Nerd Above the Net
@nerdabovethenet.bsky.social
Narrative-driven analysis, stats, and insights — exploring the game the way it deserves to be talked about. https://nerdabovethenet.substack.com/
It won't replace film study. It won't tell you which rotation to target first.
What it does: translate observable performance into leverage. It removes one of coaching's most expensive mistakes—spending a week preparing for the wrong fight.
More in the post.
What it does: translate observable performance into leverage. It removes one of coaching's most expensive mistakes—spending a week preparing for the wrong fight.
More in the post.
February 4, 2026 at 2:26 PM
It won't replace film study. It won't tell you which rotation to target first.
What it does: translate observable performance into leverage. It removes one of coaching's most expensive mistakes—spending a week preparing for the wrong fight.
More in the post.
What it does: translate observable performance into leverage. It removes one of coaching's most expensive mistakes—spending a week preparing for the wrong fight.
More in the post.
The model shows you:
→ Which skills prop them up vs cost them
→ Where your edges actually exist
→ How volatile those edges are (stability matters)
→ Which 2-3 fights are worth picking
Most of the board is noise. A few rows light up. That's the short list.
→ Which skills prop them up vs cost them
→ Where your edges actually exist
→ How volatile those edges are (stability matters)
→ Which 2-3 fights are worth picking
Most of the board is noise. A few rows light up. That's the short list.
February 4, 2026 at 2:26 PM
The model shows you:
→ Which skills prop them up vs cost them
→ Where your edges actually exist
→ How volatile those edges are (stability matters)
→ Which 2-3 fights are worth picking
Most of the board is noise. A few rows light up. That's the short list.
→ Which skills prop them up vs cost them
→ Where your edges actually exist
→ How volatile those edges are (stability matters)
→ Which 2-3 fights are worth picking
Most of the board is noise. A few rows light up. That's the short list.
This framework translates opponent performance into decision-relevant terms.
It identifies which edges are decision-relevant and which ones disappear under pressure.
Not 'they're bad at digging'—but 'their in-system dig conversion edge means long rallies will be expensive.'
It identifies which edges are decision-relevant and which ones disappear under pressure.
Not 'they're bad at digging'—but 'their in-system dig conversion edge means long rallies will be expensive.'
February 4, 2026 at 2:26 PM
This framework translates opponent performance into decision-relevant terms.
It identifies which edges are decision-relevant and which ones disappear under pressure.
Not 'they're bad at digging'—but 'their in-system dig conversion edge means long rallies will be expensive.'
It identifies which edges are decision-relevant and which ones disappear under pressure.
Not 'they're bad at digging'—but 'their in-system dig conversion edge means long rallies will be expensive.'
You built your plan around a weakness without asking what sits downstream of it.
If they've built resilience out of system—stronger pins, cleaner emergency tempo—then 'getting them OOS' isn't a win condition. It's just a state of play.
If they've built resilience out of system—stronger pins, cleaner emergency tempo—then 'getting them OOS' isn't a win condition. It's just a state of play.
February 4, 2026 at 2:26 PM
You built your plan around a weakness without asking what sits downstream of it.
If they've built resilience out of system—stronger pins, cleaner emergency tempo—then 'getting them OOS' isn't a win condition. It's just a state of play.
If they've built resilience out of system—stronger pins, cleaner emergency tempo—then 'getting them OOS' isn't a win condition. It's just a state of play.
Example: you circle their OH1 because she's hitting .120 in transition. You star their libero because she shanks jump floats.
Then the match starts and the scoreboard doesn't move the way it's supposed to.
Not because you were wrong. Because you were incomplete.
Then the match starts and the scoreboard doesn't move the way it's supposed to.
Not because you were wrong. Because you were incomplete.
February 4, 2026 at 2:26 PM
Example: you circle their OH1 because she's hitting .120 in transition. You star their libero because she shanks jump floats.
Then the match starts and the scoreboard doesn't move the way it's supposed to.
Not because you were wrong. Because you were incomplete.
Then the match starts and the scoreboard doesn't move the way it's supposed to.
Not because you were wrong. Because you were incomplete.
The challenge isn't identifying what your opponent does well or poorly.
It's knowing which of those things you can realistically exploit—and which ones carry enough leverage to be worth the effort.
Not every weakness is worth your time.
It's knowing which of those things you can realistically exploit—and which ones carry enough leverage to be worth the effort.
Not every weakness is worth your time.
February 4, 2026 at 2:26 PM
The challenge isn't identifying what your opponent does well or poorly.
