SOS/QWI Thursday Filler

No matches today so I was thinking about what content I could put on the site. I know I want to do a pass through each team and check on their first-year players but that needs a little more preparation time. An article I enjoy and have done in the past has to do with Strength of Schedule (SOS) to this point in the season so that’s what I want to focus on here. I have reached out to “Ned” who has a QWI tool that actually simulates the NCAA SOS used in the selection criteria but I’m unsure if he’ll get back to me before I want to publish this. As a reminder, SOS is the winning percentage of the teams that have been played. The better the record, the tougher the schedule. The NCAA actually provides this number in their statistical database so I’ll use that below. The NCAA uses SOS a little differently for selection. They use the normal SOS as 2/3 of what they use for selection SOS. The remaining 1/3 of the selection SOS is a team’s opponent’s opponent’s record. Basically, they try to normalize the number a bit by making sure teams haven’t built up good records by beating only bad teams. “Ned” just emailed back that he’ll have something back to me tonight (Wednesday). That could alter this article or maybe I’ll just do two tomorrow.

I need to touch on something else for a second and that’s the Watchlist. Loyal readers know that I do a Watchlist of teams that I think have a chance to be regionally ranked. You can find past data on the left side menu (or pulldown). I’m thinking I will not be doing that this year or, maybe, will do it in a different form (like a separate article). I will have to see what “Ned” sends me as I like to use his data for the Watchlist. The main reason for skipping the Watchlist work, besides laziness, is the fact that I just don’t think our region has a lot of possible at-large bid teams. I mentioned the Watchlist because SOS kind of plays into it as the Watchlist attempted to use the selection criteria to determine our regional rankings. It’s fine to figure out the regionally ranked teams but if they have no chance to be picked for the NCAA Tournament then where is the fun in that.

Right now, if you look at the Region X standings (which isn’t a real thing), you’ll find our best teams (per record) are La Verne, Mary Hardin-Baylor, Austin College, East Texas Baptist, Puget Sound, Trinity, Texas Lutheran, CMS and UC Santa Cruz. These nine teams would probably have composed the Watchlist as record is one of the selection criteria. What’s interesting about this list is the absence of UT-Dallas, Pacific Lutheran, Whitworth, Colorado College, Southwestern and Pomona-Pitzer. These are some heavy hitters that typically are regionally ranked but who’s records have been lacking this season. Let’s take a look at our nine teams and see how their records (through Tuesday) have been accumulated.

La Verne is 12-0 and has constructed that record against teams that have gone 86-62 (0.581). That’s not a horrible SOS especially for our region but we always have to remember that this number isn’t used in selection. Looking forward, and this is disturbing, their future opponent’s winning percentage is 0.458. When you combine the numbers for the entire season, you get an SOS of 0.521. That will not get it done when it comes to selection without a nearly perfect record and some key out-of-region wins. Their biggest out-of-region wins are Carthage and St Catherine and they could be regionally ranked. Any damage those teams do to others will help common opponent (another selection criterion). The one saving grace for La Verne is that they still have CMS on their schedule at least twice. Win one of these and they could have a chance at the at-large even with a low SOS.

Mary Hardin-Baylor is 14-1 with an SOS to date of 0.452. That’s 309^th out of 424 teams. When I say that you can schedule your team so that they will never be considered for selection, this is an example of it. Shockingly, their future SOS is 0.411 so it’s going to get worse! Overall, the NCAA database shows a final SOS of 0.433. If UMHB went undefeated (too late, I know) and took their only loss in the conference tournament final, they would not be selected for an at-large because they lack any quality wins and would have one of the lowest SOS (if not the lowest) under consideration.

Austin College is 11-1 with an SOS to date of 0.554. That is borderline okay as long as you have a great record. Future SOS is 0.547 and overall is 0.549. You can see the benefit of playing in the SCAC versus the SCIAC and ASC teams mentioned above as their SOS doesn’t really drop too much by playing conference. This is because the SCAC accumulated a better record overall prior to conference play. (Remember that conference play will always end up as 0.500 as half the teams will win and half will lose and you play everyone equally.) (That last note was technically wrong because good teams will have an SOS below 0.500 while the bad teams will have an SOS above 0.500 because you can’t play yourself.) (The last note was probably confusing but I don’t have the desire to re-write it or explain further.) Back to Austin College because they have a borderline SOS, let’s take a look at their best out-of-region win, which was probably Rhodes. Their record is also the product of a few non-DIII matches that will not count at selection time. Neither of these factors will help at selection time. They will need wins against Trinity to have hope for an at-large.

