Other thing to consider is how are ties calculated? - I believe soccer is the only sport besides hockey that has ties and hockey still has OT......

Darn great question. Wish I would have thought of it!! I'm guessing it would be a big negative and like a loss, since wins is a huge positive over SOS and tie is not a win.

Also, why does hockey get to play it out?  Heck more dangerous plus incredibly tiring. Or do only "North American" sports get a chance to play it out? Even field hockey, though not a N. American sport, has overtime.

According to Bob Quillman a D3 BB guy, "Pairwise is a done deal"

https://x.com/IWUhoopscom/status/1776444074779725918

QuoteI'm told by #d3hoops coaches who attended the congress meeting today...

1) Pairwise is a done deal.

2) But no one knows what the "dials" are set to and what metrics/results will be rewarded the most.

Coaches confused about how to schedule for future seasons.

Bianco confirms what I said earlier that the job of Regional Advisory Committees will mostly just be (1) to make sure there aren't any North Central (MN) - North Central (IL) mix-ups because the algorithm does the ranking, (2) to set the "dials" every few years for the weight given to W-L and SoS, and (3) to help work at the regional sites

https://x.com/BMBianco/status/1776454103331147937

QuoteThere will still be RAC's, but their job will be to make sure that the data going into the algorithm is correct (also, who else will site rep if no RAC's 🤨). Re: dials- once they're "set", they can't be changed for a few years.

Quote from: Kuiper on April 06, 2024, 01:03:21 AMAccording to Bob Quillman a D3 BB guy, "Pairwise is a done deal"

https://x.com/IWUhoopscom/status/1776444074779725918

QuoteI'm told by #d3hoops coaches who attended the congress meeting today...

1) Pairwise is a done deal.

2) But no one knows what the "dials" are set to and what metrics/results will be rewarded the most.

Coaches confused about how to schedule for future seasons.

Bianco confirms what I said earlier that the job of Regional Advisory Committees will mostly just be (1) to make sure there aren't any North Central (MN) - North Central (IL) mix-ups because the algorithm does the ranking, (2) to set the "dials" every few years for the weight given to W-L and SoS, and (3) to help work at the regional sites

https://x.com/BMBianco/status/1776454103331147937

QuoteThere will still be RAC's, but their job will be to make sure that the data going into the algorithm is correct (also, who else will site rep if no RAC's 🤨). Re: dials- once they're "set", they can't be changed for a few years.

Bianco kept tweeting last night and this morning and it's worth including a few more of his points.

He points out that the dials controlled by the RAC don't really change much.  He doesn't say why, but I'm assuming that's because as long as bad wins are removed from SoS, then the difference between a team with a strong record in a schedule of murders' row opponents and a team with a strong record in a weak schedule with a few strong opponents, is not that much since SoS doesn't adjust based on number of games counted for SoS purposes on your schedule. So, it you win all of the equally high SoS games on your schedule, you have a "perfect" SoS whether 10 games count (in a difficult schedule) or only 2 games count (in an otherwise weaker schedule).  For W&L, part of the problem is likely the 6 ties, but part is that their "bad loss" to Roanoke, which finished 7-10-1, would be included in their SoS calculation.

https://x.com/BMBianco/status/1776596646224167075

Quotethe "dials" change things very marginally, unfortunately. For example, W&L soccer (FF team) wouldn't have made the NCAA's with Pairwise despite any dial you tweak.

Bianco also asserts that 1 (or maybe 2) national champions in the past 10 years would not even have made the tournament.  If I had to guess, it would be Tufts in 2016, which was 9-5-2 when the conference tournament finished and selections were made. Other possibility might be Conn College in 2021, which was 14-4 when selections were made.

https://x.com/BMBianco/status/1776606497104949575

QuoteMen's soccer committee asked for 10 years of data. They were given 3. I can almost guarantee a national champion (possibly two) in the last 10 years for men's soccer wouldn't have made the tournament using Pairwise/NPI.

Now there's a rumor that "several dozen" schools are going to press for delayed implementation until 2025-2026.  No idea whether that would have any effect. 

https://x.com/IWUhoopscom/status/1776687316737651177

QuoteI'm told a group of "several dozen" D3 schools are going to jointly press to delay the implementation of NPI (Pairwise) to 2025-26. Objective - to get the details right & build buy-in.

Another twitter account says the change is being pushed through by smaller schools/conferences that feel it will provide more equity.

https://x.com/davemchugh/status/1773879129777271055

QuoteIt's being forced through as I understand it by those on the former cmte who felt "small" schools and conferences were being disenfranchised. There. I said what others won't say.

