All scores →
Loading…
Loading…
Loading…

Recruiting rankings: stars, composite scores and what they predict

By 4Basketball Staff · · 10 min read

How high school recruiting rankings are built, what stars and composite scores mean, how well they predict college and professional success, and why the best use of a ranking is as a starting point and not a verdict.

A high school junior commits to a college, and within the hour a number appears next to his name: four stars, a national rank in the thirties, a composite score expressed to four decimal places. Fans compare the number to the commitments of rival programs. Message boards debate whether he deserves a higher grade. Coaches, who are supposed to be indifferent to such things, glance at the rankings before they glance at anything else. Recruiting rankings have become a major part of how college basketball is discussed, and they drive a large share of the conversation between March and the following November.

Rankings are useful, but they are also routinely misread. A star rating is not a prediction of a player's future; it is an estimate of his current standing relative to the rest of his class, made by a person watching him play a limited number of games. A composite score is an average of such estimates. The thing the rankings measure well is potential as scouts see it. The thing they measure less well is how a player will fit a system, how he will develop and whether he will remain healthy.

This article explains how the rankings are made, what the common terms mean, how well rankings predict outcomes and where they mislead. If you are a fan trying to understand why your team's class is ranked where it is, or a family trying to make sense of the numbers attached to a young player, the aim is to give you a working framework.

Where the rankings come from

Before the internet, recruiting was a world of newsletters, camps and word of mouth. A handful of people, including Howard Garfinkel, who ran the Five-Star Basketball Camp, were known for their ability to assess young players and to connect them with coaches. College coaches relied on personal networks and on summer camps where the best players gathered. Rankings existed, but they circulated in print, to a small audience.

The internet changed that. Starting in the late 1990s, websites began to publish player evaluations, build databases and rank prospects for a public audience. Several services grew into national operations, with scouts covering regions and events, and a handful of names became recognizable. Analysts such as Dave Telep, who worked at ESPN and elsewhere, became well-known figures in the field. The major services today include 247Sports, Rivals, On3 and ESPN, each with its own network of evaluators and its own rankings. Women's basketball has its own set of rankings, with ESPN's HoopGurlz one of the long-running examples.

Evaluators gather information from several sources: summer circuit tournaments, high school games, showcase events and camps, all-star games such as the McDonald's All American Game, which has been played since the late 1970s, and international events. They also watch video and talk to coaches. Their ratings reflect size, skill, athleticism, production, competition level and projection, meaning how they think the player will grow. A 6-foot-9 sixteen-year-old with good coordination may be ranked higher than a 6-foot-5 player of the same age with better current production, because the evaluator is estimating what each will become.

What stars mean

The star system is a simplified grade, usually on a scale from one to five.

  • Five stars are reserved for the very top of a class, a group typically numbering in the range of a couple of dozen prospects in basketball. These are players expected to be impact players at the highest level of college basketball and often professional prospects.
  • Four stars are strong high-major prospects, expected to start or play major minutes at good programs.
  • Three stars are good Division I players who may be role players at the top level or stars at mid-major programs.
  • Two stars and below describe players who are expected to contribute at smaller programs or at lower divisions.

The exact cutoffs vary from service to service and from year to year. Because the stars are a coarse tool, a player at the top of the four-star range may be more highly regarded than one at the bottom of the five-star range. The numerical rank, the player's position in a ranked list of a hundred or more, gives more detail than the star rating alone.

The composite

If different services rate the same player differently, whose view is right? The composite is an attempt to resolve that. The best-known version, the 247Sports Composite, combines the ratings of the major services into a single score by a weighted averaging process. The composite is regarded as more stable than any one service's rating, because it smooths out the quirks of individual evaluators. If one analyst loves a player and another is unimpressed, the average lands somewhere in the middle.

Composite rankings are also used to rank teams' recruiting classes, by adding up the scores of the players who commit. A class with several top players will rank above a class with a single star and a group of lower-rated players, even if the single star is better than any of the top players in the first class. That is one limit of class rankings: they reward quantity as well as quality.

What rankings predict well

The evidence is that rankings carry real information, especially at the top.

Professional outcomes. The players ranked among the very best in their class are far more likely to be selected in the NBA draft than lower-ranked players, and the connection is strong at the top. A high school ranking among the top ten or twenty in a class is a good predictor of a first-round selection. This makes intuitive sense: scouts for the professional teams and the people who rank high school players are looking at many of the same things, and they often watch the same events. For the college-to-pro evaluation process, see how draft prospects are evaluated with data.

College production for top players. Highly rated recruits are more likely to play major minutes early and to have high usage and efficiency numbers at good programs. A five-star freshman is more likely to start than a three-star freshman at the same school.

Team success. At the level of teams, programs that consistently sign top classes tend to win more games and advance deeper in the tournament. The relationship is not perfect, but it is robust. The programs that reach the Final Four in a given year have usually recruited and developed a number of high-level players.

Where rankings fall short

The limits matter as much as the strengths.

The middle is noisy. The difference between a three-star player and a four-star player is real on average, but within any given class there is a huge overlap. A large number of all-conference players were rated as three-star prospects, and a large number of four-star prospects never became significant college players. Evaluating a player in high school is difficult, and the margin of error is large for everyone below the top.

