling apple cart are events like coaching changes. Central Michigan, for example, was looking like a great team to schedule. Then, their coach got hired away last January -- by Cincinnati, I think. This most likely has had a very significant effect on them. No doubt the Pilots scheduled them well before this occurred. They still may do well in their conference and improve what they do for the Pilots' RPI, but it's the kind of situation that is an inherent risk in the scheduling process.

Portland State def Northern Colorado today 3-1. From the stats, it looks like the Vikings were much the better team. Good win, since NC was picked by some to be one of the top teams in the Big Sky.

UPSoccerFanatic wrote:
Although the 25 (winning record) - 50 (average of opponents' winning records) - 25 (average of opponents' opponents' winning records) looks odd at first glance, it isn't.  There is a much greater spread between what individual teams' winning records can be (wins plus 1/2 ties, with the sum divided by games payed) than there is between what the average of opponents' winning records can be and even greater than there is between what the average of opponents' opponents' winning records can be.  When you take these different spreads into account, the effective ration is roughly 50% winning record, 40% opponents' winning records, 10% opponents' opponents' winning records -- or, more simply, 50% on our record and 50% on our strength of schedule.

If you look at these bubble graphs, http://soccer.moulliet.com/rpiBubble.html , which show each team's contribution to a single team's RPI, it looks like less than 50 / 50. For example, in 2012, Stanford's own record contributed to 34% of their unadjusted RPI (this includes both Element 1 and Element 3 contributions). For Portland, it was 29%.

gxm,  how are the percentages in the different bubbles determined?  For Stanford 2013 I can figure out which bubbles are Elements 1, 2, and 3, but I don't know how the percentages are calculated.  I don't know if I could figure out which of the 2012 bubbles are Element 2 or 3, as I haven't tried, but I'd have the same question there.

Also, what would happen if you averaged out all teams' 2012 Element 1, Element 2, and Element 3 contributions?

EDIT:  I woke up this morning and had the answers to my questions.  I need to find some numbers, and then I'll email you a response.

While I was humoring myself with today's rankings, I happened to notice that Wake Forest plays Virginia on October 13th. I think I may want to follow that game.

I started looking at pairings for the first round of 32 games.  UP's possible matchup could be Seattle out of WAC or PSU out of Big Sky.   WSU again is a possibility too.  

In the state of CA not too many teams and several of those likely to be seeds so they wouldn't play each other first round even if they are from different conferences.  You have the usual suspects:   Stanford and UCLA out of Pac 12 which are in seeding position.  Then you have Cal and SCU out of WCC while Pep is on bubble right now.   Finally, the Big West will likely only have one team in the mix as LB St and UC-Irving are in bubble position.  I suspect that if both UCLA and Stanford get seeds they will bring in a low RPI team (for low I mean 100+) from somewhere to play top seed in CA.

Quite a few teams should be coming out of Col/Utah area:   Could be BYU, maybe team from WAC if Seattle does not make it, Colorado out of Pac 12 is possible, as is one team from Mt West (Utah St or Boise St maybe), could be Weber St or N. Colorado out of Big Sky, and finally Colorado C. out of Conf USA looks to be in position to make it even if they don't win conference.

I work my way back East locals as we get more toward the end of the month.

You sent me looking at our opponents schedule. One game that will become huge in its consequences for seeding will be the Stanford -UCLA match a week from tomorrow.