|
Post by fmpol on Mar 2, 2021 12:22:55 GMT -8
If you've followed Bracketology in the past, you already know that the women's selection committee primarily relies on two reports, the "Nitty Gritty" report which gives a data summary for all Division I teams and the "Team Sheets" which gives the details for each team's record. The problem is that the NCAA has previously linked fans to static, non-sortable lists and even worse, they haven't posted the 2020-21 season data on the NCAA.com website. Well, I finally found the data that the committee is using this season and it's on the NCAA. org website. Here is the Nitty-Gritty Report, which can be filtered and sorted however you wish. Note that you can also check off two or more teams and check "Filter Selected Teams" to do head-to-head comparisons. The committee will do these types of comparisons to sort out the bubble and decide who should be on what seeding line. If you click on an individual team, you will be taken to that team's statistical page but that is NOT the primary page used by the committee. You have to click on the "NET Ranking" link to the right of the team's name and that will take you to the Team Sheet that the committee uses. Here is a direct link to the Oregon State Team Sheet. The Team Sheet provides a lot of good news for the Beavs. First, the committee can immediately see how dramatically the team has improved - that sharp slope in their NET ranking is fairly rare and the committee will realize that they are now a much better team than their overall record would suggest. And if the committee didn't already notice this on the Nitty Gritty report, the Beavs have played the toughest schedule in the nation (note that Average Opponent NET Rank has replaced the old Strength of Schedule). These factors along with a NET of 29 and respectable records vs. NET 1-25 teams (as well as 26-50 and 51-100 teams) put the Beavs in great shape. The main blemishes are a nearly non-existent non-conference record and the loss to Utah. Now for the caveats. Unfortunately it is important how previous opponents continue to fare and due to arbitrary cutoffs, the distinctions can become very silly. It's bad enough that the loss to Utah now looks so much worse after the Utes lost their starting PG and fell out of the Top 100. But now one has to hope that Colorado can cling to the Top 50 and San Francisco can stay in the Top 100 so that those victories can retain their current value. And of course as soon as we take the floor against Cal, "the toughest schedule in the nation" will no longer be so tough, but the Average Opponent NET Rank will still remain very high, maybe #11 or #12. As for upcoming games, obviously the first consideration is to simply beat Cal, who has a terrible record but is playing noticeably better late in the season. In the old RPI days, simply playing the #260 team would knock down your RPI, but that drop won't occur in the NET if you win AND run up the score. Unfortunately, that isn't feasible in a conference tournament where a win means an 11 a.m. game the following morning. So the Beavs should just ignore the NET implications, focus on simply beating an improved Cal team, and hopefully do so comfortably enough to perhaps rest the starters later in the game even if it means a smaller margin of victory. Then let the remaining chips fall where they may. A final note: DePaul's strange loss last night to Butler (NET 276) and the Beavs strong Team Sheet did make me wonder about the possibility of edging up to a 9 seed, with the implication being that a win in the 8-9 matchup might mean facing a 1 seed in second game. (A couple of 1 seeds may resemble a 2 seed but a couple will definitely look and play like a 1 seed.) Anyway, that's looking too far ahead so the focus should be on beating Cal.
|
|
|
Post by sewingbeaver on Mar 2, 2021 14:00:40 GMT -8
Fantastic information. Thanks for sharing. Go BEAVS!
|
|
|
Post by rmancarl on Mar 2, 2021 14:01:11 GMT -8
Thanks for sharing this. I did a little research on NET the other day and I must say, I'm still baffled by NET for some teams. I'm sure it must be a good system, but sometimes I wonder...
|
|
|
Post by beaveragain on Mar 2, 2021 14:44:21 GMT -8
One of the odd things this year due to Covid. There have been a couple of studies about the home field advantage during the season of the covid, and the conclusion is that without the fans there is no home field advantage. The NET does not take this into account and seems like one of the reasons NET has the Ducks off on ranking from what seems logical.
