NBA Rank 2026-27: Walker Kessler Jumps 25 Spots After 5 Games, and the Flaw of a Ranking

Walker Kessler played exactly 5 games last season. Yet in the newly released NBA Rank 2026-27, he climbed 25 spots. In another corner of the same ranking, Jalen Williams — a player who once made the All-NBA Second Team at age 23 — dropped to No. 25. Between the two releases, neither played a single meaningful minute of elite basketball that changed anything. The only thing that changed was the story around them.

I sat in front of those two data points for a long time, and what I saw was not basketball. I saw a ranking scoring for the spotlight, not for on-court performance. That is not empty criticism. It is an observation verifiable through the very numbers the voting panel uses to defend its results. And if I am right, then hundreds of thousands of people are reading this ranking as a measure of ability, when it is actually a measure of reach.

What made me decide to write this piece was not an argument about a specific position. It was the moment I realized that the insiders themselves — the reporters on the voting panel — do not agree with each other on criteria. LeBron James at No. 18 is called too low by some, fair by others, appropriate by others still. When a measurement provokes that much internal division among its own creators, the problem is not the number. The problem is the method.

What NBA Rank Was Built To Do

NBA Rank is ESPN's annual product, first launched in 2026 and maintained almost continuously for more than a decade. Its structure is simple: a panel of dozens of reporters, editors, and former players votes on a list of the top 100 players, then publishes them in groups from No. 100 down to No. 1. The format's special feature is that it creates a debate thread lasting several weeks, split into release waves to maximize engagement with each wave.

In Vietnam, NBA Rank is often translated and re-shared in basketball community groups under headlines like who deserves it, who is underrated. Readers understand it as a ranking of ability. But its actual structure is a media product, not an analytical metric. This distinction sounds small, but it is the root of almost every controversy around the ranking.

When a panel of dozens votes, you do not get a measurement. You get an average of dozens of different opinions, each carrying market bias, playoff memory, fame, and the pressure to produce a list worth discussing. The panel does not vote in silence based on a spreadsheet. They vote in a newsroom, where every choice is cross-checked against the question: does this choice generate debate.

In that context, understanding NBA Rank requires a different skill than reading a stat sheet. You have to ask what this ranking rewards. And the answer, across many years, is fairly consistent: it rewards attention.

I have followed NBA Rank since 2026, when it was not yet popular in Vietnam. Back then I logged every position in an Excel file, then compared them with the actual metrics of the following season. My original purpose was simple: to check whether the ranking predicted correctly. But the more I did it, the more I realized the right question is not whether it predicts correctly, but what it is predicting. A ranking that correctly predicts a player's media reach in the coming season is a ranking that works efficiently — it just does not measure what readers think it measures.

Four Cases, Four Data Gaps

I selected four specific cases from the 2026-27 ranking for analysis, not because they are the most controversial, but because they expose most clearly how the ranking operates at the methodological level. Before going into each case, I need to state clearly one thing about the source's verifiability: several roster scenarios that appear in projected versions of this ranking — for example LeBron James placed on a Philadelphia roster, or Kawhi Leonard tied to a return to Toronto — do not match verified transaction reality. That means most roster-dependent conclusions below must be read as a scenario exercise, not a transaction report. As someone who works with data, I must say this before analyzing, because analysis based on a roster that does not exist has its value limited at the methodological level, not the factual level.

Walker Kessler Climbs 25 Spots After 5 Games

This is the number that made me pause longest. Walker Kessler was pushed up 25 spots from the previous release. The data basis for this jump, according to the panel's own commentary, is the spotlight — meaning his move to a bigger market and greater visibility. But the sporting basis for the jump, on playing data, is nearly zero: last season Kessler played only 5 games.

Five games. That is the sample size. In statistics, a sample of 5 observations is insufficient to distinguish signal from noise in almost any context. You cannot say a basketball player has improved based on 5 games; you cannot even say he has stabilized. If I were defending a ranking, I would never let a 5-observation sample move 25 spots. Yet the ranking added 25 spots to Kessler.

This only makes sense if we understand the ranking is measuring something else. If the variable actually rewarded is visibility, then the 25-spot jump is entirely consistent. A young center in a big market, playing alongside an elite creator, is a compelling story. That story has media value. And the ranking paid for that story with position.

