NCAA Week 3 Power 10: Penn State Exits the Top 10 After Tennessee Loss — and the Numbers Nobody Published

On September 21, Penn State lost 3-1 to Tennessee. In the school's own recap, the cause was compressed into four words: unforced errors. No set scores. No service-error count. No attack efficiency. Not a single statistical line for Tennessee. Only one data anchor was offered to illuminate the entire story: setter Gabrielle Nichols recorded 38 assists and 12 digs — her third double-double of the season. That is everything we have.

For a data analyst, this is the most uncomfortable kind of text: a conclusion declared without accompanying evidence. I have sat with many recaps like this over nine years of tracking sports systems, and what I have learned is that when an organization publishes exactly the numbers that favor it and omits the rest, the choice itself is data. It reveals who controls the narrative and what they are trying to protect.

Penn State exited the Week 3 Power 10. Tennessee and TCU entered. That is the entire sequence that US media called an "early-season shakeup." But if you read the original report carefully, you find a paradox: the event is described as weighty, while the dataset meant to support it is too thin to balance anything.

When data is insufficient for a conclusion, the only thing left is the order in which the numbers appear — and that order always has intent.

The Power 10 is not a vote

Before dissecting the match, one common confusion must be clarified. The NCAA.com Power 10 is not an official ranking, not the AVCA Coaches Poll, and certainly not the selection tool for the NCAA Tournament. It is an editorial power ranking, written by a single analyst — Michella Chester — and updated weekly. Authority over the 64-team postseason field rests with the NCAA selection committee, based on the RPI combined with the eye test.

This distinction is not academic. It determines how we read the entire report. When Penn State "exits the Power 10," that is a perceptual event, not a competitive one. When Tennessee is described as having "entered the sport's top tier," that is an editorial claim based on one win, not a verified conclusion across multiple matches.

I often tell young analysts to separate three layers of text in any sports report. The first layer is the raw event: who won, who lost, what the score was. The second layer is data: the quantitative metrics attached to that event. The third layer is interpretation: the story the writer builds from the two above. The problem with the Week 3 report is that the second layer is nearly empty, while the third is written very fully. A large story built on a foundation that holds nothing.

Week 3 of the NCAA women's volleyball season falls in late September, when teams are still playing mostly non-conference matches. This is the window where every ranking swings hardest, because résumés are unformed, teams have not met enough times, and each result carries more perceived weight than its real value. A Week 3 win can lift a program into the top 10; a Week 3 loss can push one out. The Power 10 mechanism is inherently more volatile than the AVCA and RPI because it is designed to react quickly to each result.

In other words, we are reading a high-sensitivity ranking applied to a low-resolution dataset. That is the perfect formula for conclusions inflated in both directions.

The crack is where nobody measures

The original report says Penn State lost because of unforced errors. That is a diagnostic label, not a tactical explanation. It is like a doctor saying a patient is "sick" without saying why. Unforced errors can concentrate in three entirely different zones, and each zone tells a different story.

If errors concentrate on serving, that is a psychological and technical issue, usually appearing when a team trails and is forced to gamble. If they concentrate on attacking, that is a timing and block-reading issue, usually tied to second-ball quality. If they concentrate on serve reception, that is a system issue, and when the reception system collapses, everything behind it collapses too.

These three scenarios lead to three different diagnoses, three different remedies, and three different severity levels. The report does not tell us which one we are in. And that is the crack: a system can break from a detail nobody measures, but to know where it broke, you must have that detail first.

What I always recall in situations like this is the night Germany collapsed at the 2026 World Cup. Germany held 74% possession and generated 1.9 xG, while South Korea had only 0.4 — and still lost 0-2. If you read only the scoreline and possession stats, you conclude Germany lost to bad luck. But when I checked PPDA — passes allowed per defensive action — the figure of 11.2 appeared, far above the sub-8 standard of elite sides. The problem was not luck. The problem was a team that would not press, and that metric only surfaced when you dug deep enough.

Back to Penn State. We have a No. 9 team losing 3-1 to a No. 16 team. Structurally, this fits the scenario of a top-10 side losing its execution discipline rather than being tactically overmatched. But to distinguish between those two possibilities, we need exactly what the report does not provide: set scores and specific error counts.

A 1-3 loss with sets of 23-25, 22-25, 25-23, 23-25 is an entirely different story from a 1-3 loss with sets of 15-25, 25-20, 14-25, 16-25. The first says Tennessee won through a few moments, and the gap between the teams is nearly zero. The second says Tennessee was genuinely stronger for most of the match. The report does not tell us which, yet concludes Tennessee has "entered the top tier."

Without set scores, every claim about tier status is a guess dressed in numerical clothing.

The 38 assists and a misread

The only fully published stat line is setter Gabrielle Nichols: 38 assists, 12 digs, her third double-double of the season. It is a beautiful line, and precisely because it is beautiful, it is easy to misread.

