% of Value: A Better Way to Measure Players

Most fantasy basketball analysis relies on z-scores to rank players. A z-score measures how far a player's performance sits from the average, normalized by standard deviation. The math is clean, widely understood, and thoroughly wrong about which players help win categories.

The problem is not that z-scores are calculated incorrectly. The problem is that "how unusual is this performance" and "how much does this performance help me win" are different questions, and z-scores only answer the first.

The Framework: Slicing the Pie

In a 12-team league where each team rosters 13 players, 156 roster spots exist. Those 156 players collectively produce all the points, rebounds, assists, steals, blocks, threes, and turnovers in the fantasy universe for that league. They also determine the aggregate field goal and free throw percentages.

Each player's value in a category equals their share of that total output. If the top 156 players average a combined 2,556.7 points per game and Cade Cunningham averages 25.3 of them, he provides 0.99% of available points. A perfectly average player would provide 0.64% (simply 1/156). Cunningham contributes about 55% more than his share in points.

Repeat across all nine categories, sum the percentages, and the result is total value. The framework requires no subjective weighting. The player pool itself determines how scarce each category is and therefore how much each contribution is worth.

Walking Through the Math

Cunningham's 2025-26 season through 47 games: - 25.3 points, 5.6 rebounds, 9.6 assists - 1.5 steals, 0.8 blocks, 1.9 threes - 46.2% FG, 80.8% FT - 3.7 turnovers

To calculate his assist value: among the top 156 players by total value, the combined per-game assist average totals 586.8. Cunningham contributes 9.6 assists per game.

Cunningham's Assist Value: 9.6 / 586.8 = 1.64%

That is more than 2.5 times the baseline of 0.64%. He provides two and a half times his fair share in assists.

For percentages, the calculation uses makes and attempts rather than the percentage itself — more on that below. For turnovers, the sign flips: a player's turnovers represent negative value, and more turnovers mean a larger negative share.

The full breakdown:

Total Value: 6.76%

A baseline player — average across all categories — would provide 4.49% total value. Cunningham provides about 51% more, driven primarily by elite assists (+0.99%) and strong free throw volume (+0.63%).

Where Z-Scores Go Wrong

Running the same data through z-scores, using the league averages and standard deviations from the same 156-player pool:

Both frameworks rank Cunningham well. But look at the turnovers row. His 3.7 turnovers per game is 1.77 above the league average of 1.93, divided by a standard deviation of 0.80. The z-score: -2.21. That single category wipes out more than half his blocks, threes, and FG% contributions combined.

In Percent of Value, the same turnovers cost -0.59% above baseline. That is a real cost — Cunningham does turn the ball over more than a replacement player — but it is proportional to the actual damage. His 3.7 turnovers are 1.23% of the league's total 300.4 turnovers per game. That cost is real but does not erase three other categories the way z-scores suggest.

The distortion comes from standard deviation. Turnovers have a relatively low standard deviation (0.80) compared to their mean (1.93), which inflates z-scores for any player above average. The z-score framework treats "unusualness" as importance. Percent of Value treats "share of the league's production" as importance. These are not the same thing.

The Efficiency Category Problem

The gap is clearest with field goal percentage. Consider two real players from this season:

Cade Cunningham: 8.9 makes on 19.2 attempts per game (46.2%) Jaxson Hayes: 2.9 makes on 3.7 attempts per game (77.2%)

Hayes has a far higher field goal percentage. In a z-score framework, his 77.2% sits nearly 4.6 standard deviations above the mean — a massive positive z-score. Cunningham's 46.2% sits 0.32 standard deviations below.

But field goal percentage in fantasy is not an individual stat. It is a team stat. When calculating a fantasy team's weekly FG%, each player's makes and attempts get pooled together: total makes divided by total attempts. Cunningham's 8.9 makes on 19.2 attempts move a team's FG% far more than Hayes's 2.9 on 3.7, because Cunningham's volume dominates the denominator.

Percent of Value handles this by measuring the share of field goal makes contributed to the league pool. Cunningham provides 0.97% of the league's total makes. Hayes provides 0.32%. The percentage itself is misleading without the volume context, and z-scores don't capture that volume.

No Artificial Weighting Needed

Z-score rankings often require manual adjustments. Managers notice that turnovers dominate the results, so they reduce the weight to 0.5x or 0.75x. Or they realize FG% z-scores overrate low-volume players and add a volume filter. Each adjustment is a judgment call, and there is no consensus on the correct numbers.

Percent of Value sidesteps this. The league's total production in each category provides natural weighting. If 156 players combine for 2,556.7 points per game and only 94.4 blocks per game, the framework inherently recognizes that 1% of blocks is scarcer than 1% of points.

This also means the framework adapts as the player pool changes. In a league where blocks are concentrated among fewer players, the scarcity value of each block increases automatically. The math follows the data.

Compare Two Builds

Cunningham and Alperen Sengun illustrate how different player profiles create different value shapes. Sengun averages 20.7 points, 9.4 rebounds, 6.3 assists, 1.3 steals, 1.0 blocks, and 0.6 threes through 46 games. He shoots 49.7% from the field and 69.1% from the line.

Their z-scores: Cunningham at +4.04, Sengun at +1.57. The z-score gap is 2.47. The Percent of Value gap is 1.01%. Z-scores exaggerate the difference between these players because Sengun's weak FT% (69.1%) gets hammered — it is 1.13 standard deviations below the mean — while Percent of Value recognizes that his free throw volume (4.0 makes per game) still contributes 0.89% of the league's total, above the 0.64% baseline.

This matters for roster construction. A manager trying to decide between these two player profiles needs to know the actual category trade-offs, not the statistical unusualness of each stat line. Sengun provides more rebounds and blocks. Cunningham provides more assists, threes, and free throw volume. Both are well above a replacement-level player. The choice depends on which categories the team needs, not which player has the higher total z-score.

What This Means for Drafts

The practical differences surface most often in the middle rounds, where z-scores and Percent of Value can disagree by 20 or more picks.

Players who typically rise in Percent of Value rankings: high-volume scorers whose efficiency penalties get softened by counting output; three-point specialists, because threes are scarce in the player pool; and playmakers whose turnover costs get right-sized instead of inflated.

Players who typically fall: low-volume efficient bigs whose field goal percentages look elite in z-scores but contribute little to a fantasy team's pooled FG%; extreme specialists with one standout category that happens to have high variance; and players whose low turnover rates inflate their z-score totals without providing much actual edge.

Why This Framework Powers Everything Else

Percent of Value is the foundation for every strategic concept in this series.

Punt strategies work by giving up share in certain categories to gain more in others. Evaluating those trade-offs requires knowing the actual value exchanged, not just the z-score delta.

Streaming decisions require understanding which categories provide the most value per roster move. A streamer who adds 0.15% of the league's blocks in a single week is more valuable than one who adds 0.15% of the league's points, because blocks are scarcer.

Trade evaluation works better when both sides can see the category-level percent of value being exchanged. A deal that trades 1.5% of total value for 1.2% might still be favorable if the 1.2% is concentrated in categories the receiving team needs.

The remaining posts in this series use Percent of Value as the baseline for evaluating draft strategy, punt builds, and trade decisions.

Look at your roster through this lens: which categories are your top players contributing well above baseline, and which are they dragging below? The answer shapes every move that follows.

Dean Pitton 15 Feb 2026

All stats from the 2025-26 NBA season through February 12, 2026. Player pool: top 156 players by total value (minimum 20 games played, 15 minutes per game). League category totals computed as the sum of per-game averages across the pool.

← Previous