The issue is not that Kessler lacks potential. The issue is that we have no data to evaluate that potential. He might be an All-Defensive-caliber center next season. He might also be a player who plays 5 games and then gets injured. The ranking chose to believe the first scenario without evidence, simply because that scenario is the one more people want to believe. In risk analysis, this is a classic error: assigning high probability to the most appealing scenario instead of the one best supported by data.

I remember a principle I learned building injury prediction models for a domestic basketball team. Before adding a new variable to a model, I have to ask: if this variable disappeared, would the result change. If the answer is no, that variable is meaningless. In Kessler's case, if I remove the spotlight variable, the 25-spot jump disappears. That tells me which variable is actually operating.

Jalen Williams Drops to No. 25 For Reasons Unrelated To Ability

Opposite Kessler, Jalen Williams was pushed down to No. 25. But looking at the resume, this is a 23-year-old who once made the All-NBA Second Team — meaning he ranked among the fifteen best players in the league at that age. He did not lose skill. He did not decline in the games he played. What he lost was games: 49 absences due to injury.

This is a classic analytical error: confusing ability with availability. An excellent but injured player is not a lesser player. He is a less frequently present player. The ranking punished absence as if it were decline, and in doing so mixed two completely different variables into a single number.

The paradox is this: if the panel truly wanted to measure contribution to a team, they would have to separate these two variables. An All-NBA player missing 49 games still has higher professional value than an average player appearing in all 82 — it is just that value spread across fewer games. Combining them into a single number misleads readers about both.

I have worked with injury prediction models in basketball, and the biggest lesson I drew is that injury is not an attribute of a player, it is a random event with probability. A player missing 49 games in a season is not necessarily fragile; he may simply have been unlucky over a short period. Equating luck with ability is the most fundamental error of any perception-based ranking.

Notably, the ranking treats these two cases in opposite directions. Kessler, injured and unproven, is pushed up, based on spotlight. Jalen Williams, injured but with proven All-NBA caliber, is pushed down, based on absence. The same injury variable, two opposite results. This asymmetry cannot be explained by sporting logic. It can only be explained by story logic: Oklahoma City's disappointment is a less compelling theme than Los Angeles's spotlight.

LeBron James At No. 18 — The Most Interesting Methodologically

LeBron James at age 41 is ranked No. 18. This is the most interesting case because it shows the panel actually has the capacity to adjust for role — they just apply it inconsistently.

The basis for this position is a data sequence: after the All-Star break, LeBron shot 54% from the field, with 7 rebounds and 7 assists per game, at the lowest usage rate of his career. This is the profile of a player shifting from primary initiator to transition scorer and off-ball cutter. Technically, this is the standard model for aging superstars: reduce on-ball volume, maintain efficiency, extend career. In analytical language, this is a successful role conversion, and it deserves recognition.

The panel defends No. 18 with a role-adjusted argument: LeBron is a cut below All-NBA but still a legitimate All-Star. In other words, they value him by role-adjusted worth, not raw production. This is a reasonable analytical step. The problem is they do not apply the same logic to other cases.

If you age-adjust for LeBron, why not injury-adjust for Jalen Williams. If you reward efficiency on low volume for LeBron, why reward spotlight on empty data for Kessler. This inconsistency is the clearest evidence that the ranking does not operate on a unified methodological framework, but case by case based on story appeal.

There is one more point worth noting. LeBron is 41 this year and at No. 18. If you average recent rankings, a 41-year-old has never been placed in the top 20. LeBron remaining there reflects two things at once: he genuinely still plays well, and his name still carries enormous media value. Separating these two things is the analyst's job. Believing both at once without separating them is the fan's job.

Tyrese Haliburton And The Unnamed Achilles Discount

Tyrese Haliburton sits at No. 19, and the panel quickly argues this is too low. But there is one detail analysis must put on the table: Haliburton suffered an Achilles tear. And sports medicine history shows players returning from Achilles injuries typically take one to two seasons to regain explosiveness.

If the panel actually applies a health discount to Haliburton, then No. 19 is entirely defensible. The problem is they do not name that discount. They vote without explanation, leaving a gap readers fill with feeling. In data analysis, an unnamed assumption is more dangerous than a wrong assumption — because it cannot be verified.