In volleyball, a setter logging a double-double in assists and digs is often praised as a symbol of all-around excellence. But place it in the context of a loss. For a setter to have 12 digs, her team must have let many balls hit the floor in the backcourt, and she must have been the one diving in. In other words, the 12 digs measure not only individual defensive effort but also the volume of live balls Penn State had to handle in that match.

When a team generates a large defensive volume, there are usually two possibilities. First, the team is playing long rallies and winning them — a sign of a system that converts well. Second, the team is extending rallies but failing to convert them into points — a sign of inefficiency in transition attack. In a loss, the second is more probable, because if the first were true, the team would not have lost.

Meanwhile, Ava Falduto led the team with 15 digs. That figure, combined with Nichols's 12, shows Penn State generated significant backcourt defensive volume. But again, with no team totals and no Tennessee figures, we cannot know whether that volume was high or low against a standard. In a four-set NCAA women's match, a team typically records somewhere between 60 and 80 total digs. If Penn State had 55 total and Falduto had 15, she accounts for more than a quarter — normal. If the team total was 45 and Falduto had 15, she accounts for a third — a sign of individual dependency in defense. Without the total, no distinction is possible.

More notable is the silence around Ryla Jones, an outside hitter named without any stat line. In volleyball, the outside hitter carries primary responsibility for both attacking and backcourt reception. Leaving an outside hitter's line blank in a report where every other position has numbers is an editorial choice, and it hints that the line may have been unimpressive.

I am not asserting that. I am only saying that the absence of a number, in a text where other numbers are carefully selected, carries information. It is like a blank space on a survey map: the blank does not say there is nothing there, but it says the mapmaker chose not to fill it in.

The box-score structure: one-sided, unverified

Overall, the dataset in the report has three structural features worth naming.

First, it is almost entirely one-sided. Every number belongs to Penn State. There is not a single metric for Tennessee. This means we cannot assess whether Tennessee won by playing well or because Penn State played poorly. These are two different stories, and only data from both teams can distinguish them.

Second, it has no efficiency metrics. No attack success rate, no perfect-pass rate, no blocks per set. These are the most basic metrics in any volleyball box score. Their absence turns the text from a performance report into a narrative supported by selective statistics.

Third, it has no time-comparison element. No season averages for comparison. We know Nichols had 38 assists in this match, but not her average. We know this is her third double-double of the season, but not how many matches have been played.

In sports data analysis, there is a principle I always follow: a number without a comparison sample is an incomplete number. It is like knowing a player scored 20 points without knowing how long the game lasted, whether the team was strong or weak, and whether the pace was fast or slow. Nichols's 38 assists only mean something when placed beside the average assists of top NCAA setters, or beside her own average across prior matches.

There is one small but noteworthy signal: this is Nichols's third double-double of the season. If this signals a setter who is a consistent two-way contributor, that is valuable information about the team's ceiling. A setter who distributes well and defends well is a system asset, because she allows the team more attacking options from the backcourt.

But if the setter's double-double comes from the team relying on her all-around play to compensate for transition inefficiency, that is a warning sign, not a highlight. The same number, two opposite readings. Without team data, no resolution.

Tennessee and TCU: two programs, one promotion mechanism

The central positioning event of Week 3 is a perceptual tier inversion. Tennessee, ranked No. 16, beat Penn State, ranked No. 9, and was described as having entered the top tier. At the same time, TCU joined the Power 10. That two programs entered simultaneously shows this is not a single anomaly but a systemic rearrangement of the early-season perceived hierarchy.

What is notable is the timing. Tennessee's win falls in the non-conference window — the optimal period to bank a "resume win" before conference play begins. In the NCAA system, non-conference matches play a special role in building the RPI, because they allow cross-conference comparison. A win over a highly ranked team carries greater RPI value than a win over a weak team in the same conference.

So the Week 3 shakeup has a double meaning. It is a perceptual event, reflected in the Power 10. And it is a résumé event, reflected in the RPI. But only the first is widely covered, while the second is what actually affects Tennessee's future in December.

On Penn State's side, the exit is described as its first time out of the Power 10 this season. That detail matters more than it appears. It shows Penn State had sustained top-10 presence beforehand, and one loss does not erase that quality baseline. This is a perception correction based on one data point, not a decline.

I want to stress this distinction because it is often overlooked in ranking discussions. A program leaving the top 10 after one loss is not the same as a program on a downward trend. The first is a short-term event, reversible with a few wins. The second is a long-term trend, confirmed over months. The report shows us only the first, but its writing can make readers think of the second.

Risk lies at the interpretive layer, not the competitive one

When I build a risk matrix for this situation, the clearest finding is that most risks belong to the interpretive layer, not the competitive one.

The biggest risk is over-reading a Week 3 result as a durable tier change. This is medium-probability and medium-impact, because it affects how the public, recruiters, and even the athletes themselves perceive their program's standing. A team called "top tier" after one win may face expectations greater than its actual capacity. A team seen as declining after one loss may be undervalued.