It's knowing which of those things you can realistically exploit—and which ones carry enough leverage to be worth the effort.
Not every weakness is worth your time.
It's knowing which of those things you can realistically exploit—and which ones carry enough leverage to be worth the effort.
Not every weakness is worth your time.
It reframes the conversation:
→ "We need to work on everything" becomes "here's where our resources yield most impact"
→ Recruiting debates become explicit tradeoffs
More in the post. This one's practical—about using the framework to actually make decisions.
→ "We need to work on everything" becomes "here's where our resources yield most impact"
→ Recruiting debates become explicit tradeoffs
More in the post. This one's practical—about using the framework to actually make decisions.
January 27, 2026 at 4:13 PM
It reframes the conversation:
→ "We need to work on everything" becomes "here's where our resources yield most impact"
→ Recruiting debates become explicit tradeoffs
More in the post. This one's practical—about using the framework to actually make decisions.
→ "We need to work on everything" becomes "here's where our resources yield most impact"
→ Recruiting debates become explicit tradeoffs
More in the post. This one's practical—about using the framework to actually make decisions.
This framework translates performance into decision-relevant terms.
Not "your dig percentage is bad"—but "your dig quality is costing 16 points of win probability, and improving it gives the largest single-leverage payoff."
Not "your dig percentage is bad"—but "your dig quality is costing 16 points of win probability, and improving it gives the largest single-leverage payoff."
January 27, 2026 at 4:13 PM
This framework translates performance into decision-relevant terms.
Not "your dig percentage is bad"—but "your dig quality is costing 16 points of win probability, and improving it gives the largest single-leverage payoff."
Not "your dig percentage is bad"—but "your dig quality is costing 16 points of win probability, and improving it gives the largest single-leverage payoff."
Context: elite teams aren't good at everything. They're elite in dig quality and passing. Attack/block volume? Near average.
They're not swinging more. They're better at stabilizing rallies and generating clean offensive opportunities.
They're not swinging more. They're better at stabilizing rallies and generating clean offensive opportunities.
January 27, 2026 at 4:13 PM
Context: elite teams aren't good at everything. They're elite in dig quality and passing. Attack/block volume? Near average.
They're not swinging more. They're better at stabilizing rallies and generating clean offensive opportunities.
They're not swinging more. They're better at stabilizing rallies and generating clean offensive opportunities.
Here's the counterintuitive one: fix ALL below-average skills to average.
Gain? +7.0%
Less than improving dig quality alone.
Why? Improvements interact. Spreading resources dilutes impact on your most critical weakness. Focus matters.
Gain? +7.0%
Less than improving dig quality alone.
Why? Improvements interact. Spreading resources dilutes impact on your most critical weakness. Focus matters.
January 27, 2026 at 4:13 PM
Here's the counterintuitive one: fix ALL below-average skills to average.
Gain? +7.0%
Less than improving dig quality alone.
Why? Improvements interact. Spreading resources dilutes impact on your most critical weakness. Focus matters.
Gain? +7.0%
Less than improving dig quality alone.
Why? Improvements interact. Spreading resources dilutes impact on your most critical weakness. Focus matters.
Run the what-if scenarios:
Improve dig quality from bottom-tier to average: +16.6% win probability. One skill moves you from scraping by to genuinely competitive.
Improve passing to average: +8.8%
Improve serving: +4.1%
Improve dig quality from bottom-tier to average: +16.6% win probability. One skill moves you from scraping by to genuinely competitive.
Improve passing to average: +8.8%
Improve serving: +4.1%
January 27, 2026 at 4:13 PM
Run the what-if scenarios:
Improve dig quality from bottom-tier to average: +16.6% win probability. One skill moves you from scraping by to genuinely competitive.
Improve passing to average: +8.8%
Improve serving: +4.1%
Improve dig quality from bottom-tier to average: +16.6% win probability. One skill moves you from scraping by to genuinely competitive.
Improve passing to average: +8.8%
Improve serving: +4.1%
Your team mid-season: passing is solid, blocking is okay. Everything else—dig quality, freeballs, digs per rally—below average or bottom-tier.
The question isn't "what are we bad at?" It's "which weaknesses are actually expensive?"
The question isn't "what are we bad at?" It's "which weaknesses are actually expensive?"
January 27, 2026 at 4:13 PM
Your team mid-season: passing is solid, blocking is okay. Everything else—dig quality, freeballs, digs per rally—below average or bottom-tier.
The question isn't "what are we bad at?" It's "which weaknesses are actually expensive?"