|
|
|
Post by fmpol on Mar 2, 2021 17:02:21 GMT -8
I forgot to mention one important finding. There's been some talk about whether the Pac-12 is getting too little or too much respect as far as the number of potential selections. One of the most interesting changes in the Nitty Gritty Report this season is that conference ranking (i.e. does this team play in the strongest conference, the 8th best conference, the worst conference, or whatever) is no longer shown on the report. This is one of several indications that conference affiliation may be de-emphasized this season. (The procedures have always said that conference affiliation should not considered when selecting the field of 64, but I can guarantee you that the women's committee has consistently ignored that rule. Maybe they'll actually follow the rule this year.) Here's another indication that conference affiliation may be less important this season. Here are the Conference Rankings Report, and note that they haven't even bothered to compute an overall NET rating. They may have realized that it was silly to do so in a season with such limited non-conference matchups. But if you (or committee members) are curious about conference strength, the list can be sorted by Road NET and you can see that all the screaming that we hear on the Pac-12 Network about how we're in the best conference really isn't borne out by the rankings (though 4th isn't too bad). Anyway, what this means is the Beavs, WSU and Colorado may have to make the field on their own merit independent of the fact that they are from the "Conference of Champions", and that we shouldn't care too much about whether the Pac-12 gets more or less bids than the ACC, Big 12 or whatever.
|
|
|
Post by lotrader on Mar 2, 2021 19:09:43 GMT -8
Thanks for the information. I don't envy the selection committee's job this year. The issue with the Covid stop/start has made this a nearly impossible task. I like the NCAA Tournament, because even though I think the PAC-12 teams seeding seems to change for the worse as we approach the Big Dance, our PAC-12 teams seem to rise to the occasion and show the rest of the country that we are the best overall conference. This year might be different given all the issues teams in our conference have had with the start/stops. But I have hope that we will get 6 teams in, and, they will perform better than forecasted. I see Stanford as the best team in the country right now, and I'm hopeful that OSU continues to gain confidence and experience through the PAC-12 Tournament.
|
|
|
Post by beaveragain on Mar 2, 2021 20:00:24 GMT -8
I forgot to mention one important finding. There's been some talk about whether the Pac-12 is getting too little or too much respect as far as the number of potential selections. One of the most interesting changes in the Nitty Gritty Report this season is that conference ranking (i.e. does this team play in the strongest conference, the 8th best conference, the worst conference, or whatever) is no longer shown on the report. This is one of several indications that conference affiliation may be de-emphasized this season. (The procedures have always said that conference affiliation should not considered when selecting the field of 64, but I can guarantee you that the women's committee has consistently ignored that rule. Maybe they'll actually follow the rule this year.) Here's another indication that conference affiliation may be less important this season. Here are the Conference Rankings Report, and note that they haven't even bothered to compute an overall NET rating. They may have realized that it was silly to do so in a season with such limited non-conference matchups. But if you (or committee members) are curious about conference strength, the list can be sorted by Road NET and you can see that all the screaming that we hear on the Pac-12 Network about how we're in the best conference really isn't borne out by the rankings (though 4th isn't too bad). Anyway, what this means is the Beavs, WSU and Colorado may have to make the field on their own merit independent of the fact that they are from the "Conference of Champions", and that we shouldn't care too much about whether the Pac-12 gets more or less bids than the ACC, Big 12 or whatever. When Louisville's coach was talking about needing to replace a game on his schedule he mentioned that there were 14 Div. I teams within an hour's drive of him that he was drumming up a game with. When OSU needed a game to fill in they had to get a Div. II team from Montana. In the best of times comparing Road NET is unfair to the west coast, but in the year of the Covid it's apple's and Moon's.
|
|
|
Post by sweetpea on Mar 3, 2021 14:28:29 GMT -8
“And if the committee didn't already notice this on the Nitty Gritty report, the Beavs have played the toughest schedule in the nation (note that Average Opponent NET Rank has replaced the old Strength of Schedule).”