The second risk is confusion between the editorial ranking and the official selection mechanism. The Power 10 swings harder than the RPI and AVCA because it is designed to react quickly. If readers treat the Power 10 as a gauge of NCAA Tournament chances, they get a distorted picture of teams' real standing.

The third, lower risk is a soft dependency on the setter position at Penn State. If Nichols is the primary and nearly sole distribution option, having her carry both defense and offense creates a structural weak point. This is a risk to monitor, not a confirmed conclusion, since the report gives no information on backup options.

The fourth risk is that missing set scores and error counts may mask the true magnitude of the upset. If the match was close, its impact on both résumés is far smaller than if it was lopsided. Without that data, any assessment of the upset's magnitude carries large error in both directions.

Overall, the situation's risk rating is low to medium. The report describes reputational movement within a media ranking, not a competitive, personnel, or compliance crisis. The main risk is analytical: reading too much into one early-season result.

The counter-intuitive angle: perhaps the upset was inflated, not underrated

The natural reaction to a data-thin report is to doubt the positive conclusion. If Tennessee is called "top tier" without sufficient evidence, perhaps they do not deserve the label. This is the obvious reading, and I think it is partly right.

But there is a more counter-intuitive reading, and it is more interesting. Perhaps the problem is not that the upset was inflated, but that both conclusions were inflated — both Tennessee's promotion and Penn State's demotion. And what is really fluctuating is not the quality of the two programs but the sensitivity of the measuring tool.

Think of it this way. The Power 10 is a one-writer, weekly-updated ranking designed to react quickly to each result. Structurally, it is like an index with a very short smoothing window. When you apply such an index to a sparse early-season data series, you get large swings that do not correspond to any real change in the measured system.

This is a phenomenon I have encountered many times in football data analysis. Short-window metrics like five-match form often swing harder than a team's true quality. A team can jump from fourth to second in a form table simply because of two wins over weak opponents, while its true quality is unchanged. Readers of short-term form tables easily mistake index fluctuation for system change.

The same may be happening here. Penn State leaving the Power 10 and Tennessee entering may reflect the tool's sensitivity more than a real change in tier. If so, both conclusions — "Penn State is declining" and "Tennessee is top tier" — are illusions created by the measuring tool.

The way to verify is simple and requires no additional match data. Just compare the Power 10 with the AVCA and RPI over the next few weeks. If the Power 10 swings hard while the AVCA and RPI stay stable, the editorial tool's sensitivity is the dominant factor. If all three move in the same direction, the change is real.

When three different measures tell three different stories, what is fluctuating is usually the ruler, not the thing being measured.

There is one more point worth reflecting on. The original report is written in an objective style, without hyperbolic language. But it still participates in the early-season hype cycle by amplifying a promotion based on one match. This amplification does not come from wording but from structure: it gives much space to the claim and little to the evidence. Sometimes, how a text allocates attention matters more than what it says directly.

Signals to track in the next round

If I had to pick the signals that will confirm or refute the Week 3 conclusions, I would pick five.

The first is Power 10 movement in Weeks 4 and 5. If Tennessee and TCU hold their positions, the promotion claim has a basis. If they drop out soon after, the Week 3 shakeup was merely tool fluctuation.

The second is Penn State's conference-play results. If they return to the top 10, the Tennessee loss is confirmed as an anomaly, not a decline. If they continue to drift away, Week 3 is the start of a trend.

The third is Tennessee's performance against ranked opponents in conference play. One résumé win needs to be confirmed by other résumé wins. If Tennessee only beats Penn State and loses to the rest of the strong opponents, the "top tier" label will not hold.

The fourth is the set scores of the September 21 match. If the full box score is published, we can re-measure the upset's true magnitude. A close loss and a lopsided loss lead to entirely different assessments of the gap between the two teams.

The fifth is the divergence between the Power 10 and the AVCA or RPI. If these two systems diverge, that gap is the gap between perception and reality.

What remains after the arena lights go out

In US college sports, there is an under-noticed feature: programs do not just compete to win, they compete to be seen. Rankings, editorial or official, serve a recruiting function. A program called "top tier" attracts better young athletes, and better young athletes create a stronger program over the following seasons. This spiral is slow but powerful.

That is why tier claims, even based on one match, have real effect. Not on this season's scoreboard, but on the scoreboard three years from now. And that is also why careful analysis of such claims matters. Not to deny them, but to know what they rest on.

I do not believe in luck. I believe in the frequency with which luck appears. One surprise win may be luck, but if the same program repeats it many times, that is frequency, and frequency is data. The question of Week 3 is not whether Tennessee deserves it. The question is whether they can repeat this result as the data thickens.

People look at the ranking. I look at the gap before the ranking was written. And the biggest gap of Week 3 lies in the numbers never published: set scores, specific error counts, efficiency rates for both teams. Those are the numbers that will tell us what actually happened on the court on September 21.

Data never lies; only people lie to themselves. And when a report chooses not to publish data, the careful reader's first move is to ask: what are they holding back, and for whom.