East Texas Baptist is 10-2 and has constructed that record against teams that have gone 0.597, which is pretty good. Unfortunately, by playing in the ASC, their future SOS is 0.449 and their overall is 0.515. Another SOS number that will not play come selection time.

Puget Sound is 10-2 and has constructed that record against teams that have gone 0.515. Their future SOS is 0.452 and their overall is 0.483. Another team probably out of the at-large discussion.

Trinity is 11-3 and has an SOS of 0.601. Their future SOS is 0.582 with an overall of 0.591. That will play come selection time. They also have out-of-region wins over Franklin & Marshall and MIT and I know that MIT will help if they need an at-large bid. On the flip side, losing that 5-set match to Tufts may hinder their chances. We’ll have to see how Tufts and MIT do the rest of the way, but Trinity is a team that will look strong come selection day.

Texas Lutheran is 9-4 and has an SOS of 0.551. Their future SOS is 0.622 and their overall is 0.588 so they are still in the at-large discussion. They have an out-of-region loss to Aurora who will not be regionally ranked so it actually doesn’t hurt as bad, which is a strange fact as well as being stupid. They have a nice win over Chicago who are in the same region as Aurora so it probably will even out. I don’t think Chicago will be regionally ranked but they are better than Aurora even though Aurora has a better record. So, not much help out-of-region. TLU, of course, still has Trinity on the schedule, too.

CMS is 8-3 with an SOS of 0.607 but a future SOS of only 0.481. That makes their overall number 0.535. Again, this won’t be their selection SOS and I probably need to stop down a second to explain more.

The numbers I’m providing are based on the current records of teams. If you played an out-of-conference team then your SOS is going to fluctuate based on how well they do the rest of the way. The NCAA database numbers are not factoring that in. The more teams you played out-of-conference, the more variable your SOS will be.

Back to CMS and that SOS is a concern. With that said, they’ve played a number of teams that should do well the rest of the way and that includes teams they have lost to as SOS doesn’t care whether you won or lost. So, for a different team, I might say that SOS will keep them out of the at-large game but that’s not the case here. Out-of-region teams they have played are Carthage, UW-Eau Claire, WashU (lost), Johns Hopkins (lost), Chicago and Juniata (lost). They would be an at-large lock if they had won one of those lost matches, but I still think they are alright.

UC Santa Cruz is 12-7 with an SOS of 0.596. They have a future SOS of 0.748 (!) and an overall SOS of 0.650. So, no issues here and you can actually see one of the benefits of NOT playing in a conference that plays matches. UCSC also has some nice out-of-region wins against Gustavus Adolphus, Tufts and Endicott. They also have two more tournaments in the future that should help (as you can tell from that future SOS number). If you want to look at a program that scheduled well, stop down here. Those 7 losses are problematic but their SOS and ranked wins should play come selection time.

So, if you wondering why I was talking about my Watchlist earlier then you’ve probably realized that this is what this article has turned into. Of the nine teams I mentioned, the one’s I see with an at-large chance, right now, are La Verne, Trinity, Texas Lutheran, CMS and UC Santa Cruz. There could be teams I didn’t mention that start to play well and have a chance down the road (Colorado College comes to mind) but that’s a small number of teams when you consider, at least, two of them should probably get an automatic qualifier. My last little caveat to this portion of the article is just because I think these teams have a chance at an at-large, doesn’t mean they will be regionally ranked so that they are positioned to be considered. UCSC has a number of bad in-region losses, which won’t be helped by the fact they don’t play in a conference that is within the region. (No chance to compensate down the road.)

Some other tidbits about SOS so far within the region:

• Colorado College leads the region (and SCAC) in SOS at 0.656 (19^th)

• Howard Payne is last in the region at 0.376 (383^rd)

• UT-Dallas leads the ASC with 0.627

• Pacific Lutheran leads the NWC with 0.618

• Caltech leads the SCIAC with 0.630

I guess the thing to gleam from the tidbits is that Colorado College still has an excellent chance at an at-large. They just need to win.