McHugh is directly calling out the commissioner of the Empire 8 conference in that tweet. The E8 commissioner was formerly on the championships committee and McHugh claims that he's doing this because E8 men's basketball teams got left out of the tournament in 21/22 and 22/23 despite winning 21+ games and this is his parting gift.

I have a lot of things to say about this, and I'm not sure I should say them quite yet before carefully thinking about it first.

The D3 Management Council has approved the new selection criteria starting this coming season

https://www.ncaa.org/news/2024/4/18/media-center-diii-management-council-supports-expanding-football-team-sport-brackets.aspx

QuoteAdditionally, the council approved the use of the NCAA Power Index for selection of at-large berths to team sports championships. The changes will go into effect starting with 2024 fall championships.

Coach Bianco was miffed about the way the DIII Council announced the change in its press release, implying that it was trying to hide the change amidst other changes it announced.

https://x.com/BMBianco/status/1781091102298095682

Quote"ADDITIONALLY"?!? Like this is some throw away change. This is THE CHANGE!

It appears that the next step is approval by the DIII Presidents Council, but those in the know say that the President Council doesn't tend to go against the Management Council on these matters.

https://x.com/d3hoopsville/status/1781007706003775497

QuotePairWise on to President's Council.

It's interesting as this progresses how many more people within #NCAAD3 are voicing frustration w/lack of info & clarity about it.

PC doesn't tend to vote against MC. Not sure there's enough pressure to do so. Train wreck coming?

Although the President's Council still has to rule on this as I understand it, I suspect they are preoccupied with other things and despite coaches caring a lot about this, it isn't a politically controversial enough issue for the Presidents to decide to overrule their Athletic Directors and other reps on the Management Council.  So, what would be the real effect after a year or two of transition?  I'll speculate a little:

1.  Higher SoS is still the safest bet for tournament selection (and success in the tournament)

Since they will reportedly drop "bad wins" against weak opponents, many have assumed you can schedule weaker opponents.  However, you still need a couple of strong opponent wins (or maybe really strong opponent ties/losses) to get your SoS level to a level that puts you in contention without the "bad wins."  It's not clear exactly what you need, though, but that's not really different than the current system.  Under the old system, coaches were always guessing, many times years in advance, about opponent strength when they set their non-conference schedules.  We all know schools that scheduled really strong opponents that turned out to have bad seasons and became "bad wins/losses."  That means that if you want to be sure, you still need to schedule a mix of historically strong opponents, hope some of them are actually strong that year, and win those games.   

All that has really changed is that coaches may not understand how far down you can go for a win to be dropped as a "bad win."  Under the new system, coaches would love to schedule teams that they are (1) sure to beat, but (2) will be dropped so it won't hurt their SoS.  Good luck with that.  Even with years of experience with the system, that is going to be difficult to predict except for the very worst programs.  Moreover, under both the old system and the new system, the scheduling incentives of the awful teams haven't changed.  They would like to schedule opponents where their teams can have somewhat competitive games and they aren't going to travel far to get slaughtered.  That means patsies aren't going to have much incentives to agree to a game against a non-local strong opponent.  The bottom line is that not knowing the exact definition of a "bad win" means that because of the incentives for both teams in a scheduling decision, the strong team is going to end up with a mix of opponents that looks similar to the way they do now for the most part.

2.  Lower priority for travel to play highly ranked teams

The big exception to the first point is that, given the uncertainty, teams will have less incentive to schedule expensive trips in search of higher SoS.  Some of the far flung conferences in Region X may still need to do it just to get enough games, but the market will be smaller for high travel games.  That's probably good for everyone because the cost and classes missed are problematic.

3.  Conference will matter more (but coaches won't have much say in that)

I haven't heard anyone discuss this, but this will affect conference affiliation.  It's probably better to be in a conference with a few great teams and a bunch of easy wins, than an evenly matched bunch of decent teams. Coaches of individual sports can't affect that for the most part though.

Did we ever get any confirmation on whether the NPI is going to actually drop bad wins or is this still based on a kneejerk reaction from a coach who saw PairWise and made assumptions that may not even be relevant to soccer?

Maybe the NCAA will attempt to reinvent the wheel in a dumb way with the new ranking system for D3 but they already had an adjusted RPI system in place for years at the D1 level.  It would make sense for the D3 system to closely mirror that.  In the D1 system, teams would receive scaling bonuses for wins or ties against top ranked teams and penalties for losses or ties against low ranked teams.  Every game still counts for win% and SOS purposes.   