I see the same thing in how Kawhi Leonard is ranked. Leonard is placed at No. 17 and tied to a scenario of moving to Toronto — a scenario I stated above does not match known transaction reality. Regardless of the scenario, the question remains: a player with Leonard's dense injury history ranked No. 17 on what basis. If based on peak skill when healthy, then that is an assessment of ceiling, not expected value. These are different things, and conflating them misleads readers. Ceiling is what a player can do. Expected value is what a player will do, multiplied by the probability he is on the floor. This ranking continuously measures the first and labels it as the second.

The Health Discount — The Ranking's Biggest Omitted Variable

If I had to point to a single variable this ranking handles worst, it is health and availability. At least four of the stars mentioned in the ranking carry serious injury flags: LeBron at 41, Kawhi Leonard with a dense injury history, Jalen Williams missing 49 games, and Haliburton with an Achilles injury. Four of the six most prominent names on the list have availability issues.

A serious ranking would separate these two variables: ability when healthy, and projected games played. This ranking does not. The result is that readers do not know what they are reading. A player at No. 10 with a dense injury history may have lower actual value than a player at No. 40 who is healthy — depending on purpose. The ranking's failure to separate these variables is not just a technical inconvenience. It is a way of obscuring the truth.

I have seen this on a smaller scale, in consulting work for domestic clubs. When a club wants to sign a foreign player, the prettiest stat sheet has never been the deciding factor. The deciding factor is how many games he played over the last three seasons. A player scoring 20 points a game but playing only 30 games a season has lower value than a player scoring 14 but playing all 50. This is a calculation any sporting director must make. NBA Rank does not make that calculation, and so it does not serve the purpose many assume it serves.

Contrarian: Correlation Is Not Causation — Story Is Not Ability

Here I need to state plainly what I believe is the center of the whole issue, even if it is unpopular.

The NBA Rank ranking does not measure player ability. It measures the relevance of a player's story at a specific media moment. The two correlate — good players usually have more stories — but correlation is not causation, and in many cases that correlation breaks down entirely.

Numbers do not lie, but they also do not know how to tell stories. The problem is that when a ranking prioritizes storytelling, it gradually drifts from the numbers. This does not require a conspiracy. It only requires a set of voters sharing the same bias about what makes a list worth discussing.

Look at the structure of the jumps. Kessler climbs 25 spots despite nearly empty playing data, and the named reason is spotlight. That is a market variable, not a playing variable. Jalen Williams drops because of injury, not decline. That is an availability variable, not an ability variable. Both cases show the ranking adding and subtracting points based on variables outside the court.

In econometrics, this phenomenon has a name: omitted variable bias. The ranking tries to measure one variable — ability — but continuously introduces other undeclared variables: market, injury, exposure, playoff memory. The result is a number that looks like a measurement but is actually a weighted combination of many things. And because the weights are not published, readers cannot verify.

What bothers me most is not that the ranking is wrong. It is that it is read as right. When a Vietnamese fan reads Kessler at No. 41, Jones at No. 56, they may believe these are two comparable measurements. But if one number was added 25 spots for market and the other subtracted for injury, that comparison is meaningless.

I think about how public opinion handles rankings like this. They argue about each position, as if each position were an objective fact to defend or overturn. But the system producing them does not even claim to be objective. It is a panel. It is opinion. And opinion, even packaged as a very scientific-looking table, is still opinion.

Every coach talks about feeling. I have no feeling, I have standard deviation. And the standard deviation here is large. The dispersion of opinion on the panel — with conflicting views on LeBron at No. 18 — shows the voting body itself does not agree on criteria. When a panel publishes a single number but is internally divided, that number is not a measurement; it is a compromise. And compromise, in analysis, is a form of data blurring.

There is another aspect I want to dig into: the 2026 rookie class. The panel pushes several rookies high and predicts someone is bound to arrive All-Star ready. This is a classic form of optimism. In any rookie class, most top picks do not reach All-Star level in their first season. The rate of rookies making the All-Star team in year one is far lower than media expectations. But the ranking rewards potential because potential generates stories, while disappointment goes unpredicted.