Alright, I was debating on how to present this next set of data but let’s just do it here and you guys can have a long read. “Ned” did send his QWI update, which includes his simulated selection criterion SOS. The first thing I wanted to do was compare the SOS from the NCAA database to what “Ned” has. Here’s a quick breakdown (“Ned’s” number is to the right):

• La Verne 0.581 -> 0.564

• Mary Hardin-Baylor 0.452 -> 0.461

• Austin College 0.554 -> 0.492

• East Texas Baptist 0.597 -> 0.554

• Puget Sound 0.515 -> 0.506

• Trinity 0.601 -> 0.580

• Texas Lutheran 0.551 -> 0.557

• CMS 0.607 -> 0.599

• UC Santa Cruz 0.597 -> 0.590

For the most part, the numbers are within 0.015 of each other (Mary Hardin-Baylor, Puget Sound, TLU, CMS and UC Santa Cruz). La Verne is close to that tolerance as is Trinity. The two outliers are Austin College and East Texas Baptist. My assumption is that these variances are due to the opponent’s opponent’s record that counts as 1/3 of the selection criterion SOS. If true, then the Austin College and ETBU numbers probably mean that they are not doing well in this regard (in other words, the teams they have played have played bad teams).

“Ned’s” data includes more than just SOS calculations. He also attempts to rank teams by regions (as well as overall). This isn’t done strictly by the selection criteria but by his QWI number. We have to remember that QWI doesn’t factor in head-to-head and common opponent. Still, it’s as close as we are going to get until the rankings come out and once you start comparing out-of-region teams, head-to-head and common opponent become rarer. Here is what QWI has for each region:

Region I

Region II

Region III

Region IV

Region V

Region VI

Region VII

Region VIII

Region IX

Region X

Lots to take in above. I certainly don’t agree with some of it but this is what you get with a mathematical formula. Still, it’s a great first pass.

Just scanning down the teams and you can see that QWI shares my thoughts on the quality of the New England and New York Regions. Those are the only four regions that have teams in the top 7 with a zero or negative QWI. There are two teams in Region VIII that didn’t make their rankings that would lead three of the NE or NY regions (Carthage and UW-Platteville). Region V has done a great job getting at-large bids in the past couple of years but they seem a little top heavy this year. Still, don’t be surprised if most of those teams get bids. I’ve struggled with Mary Washington in my Top 25 due to their schedule. QWI has them 58^th overall. Region VII and QWI has Calvin over Hope. Okay. QWI has Oshkosh (Region VIII) as the number 1 team in the nation, which shouldn’t be surprising since it rated their schedule as the toughest and they haven’t lost. Southwestern regionally ranked? I think “Ned” has a bonus in the formula for my Pirates.

Staying in Region X and that ordering would be a problem if both Puget Sound and UC Santa Cruz need an at-large bid. UCSC might get one but I don’t think UPS would so they would effectively block them. Just eyeballing teams and out-of-region teams that are regionally ranked that could help us come selection are Tufts, MIT, Endicott, Franklin & Marshall, Eastern, Birmingham-Southern, Saint Benedict, Gustavus Adolphus and Saint Catherine. As a region, we are all pulling for these teams to do as well as possible moving forward while not getting an automatic qualifier.

One last thing I wanted to look at were the number of Ranked Results each of the top 7 teams in a region has accumulated. This is one of the ways the New England and New York regions gain an advantage as they have more teams ranked that they can play.

• Region I – 27

• Region II – 31

• Region III – 24

• Region IV – 20

• Region V – 30

• Region VI – 27

• Region VII – 38

• Region VIII – 42

• Region IX – 30

• Region X – 37

Well, interestingly, exactly the opposite that I would have expected. Still, with conference play in full swing, we should see the numbers reverse by the end of the season. For instance, in Region I all ranked teams are in the same conference (NESCAC). Region II has four from the NEWMAC. It sure is nice to pick up a ranked result every time you have a conference match. I guess this shows, with minimal conference play so far, which regions have tried to play tough schedules and which haven’t. These numbers can also be skewed by teams that go out of their way to schedule tough. Both Trinity and UC Santa Cruz have 8 ranked results so far. That’s almost half of the Region X total.

I’m on a roll so more tidbits:

• UW-Whitewater has the most ranked wins (8)

• Wittenberg and Chicago have the most ranked losses (8)

• Middlebury and UW-Whitewater have the highest SOS at 0.668

• Warren Wilson has the lowest SOS at 0.275

• St Joseph and Wilson (not Warren Wilson) are the lowest ranked teams by QWI

• McMurry is one place ahead of them and is the lowest ranked team in Region X

Middlebury looks like an early contender to be a team that you just shake your head that they received an at-large bid.  

Lots of fun stuff but that’s my time and I hope I took up a small part of your Thursday. “Ned” usually posts on Volleytalk (link on the left or pulldown) and he attempts to keep a DIII thread going if you ever want to check it out.