from AI:
how NET works...
Core idea
NET is a performance-based ranking built from game results and efficiency metrics. It is not just wins/losses and not just margin of victory it blends several factors.
Main components (what goes into NET)
1) Team Value Index (TVI)
This is results-based.
It rewards:
2) Adjusted Net Efficiency
This is performance-based.
It measures:
Net Efficiency = Offensive Efficiency Defensive Efficiency
Example:
Road wins count more than home wins.
Difficulty order:
[ol]
Road win (most valuable) Neutral win Home win (least valuable) [/ol]
Losses work the same way in reverse.
4) Scoring margin (capped)
Margin of victory is included but limited so teams don't run up scores.
Everything is calculated per possession so fast teams aren't automatically rewarded over slow teams.
What NET is NOT
It is NOT just:
The committee uses NET as the main sorting tool.
Key quadrant system:
Quadrant 1 (best wins):
Simple way to think about the formula
While the exact math is hidden, conceptually it behaves like:
NET
(Win quality + opponent strength + location value)
+
(Adjusted offensive efficiency adjusted defensive efficiency)
with margin capped and everything adjusted by possessions.
Why this matters for Cal (example context)
For a team like Cal:
how NET works...
Core idea
NET is a performance-based ranking built from game results and efficiency metrics. It is not just wins/losses and not just margin of victory it blends several factors.
Main components (what goes into NET)
1) Team Value Index (TVI)
This is results-based.
It rewards:
- Beating strong teams
- Winning road games
- Neutral-court wins
- Penalizes bad losses (especially at home)
Quote:
Quote:
"Who did you beat, where did you beat them, and how hard was it to do?"
2) Adjusted Net Efficiency
This is performance-based.
It measures:
- Offensive efficiency (points scored per possession)
- Defensive efficiency (points allowed per possession)
- Adjusted for opponent strength and game location
Net Efficiency = Offensive Efficiency Defensive Efficiency
Example:
- Score 1.12 points per possession
- Allow 0.98 points per possession
- Net efficiency = +0.14
- Opponent quality
- Home vs road vs neutral
Road wins count more than home wins.
Difficulty order:
[ol]
Losses work the same way in reverse.
4) Scoring margin (capped)
Margin of victory is included but limited so teams don't run up scores.
- The cap is 10 points
- Winning by 10 = same as winning by 30 in the model
Everything is calculated per possession so fast teams aren't automatically rewarded over slow teams.
What NET is NOT
It is NOT just:
- AP poll voting
- Strength of schedule alone
- Record alone
- Computer stats alone
- Results + efficiency + opponent strength + location
The committee uses NET as the main sorting tool.
Key quadrant system:
Quadrant 1 (best wins):
- Home: beat teams ranked 130
- Neutral: 150
- Away: 175
- Home: 161+
- Neutral: 201+
- Away: 241+
Simple way to think about the formula
While the exact math is hidden, conceptually it behaves like:
NET
(Win quality + opponent strength + location value)
+
(Adjusted offensive efficiency adjusted defensive efficiency)
with margin capped and everything adjusted by possessions.
Why this matters for Cal (example context)
For a team like Cal:
- Road wins vs mid-tier ACC teams boost NET a lot
- Close losses to strong teams don't hurt as badly
- Blowing out weak teams doesn't help much after +10