This is the crux. A good model must be able to be wrong and must state in advance where it can be wrong. NBA Rank does not do this. It offers a number without conditions, without confidence intervals, without counter-scenarios. A number without a confidence interval is an unverifiable number. And an unverifiable number is not data — it is an opinion with a number attached.

I am not saying this to belittle the panel's work. Selecting the top 100 players is a genuine challenge, and the participants clearly understand basketball far better than I do. I am saying this to distinguish two things often conflated: an entertainment product with its own value, and an analytical tool it is not. A good entertainment product does not have to be a good analytical tool. The problem only arises when people use the first as the second.

A Lesson From When I Was Called A Lab Scientist

In 2026, while interning at a sports outlet, I wrote a prediction that Germany would be eliminated in the World Cup group stage. The basis was the team's PPDA in qualifying: 12.5 — far above the 9.8 average of the last five World Cup champions, with an average distance covered of only 98 km per match. Colleagues laughed, calling me a lab scientist. The result is known: Germany finished bottom of the group, lost 0-2 to South Korea, and were eliminated.

I tell this story not to brag. I tell it because it taught me a lesson directly applicable to NBA Rank today. When I predicted Germany's elimination, what I did was not offer a better opinion than the experts. What I did was isolate one variable — pressing intensity — from the general story of a defending champion. And that variable said something different from the story. With NBA Rank, the variable to isolate is availability. And it says something completely different from the story the ranking is selling.

In 2026, the whole world mourned Germany. I just quietly re-read my model's log file. Today, the whole basketball community argues about every position in NBA Rank. I just quietly count the games each player on the list has played over the last three seasons. Two tasks that sound different, but in essence are identical. Both are about separating signal from noise.

What To Track, Not What To Argue About

If the ranking is not worth arguing over position by position, what is worth tracking. My answer is signals that are measurable and predictable.

First, Jalen Williams's return. If he is healthy and plays over 65 games, the All-NBA resume remains intact. No. 25 will become an undervalued prediction, not a decline. I will track his games and efficiency in the first half of the season. If he plays over 65 games and maintains efficiency comparable to his All-NBA season, then the ranking erred here, and this is a fixable error.

Second, LeBron's off-ball efficiency. The 54% shooting, 7 rebounds, 7 assists at career-low usage pattern is a more trustworthy sample than Kessler's 5 games, but still a small sample if counting only the post-All-Star stretch. If it is maintained all season, this is one of the most impressive role conversions in league history. If it collapses, we will know the previous post-All-Star stretch was a small sample, not a trend. I lean toward the first possibility, because this role-conversion model has been proven across generations of players, but I keep a margin of doubt. A 41-year-old can maintain off-ball efficiency for a short stretch, but sustaining it all season at 41 is an entirely different physical question.

Third, the Kessler — Doncic pick-and-roll. This is the most concrete tactical scenario in the entire ranking. The logic is sound: a rim-finishing center plus an elite creator is an efficient formula. But it has never been tested on the floor. I will track Kessler's conversion rate in pick-and-roll situations, and I will wait for that number to pass a minimum sample before believing the 25-spot jump. In basketball, pick-and-roll pairs need time to build chemistry, and playoff defenses have many ways to neutralize a rim threat: switching, blitzing, or forcing the creator into mid-range shots. A pick-and-roll pair never tested in the playoffs is a pick-and-roll pair not yet evaluated.

Fourth, Haliburton's Achilles discount. This is a medical variable, not a basketball variable. The way to observe is first-step speed and rim-attack frequency. If he returns to pre-injury efficiency, the argument that No. 19 is too low will be validated. If not, No. 19 will look like a reasonably pessimistic assessment. In either case, I want to see a player returning from an Achilles injury play at least two full months before re-evaluating him. A player returning from a major injury typically goes through an uneven acceleration and deceleration phase, and evaluating him during that phase is a trap.

Fifth, the multi-star fit problem — regardless of the specific roster scenario. Any team with three ball-dominant players raises a question about spacing and touches. This is the type of problem often underpriced in the regular season, then exposed in a seven-game playoff series. I will track the distribution of usage rates and assists among the stars. If one star accepts a reduced role, this is a positive cultural indicator. If not, this is a risk. History shows three-superstar teams often fail not for lack of talent, but because no one agrees to reduce their role.