Is “Average opponent NET Rank” better than RPI strength of schedule or just a different algorithm? RealTimeRPI shows Beavs as 41st in SOS.
|
|
|
Post by beaverstever on Mar 3, 2021 16:03:12 GMT -8
|
|
|
Post by bennyskid on Mar 3, 2021 16:09:35 GMT -8
RPI SoS is pretty much garbage. RPI only looks at opponent raw records and opponent-opponent records, so it values a game against, say, Mudpuddle City College (undefeated champions of the Cornfield Conference) more than a win over a decent major conference teams. You get the SoS by simply adding up the opponents win-loss records, taking the percent and multiplying by two, and then adding the opponents-opponents records, taking the percent, and adding the two percents together. When RPI was the measure, the trick to scheduling was to get as many games as possible against the BEST teams from the WORST conferences.
|
|
|
Post by fmpol on Mar 3, 2021 19:38:45 GMT -8
“And if the committee didn't already notice this on the Nitty Gritty report, the Beavs have played the toughest schedule in the nation (note that Average Opponent NET Rank has replaced the old Strength of Schedule).” Is “Average opponent NET Rank” better than RPI strength of schedule or just a different algorithm? RealTimeRPI shows Beavs as 41st in SOS. RPI is no longer used by the women's Selection Committee, so any measure derived from RPI (such as the old SOS that was based on RPI) would also be ignored. If you look at the factors used by the committee, the purely statistical factors are all included on the Nitty Gritty report and Team Sheets, so those are the only data sources that you need to look at if you want to guess at how they might be deliberating. I did notice that on tonight's broadcast, the Pac-12 network was still erroneously using the old SOS when assessing WSU's profile as a potential at-large team. That's typically pathetic of the network, but in fairness the NCAA has done a terrible job of publicizing the location of the selection data. On the page linked above, their link to the NET rankings had always led to the simple numerical ranking on NCAA.com. I only realized that they had corrected the link after I had found the Nitty Gritty on my own at NCAA.org. As to whether Average Opponent NET Rank is a better measure than SOS, I think any ranking based on the NET is better than a ranking based on RPI, but you could probably find hundreds of articles explaining why the still-imperfect NET is far better than RPI.
|
|
|
Post by fmpol on Mar 4, 2021 9:42:03 GMT -8
Here's a bracketology update after the Cal game. Note: The Nitty Gritty link in my original post is a static link to the 3/1/21 update. To reach the most recent update, either use the pull-down menu at the top (to the right of "Thru Games") to scroll up to the most recent date available, or go to the NCAA Nitty Gritties Index which will always have the link to the latest update for women's basketball. As expected, the NET dropped after a narrow win over Cal so the Beavs are now at #36 (that's still very good). Their strength of schedule (i.e. Average Opponent NET Rank) is no longer the best but is still #9 (excellent). The worst development (and this is where bracketology gets silly) is that Colorado's loss drops the Buffs down to #54, so our record vs. 26-50 teams goes from 2-2 to 0-2 (that looks bad but not disastrous when the committee looks just at the Nitty Gritty). ASU and San Francisco remain in the Top 100 so our record vs. 51-100 teams goes from 3-0 to 5-0 with the addition of Colorado. Overall the Beavs are still in good shape (but I am still pissed about the Colorado game). Today's game vs. Oregon will of course improve our strength of schedule and cannot hurt NET ranking too much barring a blowout loss. But since we are stuck with an 0-2 record vs. 26-50 teams, the record vs. 1-25 teams seems a bit more important. A win today even coupled with a loss tomorrow yields a 3-4 record which would be outstanding especially when the committee sees that most of the competition came vs. the top half of the top 25.
|
|
|
Post by fmpol on Mar 6, 2021 8:22:06 GMT -8
Here's the final update for this thread now that the Beavs have completed their NCAA resume after the Stanford game. (Be sure to see my 3/4/21 post on how to find the most recent Nitty Gritty.)
NET ranking has dropped to 35 (blowout losses even to the #1 team will do that) but that is still a good ranking. Their strength of schedule (Average Opponent NET rank) has risen to #5 which is excellent.