Late bloomers. Players develop at different rates. A teenager who has not yet grown, who started playing late or who attends a small school against weak competition may be overlooked. Some of the most celebrated professionals were lightly recruited. Stephen Curry drew few offers from major programs and played at Davidson. Damian Lillard played at Weber State. Steve Nash came out of a small high school in British Columbia and played at Santa Clara. Jimmy Butler spent time at a junior college before playing at Marquette. These stories are well known precisely because they are exceptions, but they illustrate that the system misses people.

Selection bias and circularity. Rankings are shaped by exposure. A player who attends the major events and plays on a top summer team is seen by more evaluators than a player who does not. And once a major program offers a scholarship, the player's ranking often rises, because the offer is viewed as a vote of confidence. That means rankings partly reflect the decisions of coaches. In some cases, a player may be ranked highly because he received offers, and he received offers because he was ranked highly.

Fit and role. A star rating says nothing about whether a player fits a particular system. A skilled shooter may thrive in an offense built on spacing and struggle in one built on post play. A quick, athletic guard may be ideal for a pressing team and a poor fit for a slow half-court team. The best coaches recruit for fit, not just ranking, which is one reason some programs outperform their class rankings. Our piece on how college coaches evaluate high school players looks at what coaches add to the picture that rankings do not.

Development and health. What a player becomes depends on coaching, strength training, health and opportunity. A highly ranked recruit who suffers a serious injury may never fulfill his rating. A lower-ranked player who works with a strong development staff may exceed expectations. The rankings, made at a fixed moment, cannot capture this.

The transfer portal and the value of class rankings

The usefulness of team recruiting rankings has declined in recent years because rosters are no longer built only from freshmen. With the transfer portal, a team may add several experienced players each offseason. A class ranking based only on high school commitments misses a large part of the picture. Many analysts now supplement the rankings with a view of the whole roster. If you are comparing programs, look at the overall composition: how many returning players, how many transfers, how many freshmen. We explain the portal in how the college transfer portal works.

The same concern applies to the one-and-done pattern. A program that signs the nation's best freshman may get one season of his time. That can produce a great team for a year, but the class ranking tells you little about what the team will look like the following season. See the one-and-done era for how this works.

How to read a ranking sensibly

A few habits help.

  1. Pay attention to the tier, not the exact number. The difference between the 12th and 15th player in a class is not meaningful. The difference between the 5th and the 80th is.
  2. Check several sources. Use the composite, or compare more than one service. If the services agree, the signal is stronger.
  3. Remember the age. A player who is young for his class may have more room to grow than one who is old for it.
  4. Look at production when available. A player's statistics against strong competition, particularly on the summer circuit, add useful detail.
  5. Do not read too much into class rankings. They are an average of individual ratings and do not account for the transfers a team might add or lose.
  6. Think about fit. Ask what a player does well and whether the program's style will use it.

For families: what the numbers do and do not mean

If you are the parent or guardian of a player, it is important to treat the rankings with care. A high ranking can bring attention, scholarship offers and opportunities. A low ranking or none at all does not mean a player cannot succeed in college. Many players earn scholarships through steady improvement, strong academics and good communication with coaches. For practical guidance, see our guide to how to get recruited. It is also worth remembering that rankings are made by businesses whose revenue depends on audience interest, which creates an incentive to generate attention. That does not mean the evaluators are not skilled; many are. It does mean that the product is partly entertainment.

For young players, being ranked is a lot of pressure. Comparisons to peers, social media commentary and early labels can affect confidence. Coaches and parents should help a player keep the focus on what he can control: his work, his health and his schooling.

Why early rankings move so much

One of the least appreciated features of the rankings is how much they change over time. Services now publish lists of players years before they reach college, and the ratings of sophomores and even younger players are often the subject of debate. Early rankings lean heavily on physical development. A boy who matures early, who is taller and stronger than his peers at fifteen, can dominate and be ranked accordingly. By eighteen, the others have caught up, and his ranking may slip. Meanwhile a player who grew late and was ranked in the middle can climb.

This is a form of regression to the mean, an idea we discuss in regression to the mean: why extreme seasons rarely repeat. A player at the extreme top of a list, rated on limited evidence, is more likely to drop than to rise, simply because extreme ratings contain more luck and measurement noise than average ones. The practical lesson is to trust a ranking more the closer it is to the player's senior year, and to treat anything earlier as a sketch.

The same applies to commitments. A player who commits early, as a sophomore or junior, locks in a relationship before he and the program know how each will develop. Some of those commitments are later reversed or reopened. This is one reason many coaches and advisers urge patience.

What to take away

Recruiting rankings are best understood as informed estimates, not as verdicts. They say a great deal about who the top prospects are, and they carry real information about who will reach the professional level. They say much less about the middle of the pack, about late developers or about how a player will mesh with a particular coach. Use them as a starting point, look at the whole roster and remember that development matters as much as the number next to a name. For where the numbers lead in the standings, check the rankings page and the college basketball section.

Search more on this

Keep reading