Sixth, the NBA's 65-game rule. This is a competition-structure variable, not a player variable, but it directly affects how any ranking should be read. Since 2026, a player must play at least 65 games to be eligible for awards like All-NBA and MVP. This means availability has become part of the league's own structure. A ranking that ignores this variable is ignoring a rule that has been institutionalized. If a player cannot qualify for All-NBA because he plays 60 games, then placing him in that group in an individual-ability ranking is inconsistent with the system the league itself established.

Data's Blind Spot: What The Ranking Cannot Say

There is one thing I always remind myself when analyzing any ranking: data does not play basketball. It only describes what happened, and even that is imperfect. NBA Rank does not even describe what happened. It predicts what might happen, based on what happened, filtered through the lens of the voters.

All data has blind spots. Performance data cannot speak to team chemistry. Offensive metrics cannot speak to defense. And an individual ranking can say nothing about a team sport, because basketball is decided by interactions, not by individuals lined up side by side. The strongest lineup is never eleven beautiful names, but eleven equations in harmony.

In this case, the ranking's specific blind spot is availability. A ranking ranks ability, but ability only has value when expressed. A No. 10 player playing only 40 games has lower actual value than a No. 40 player playing all 82, depending on purpose. The ranking's repeated conflation of these two concepts shows it is designed to provoke debate, not to forecast.

People look at goals to remember a match. I look at xG to understand the match that did not happen. Applied here: people look at positions in a ranking to remember the season. I look at confidence intervals to understand what the ranking could not say. And in this case, what the ranking could not say is a simple truth: availability matters as much as ability, and no ranking measures both at once with a single number.

The Market Structure Behind A Ranking

There is one more analytical layer I want to touch, because it explains why this kind of ranking exists the way it does.

NBA Rank is a media product in a peak content cycle. It is designed to generate engagement within a short window — a few weeks before the season starts. Within that window, the goal is not analytical accuracy. The goal is debate volume. Every decision about the list — who is pushed up, who is pushed down, who is placed in a compelling roster scenario — serves that goal.

This is not ethically wrong. It only means one thing: this ranking is not optimized for truth. It is optimized for attention. And when the two goals conflict — which happens more often than people think — attention wins.

Fans in Vietnam, as anywhere, consume this product as debate material. That is entirely reasonable. Debate is part of the joy of sports. The problem only arises when that debate is anchored to a number believed to be objective. At that point, fans are no longer debating basketball; they are debating a product selling them a feeling of objectivity it does not possess.

I think this is especially true in a market like Vietnam, where most fans access the NBA through translated content and summary rankings. A fan in Da Nang may not have time to watch enough games to judge a player themselves, so they rely on a ranking as an anchor point. This is entirely understandable. But it means the quality of the ranking affects the quality of understanding of an entire community. And when that ranking has structural bias, the bias spreads with it.

Conclusion: Signals For The Next Cycle

NBA Rank 2026-27 will continue to be shared, continue to provoke debate, and continue to be read as a measure of ability. That is hard to avoid. But if I leave one thing behind in this piece, I want it to be a different reading habit: before arguing about a number, ask which variable produced it.

With this ranking, the variables producing the numbers are largely spotlight, playoff memory, and story appeal. That is not a sin. It is the nature of a media product. But recognizing it is the first step to not being led by it. And the second step is returning to what can be measured: games played, efficiency on a sufficient sample, and the conditions under which a forecast can fail.

Data is a monastery: the less noise, the more clearly you hear something trying to speak. This ranking is noisy. And within that noise, what is worth hearing is not the number, but the silence of what it does not say: games played, sample size, and the conditions under which the model fails. If you can track those three things, you have surpassed most people arguing about the ranking without ever opening it to read the methodology.

When a young coach tells me this ranking is meaningless and he does not need it, I smile. I touch the future with a keyboard. And the keyboard does not need a panel — it only needs to know which variable it omitted. This ranking omitted many variables. The question for next season is not who stands at which position. The question is who among the names pushed up by spotlight will prove their value through actual games played, and who among the names pushed down by injury will reclaim their position through demonstrated efficiency. Those are questions answerable with data. And those are the only kind of questions worth asking.