Quote from: paclassic89 on April 19, 2024, 03:42:00 PMDid we ever get any confirmation on whether the NPI is going to actually drop bad wins or is this still based on a kneejerk reaction from a coach who saw PairWise and made assumptions that may not even be relevant to soccer?

Maybe the NCAA will attempt to reinvent the wheel in a dumb way with the new ranking system for D3 but they already had an adjusted RPI system in place for years at the D1 level.  It would make sense for the D3 system to closely mirror that.  In the D1 system, teams would receive scaling bonuses for wins or ties against top ranked teams and penalties for losses or ties against low ranked teams.  Every game still counts for win% and SOS purposes.   

According to the FAQ section the Championships Committee released as part of its recommendation for using the NPI, they specifically mentioned one of the benefits of the system was that it dropped "bad wins," so I think it's a fair assumption that it at least does that.  What we don't know is where the cutoff line is drawn.

https://ncaaorg.s3.amazonaws.com/committees/d3/champs/D3CC_SelectionCriteriaDatabaseFAQ.pdf

QuoteQ: What are the benefits of the selection criteria database?

A: One benefit of the selection criteria database is that all data will be analyzed, and the criteria
will be applied objectively to establish the championships field (the at-large berths remaining
after AQs are awarded).

Another benefit of the database is that teams are not penalized for winning a contest that it
should win. For example, in the current system if a strong team plays a weaker team its
strength-of-schedule will be negatively impacted even though it won the contest – as it should.
In the selection criteria database, the team that wins a contest against a lesser opponent will
not be statistically impacted by winning a contest it should win. This will greatly benefit
stronger teams in weaker conferences

Thanks for that.  I guess i'll withhold judgement until we get more data.  I'm inclined to give the NCAA the benefit of the doubt and say the selection impacts will probably effect what we would consider bubble teams rather than some egregious change where a W&L type of team (easily top 25 in RPI last year) (I think they were brought up as missing out under this potential system??) gets left out.   I'm firmly on the NPI side and have been beating the algorithmic ranking drum for years. There's no reason to have regional ranking committees eyeballing RvR's and SoS in 2024.  It's a waste of time and resources

Let's be honest, every year some teams get screwed and that's not going to change with the new format. And furthermore, any given year any team can win. You don't need to be a "top 10" program to have a shot.   

D3 Datacast, a basketball podcast and blog, did an update on the NPI/Pairwise process.  It has enough inside scoop that it is worth breaking down in some detail, but you should read the whole thing so they get the credit for doing the work (for whatever that is worth to them).

What We've learned about the Coming of NPI

Here's the background to get people up to speed

QuoteThe winds of change have been blowing throughout Division III, including in the way teams will be ranked and selected for national tournament play. As most are likely aware, the Championships Committee has investigated and recommended the implementation of a new ranking and selection system, known as the NCAA Power Index (NPI), sometimes also referred to as "Pairwise" or the Selection Criteria Database.

That recommendation, with an effective date of the 2024-25 season, was subsequently approved in concept by the Division III Management Council and President's Council. While many in the Division III community have known about the ongoing process to move toward NPI there seems to be a great deal of consternation and confusion over its details and implications. While certain details are still being set and the full implications of the new system may not be known until it has been in place for at least a year, I wanted to put together this blog post to lay out some information that we have been able to gather.

The author notes that there has been some opposition trying to slow things down and it identified the principal groups who are complaining.

QuoteWhile change is sure to come with a dose of low-level grumbling there have been some entities that have logged formal concerns or at least expressed a desire to have the process slowed down or delayed a year so there can be greater understanding among the membership before NPI is put into place. That list includes at least the North Coast Athletic Conference, the Liberty League men's and women's soccer coaches, NESCAC men's soccer coaches and the United Soccer Coaches Division III Advisory Council.

Notice a common theme among these critics?  Yep.  They are basically all from soccer and you can bet that the NCAC concerns are being driven by the soccer coaches since Denison men's soccer HC Brandon Bianco has been leading the charge to slow this down ever since word leaked out about the proposals.

Does that mean there's a chance it won't happen for 2024?  The author doesn't think so, in part because all of the opposition has come from soccer.  It sounds like the soccer coaches either haven't gotten the other sports on board or they haven't figured out this politics thing where you need allies to push through or block changes (to be fair, soccer may be uniquely impacted because they are a fall sport, whereas winter or spring sports have some time to alter their schedules, and some other fall sports like football are just trying to find opponents (see former ASC schools) and may just not be as organized from a coach's perspective as soccer).

