VBelo Rankings: The 2025 Debut

Let's look at the VBelo ratings that power the daily projections. Don't worry, it will be very light on the math, but you might be surprised about where teams currently rank.

VBelo Rankings: A Quick Exploration

Every day, the VBelo model spits out the probabilities of either team winning each match. Those probabilities look like this (in case you don’t subscribe to the daily emails.) Here are a few matches for today over in the MIVA.

VBelo Report | February 6, 2025

This whole “VBelo" (pronounced vee-bee-ee-low) thing started with these percentages on a website formerly known as Twitter. At the heart of this passion project is a desire to grow the game by providing something I didn’t see in the men’s game coverage: an elo ranking model.¹

One of the fundamental questions that the VBelo model tries to answer is this: how good is a specific volleyball team? This is a simple sounding question, but it starts to get really complicated, really fast.²

Today, I am going to try and not dive too deep into the math of the model. Instead, I want to focus some broader strokes before finally getting to the first release of the VBelo Rankings of the season. (Yes, I guess you can just scroll to that part, but then you miss out on all the fun.)

What doesn’t the model care about?

• What the coaches think - The coaches’ poll means nothing to the model. It doesn’t matter if a team is ranked or not. This is a great way to remove narratives about who should beat whom.

• What the media think - Same thing for the media poll or any stories that might be going around. The model doesn’t care about opinions or off-court rumors. Those are for the message boards.

• What the fans think - This is probably obvious by now, but things like rivalries just aren’t in the model. It also doesn’t matter if a team has a lot of devoted fans or it is a smaller program with lighter attendance.

• The exact players on the court - Maybe more surprisingly, the model doesn’t actually care which players are on the court. Injuries don’t matter before a match (but they might end up having an impact on the results.) It also doesn’t matter if someone is having an off night or if so-and-so used to play for the other team.

What does the model care about?

• Winning - I can’t stress this enough. This model is fundamentally about winning. If a team wins, their ranking goes up.³ At the end of the day, teams that win more are ranked higher. But not all wins are created equally…

• How good the opponent is - This is where it starts to get fun. In the model, teams get rewarded more for beating better teams. The best way to increase your VBelo rating is to beat a team with a better VBelo rating than you. The catch here is that how good a team is continually being evaluated by the model. It is learning about these teams just like everyone else is while they watch them play.

• How many sets a team wins - This is a smaller point, but winning in 3 sets is better in the model than winning in 5 sets. Teams that are dominant get rewarded for that dominance.

• Where the match is played - Lastly, home court and distance travelled are taken into account. Travelling is hard and the farther you have to travel, the harder it is. Then there is home court. Let’s just say, home teams win. A lot. So far this season, home teams have won 68% of the time!

Rankings

That was a lot of words, I know. But let’s get to it. A small disclaimer, I am planning to make a prettier graphic eventually, but this is what my code gives me every time I run the model.⁴ It could be worse.

There are probably things you disagree with here. That is part of what makes this fun. Is Lincoln Memorial the 7th best team in the country? The answer is: it depends how you answer that question. If you focus on winning…then maybe they are.

One thing I do love about an elo model is that it provides more than just who is #1 and who is #2, but how how apart is #1 from #2. In this case, they are basically on top of each other! And then you look at the difference between #21 and #22 and it is only 1 point, the slimiest of margins! Then you can start to see groupings of teams, or tiers, if you will. This starts to create a more nuanced landscape of ranking teams. I like nuance.

So what?

This is one of my favorite questions. So what? Do I think these rankings are “perfect” and all other polls or rankings should be thrown in the trash? Absolutely not! Every ranking or poll tells us something. The real question is how do we take all of this data to form our sports opinions.

For me, I like to think of VBelo as a “jersey-blind” ranking. For example, if we didn’t know a team was UCLA, would we still rank them the same? Hard to say. Can we actually ignore those 21 championship trophies when saying who is currently the best team based on results? Probably not. If we just looked at the results, what would that look like. I think it looks a lot like the ranking above.

I also think there is another, very compelling “so what?” for VBelo rankings, but I am going to save that for next week. I’d love to know what you think, though. Let me know in the comments.

---

Rock Chalk, Jayhawk

On Tuesday, a tremor was felt in the men’s volleyball world as Grand Canyon Head Coach Matt Werle announced his departure. In his 9+ seasons⁵ with the Lopes, Werle racked up 161 wins and just 94 losses, good enough for a .631 winning percentage.

The 2024 season was Werle’s crowning achievement in the men’s volleyball world. He led GCU to their first MPSF championship, upsetting eventual national champions UCLA. That team then went on to win their first NCAA tournament match before falling in the semifinals. For his impressive year, Werle was recognized as the AVCA Coach of the Year, a well-deserved honor.

For the remainder of the 2025 season, assistant coaches Bryan Dell'Amico and Jon Girten will lead the Lopes. While it is no small feat to step into a role like this mid-season, Werle has set this program up for continued success.

Grand Canyon is currently 7-1 and does not play their next match until February 14th when they host the surging UCSD Tritons.⁶

gcu_mvolleyball

A post shared by @gcu_mvolleyball

---

Latest Polls

Coaches’ Poll

Okay, we are starting to see some movement (just not at the top). We finally have a new name to add to the poll: Lincoln Memorial. I am a big fan of the IVA getting the love and respect it deserves. For the Railsplitters, I think #19 isn’t high enough. They are a solid team with all of the parts to give just about any team a run for their money.

AVCA

Media Poll

Off The Block took the week off from the media poll. I just want to use this space to say Off The Block is fantastic and you should definitely be reading everything they are putting out. They will be back next week with a media poll and I am so glad that they get a small break in the busy volleyball season.

VBelo Rankings

I already put those in earlier. Did you miss it?

---

In’s & Out’s

A section for all of the random things that don’t fit anywhere but belong somewhere.

• Want to check on players who went on to play internationally? One of my favorite places to visit for this (and lots of other things) is Volleybox. It is really interesting to see all the teams these players end up playing for when they turn pro.

• AVCA Player of the Week: Dillon Klein (Outside Hitter, USC)

  • For those wondering, Klein is the 3rd OH to win the award this year and the first player from the MPSF.

    uscmensvolley

    

    A post shared by @uscmensvolley

• Triple-Double Watch 2025: Still nothing…BUT, let’s talk about a setter in the Big West not named Moni Nikolov. That’s right, Tread Rosenthal flirted with a Triple-Double against BYU last week!

  • 43 assists, 9 kills, 8 blocks, 6 digs

1. I was significantly inspired by FiveThirtyEight, several acquisitions ago, when their sports coverage was some of the best in the game, in my opinion. All of their sports models were elo models and I wanted something like that for men’s volleyball.

2. This is true for sports, in general.

3. Okay, there are actually a few cases where a team can win and their overall VBelo rating does go down slightly, but I’m not going to get into that. It is rare and makes sense for those situations. Just trust me on this one for now.

4. You can partially blame Hawaii and Stanford going to 5 for this graphic.

5. I guess this is technically 8 full seasons, 1 covid-shortened season, and 1 job-change-shortened season.

6. The model does not take into account personnel changes. Honestly, I don’t think it would be a great data input mid-season. Or rather, I think it would be a very noisy data input with lots of variability.