The committee uses the Nitty Gritty report as a starting point and there is a big focus on the columns of record vs. top 25 teams, 26-50, 51-100, and 101+. On that basis alone, I think the Beavs will have a slight chance at a 6 seed but is much more likely to be a 7 or 8 seed. There are too many teams still playing to be definitive but right now I don't see much likelihood of moving outside that range.
The ESPN bracketologist has said (paraphrasing) that the committee may penalize the Beavs for not playing enough games and that is why he is still warily putting them at a #9 seed. I believe he is basing this opinion solely on a comment that the committee chair made after the first Top 16 reveal on 2/15/21 re: Michigan and South Florida, two teams who had only played 12 games at the time and had each only played 2 games vs. top 50 teams. The chair said there wasn't enough evidence to put them in consideration for a higher seed. That lack of evidence hardly applies to the Beavs, who now have 17 games vs. Division I opponents and have 9 games vs. top 50 teams. That's more than enough evidence and better evidence than many top mid-majors will have, so I believe (and hope) that the committee will judge the Beavs straight up and without any extraneous consideration such as total games played.
Finally, I've noticed that the columns on NET Non-Conference performance includes every game outside the regular conference season, so it stupidly counts the conference tournaments as non-conference games. That effectively negates any meaningful consideration of non-conference record, which is fine this season given the problems with covid scheduling. And this is good for the Beavs, as their skimpy one-game non-conference Division I schedule is no longer a blemish on their resume. Their only blemishes now are the 0-2 record vs. 26-50 teams (that's not a weakness but more an oddity that might raise some scrutiny) and the loss to #113 Utah. Hopefully San Francisco will do well in the WCC tournament so that they remain in the Top 100 (every little resume point counts).
|
|
|
Post by beavs6 on Mar 6, 2021 8:41:09 GMT -8
Here's the final update for this thread now that the Beavs have completed their NCAA resume after the Stanford game. (Be sure to see my 3/4/21 post on how to find the most recent Nitty Gritty.) NET ranking has dropped to 35 (blowout losses even to the #1 team will do that) but that is still a good ranking. Their strength of schedule (Average Opponent NET rank) has risen to #5 which is excellent. The committee uses the Nitty Gritty report as a starting point and there is a big focus on the columns of record vs. top 25 teams, 26-50, 51-100, and 101+. On that basis alone, I think the Beavs will have a slight chance at a 6 seed but is much more likely to be a 7 or 8 seed. There are too many teams still playing to be definitive but right now I don't see much likelihood of moving outside that range. The ESPN bracketologist has said (paraphrasing) that the committee may penalize the Beavs for not playing enough games and that is why he is still warily putting them at a #9 seed. I believe he is basing this opinion solely on a comment that the committee chair made after the first Top 16 reveal on 2/15/21 re: Michigan and South Florida, two teams who had only played 12 games at the time and had each only played 2 games vs. top 50 teams. The chair said there wasn't enough evidence to put them in consideration for a higher seed. That lack of evidence hardly applies to the Beavs, who now have 17 games vs. Division I opponents and have 9 games vs. top 50 teams. That's more than enough evidence and better evidence than many top mid-majors will have, so I believe (and hope) that the committee will judge the Beavs straight up and without any extraneous consideration such as total games played. Finally, I've noticed that the columns on NET Non-Conference performance includes every game outside the regular conference season, so it stupidly counts the conference tournaments as non-conference games. That effectively negates any meaningful consideration of non-conference record, which is fine this season given the problems with covid scheduling. And this is good for the Beavs, as their skimpy one-game non-conference Division I schedule is no longer a blemish on their resume. Their only blemishes now are the 0-2 record vs. 26-50 teams (that's not a weakness but more an oddity that might raise some scrutiny) and the loss to #113 Utah. Hopefully San Francisco will do well in the WCC tournament so that they remain in the Top 100 (every little resume point counts). I hope your analysis is spot on. If we get a 7 seed or better I cannot see OSU not making the sweet 16. When that occurs, then let the chips fall where they may. Thanks for all of your work and sharing it with us. Go Beavs!
|
|
|
Post by grumpybeaver on Mar 6, 2021 9:33:02 GMT -8
|
|