QuoteSo what are the odds the Management Council slams on the breaks and defers for a year? At this point that seems unlikely given that it appears the bulk of the comments/complaints/concerns have all come from one sport: soccer. With NPI being a division-wide initiative, I have a hard time seeing one sport being able to mount an effort to derail what looks to me to be more of a certainty.

This is especially true based on what we've gathered based on the response from sport committees. If there was a group of people who were best positioned to sidetrack the NPI process it would have come here but responses have been generally positive. Level of support varied, particularly when it came to implementing for Fall 2024 instead of holding off until Fall 2025, but it appears only the men's soccer committee was willing to go further and voice an opinion that the system is not ready to be implemented.

Once again, is a single sport going to be enough to hold off NPI for at least another year?

So what are the next steps?  Apparently, the "dial" settings for soccer, as a fall sport, should have already been submitted because they were due by June 1 and will be posted by June 15. Then, the Management Council will formally approve the NPI in its final form.

QuoteThe next part of the process following the approval of the recommendation to proceed with NPI was for the individual sport committees to consider and submit their recommended "dial" settings. The so called "dials" allow sport committees to tune relative weights of components such as winning percentage, strength of schedule, home/away multipliers, etc. that feed into the NPI calculation. The dials will be set by each committee, not by sports. This means we could see different dial settings in men's and women's basketball. This is not unlike what we have had in the past were the strength of schedule calculation for men's basketball includes a home/away multiplier while the women's basketball SOS does not.

Dials for Fall sports were due to be set by June 1 with a posting date set for June 15. Winter sports face a June 15 date for dial settings and a posting date of July 1 while Spring sports will have their dials set by July 1 to be posted July 15.

This all leads to one big final decision point: the Management Council meeting in July. It is at this meeting that the Management Council is to review and approve the NPI in its final form, taking the necessary legislative action to officially put it into place.

They also will supposedly have educational webinars for coaches in July.  Of course, that will be too late to change anyone's schedule for Fall 2024, but at least they might start to learn how individual games can affect their ranking/selection, which could inform their decisions on how they prioritize games for player selection/rotation/substitution purposes etc.

QuoteIt may be "too little, too late" for some coaches and administrators but for those willing to accept "better late than never" it seems more information is not far off. The plan is to hold educational webinars in July for Fall sports, late July/early August for Winter sports and late August/early September for Spring sports.

More from Coach Bianco about NPI

https://x.com/bmbianco/status/1799144274128163193?s=46&t=uXiupHZfR0TxrRyWF9BmRg

"NPI has always felt from the start as a "done deal" by those inside and outside the room. Now that more info is out, there's a lot to digest. But make no mistake, there was an agenda here to push this through. At least @NCAADIII said the quiet part out loud."

(He's referring here to the fact that this gives second place teams in weak conferences a pathway to qualification over middling teams in really strong conferences with high SoS)

The DIII Championships Committee has set the "Dials" for the Fall 2024 sports

https://ncaaorg.s3.amazonaws.com/committees/d3/champs/2024D3CC_NPIWeights.pdf

Men's soccer is the following

Win%/SoS

15%/85% SOS

Home/Away Win-Loss
 
equal weight for home and away wins/losses

Quality win Bonus

54.0 (e.g., bonuses for all wins/partial wins against teams with an NPI of 54 or higher)

QWB Multiplier

.750 (default is .500 - meaning soccer has a higher multiplier before it is a quality win bonus)

Overtime

100/0

Minimum Wins

10

I read this as basically the Men's Soccer Committee setting dials as far toward SoS as they could (since 25/75 is the default between Wins/SoS). Still have to have 10 wins to receive the NPI adjustment, which means it has a high bar for when wins only can help. (I could be wrong on my interpretation though)

First few parts of the Q&A

Q: What is the NPI?
A: The NPI is a system that houses the division's statistical data and objectively applies the
selection criteria to the data based on the sport committee's established weighted criteria. NPI
is a combination of the following: winning percentage; strength of schedule; home-away
multiplier; quality win bonus; and overtime results (when relevant).

 The weight of winning percentage versus strength of schedule is often referred to as one of the
"dials" that a sport committee can adjust, placing any subjectivity within the criteria
themselves. For example, a sport committee could set the dials to 30/70; 25/75, etc. Other
"dials" a sport committee can also choose to apply are home and away multiplier, quality win
bonus (QWB), and overtime win/loss weights.

Q: Does this mean our current selection criteria goes away?
A: Most of the current selection criteria will no longer be used. The following criteria will no
longer be used: head-to-head competition; results versus common Division III opponents;
results versus ranked Division III teams; and secondary criteria. However, winning percentage
and strength of schedule will continue to be used.