The question isn't "what are we bad at?" It's "which weaknesses are actually expensive?"
One pattern kept emerging: volume matters less than the conditions that generate it.
Distribution, blocking, attacking—all the same. The model rewards sequences, not isolated events.
Distribution, blocking, attacking—all the same. The model rewards sequences, not isolated events.
January 20, 2026 at 3:02 PM
One pattern kept emerging: volume matters less than the conditions that generate it.
Distribution, blocking, attacking—all the same. The model rewards sequences, not isolated events.
Distribution, blocking, attacking—all the same. The model rewards sequences, not isolated events.
Calibration: when it assigns a 70% win probability, teams should win ~70% of the time. Otherwise the numbers are decoration, not insight.
It held up. Which means the contradictions matter.
It held up. Which means the contradictions matter.
January 20, 2026 at 3:02 PM
Calibration: when it assigns a 70% win probability, teams should win ~70% of the time. Otherwise the numbers are decoration, not insight.
It held up. Which means the contradictions matter.
It held up. Which means the contradictions matter.
The good kind of volume—controlling rallies, forcing scrambles—shows up in other metrics: freeballs received, opponent out-of-system percentage, in-system digs.
The model doesn't penalize aggression. It penalizes survival.
Before showing any of this, the model had to prove it was honest.
The model doesn't penalize aggression. It penalizes survival.
Before showing any of this, the model had to prove it was honest.
January 20, 2026 at 3:02 PM
The good kind of volume—controlling rallies, forcing scrambles—shows up in other metrics: freeballs received, opponent out-of-system percentage, in-system digs.
The model doesn't penalize aggression. It penalizes survival.
Before showing any of this, the model had to prove it was honest.
The model doesn't penalize aggression. It penalizes survival.
Before showing any of this, the model had to prove it was honest.
Not because attacking is bad, but because high volume often captures rallies that won't resolve. More swings without finishing = more chances for something to break down.
January 20, 2026 at 3:02 PM
Not because attacking is bad, but because high volume often captures rallies that won't resolve. More swings without finishing = more chances for something to break down.
he gap between elite and struggling? Over 70 percentage points in estimated win probability. No other metric approaches that.
This isn't about spectacular plays. It's about digs that consistently return rallies to neutral or better.
Attack volume showed up negatively.
This isn't about spectacular plays. It's about digs that consistently return rallies to neutral or better.
Attack volume showed up negatively.
January 20, 2026 at 3:02 PM
he gap between elite and struggling? Over 70 percentage points in estimated win probability. No other metric approaches that.
This isn't about spectacular plays. It's about digs that consistently return rallies to neutral or better.
Attack volume showed up negatively.
This isn't about spectacular plays. It's about digs that consistently return rallies to neutral or better.
Attack volume showed up negatively.
Attacking still matters. But it isn't where the biggest leverage shows up. The conditions that create quality attacks matter more than the attacks themselves.
In-system dig percentage is the strongest single driver. By a lot.
In-system dig percentage is the strongest single driver. By a lot.
January 20, 2026 at 3:02 PM
Attacking still matters. But it isn't where the biggest leverage shows up. The conditions that create quality attacks matter more than the attacks themselves.
In-system dig percentage is the strongest single driver. By a lot.
In-system dig percentage is the strongest single driver. By a lot.
The model isn't choosing between offense and defense. It's pointing out they're collaborating.
Attacking is how points end. Access to in-system offense is where leverage is created.
One thing I expected to dominate—and didn't—was attack quality on its own.
Attacking is how points end. Access to in-system offense is where leverage is created.
One thing I expected to dominate—and didn't—was attack quality on its own.
January 20, 2026 at 3:02 PM
The model isn't choosing between offense and defense. It's pointing out they're collaborating.
Attacking is how points end. Access to in-system offense is where leverage is created.
One thing I expected to dominate—and didn't—was attack quality on its own.
Attacking is how points end. Access to in-system offense is where leverage is created.
One thing I expected to dominate—and didn't—was attack quality on its own.
The strongest drivers aren't about how well teams attack.
They're about how often teams get to attack cleanly.
In-system digs. Freeballs received. Pass rating. Three of the four most influential metrics are doing the same job: deciding whether the offense operates in-system or just survives.
They're about how often teams get to attack cleanly.
In-system digs. Freeballs received. Pass rating. Three of the four most influential metrics are doing the same job: deciding whether the offense operates in-system or just survives.
January 20, 2026 at 3:02 PM
The strongest drivers aren't about how well teams attack.
They're about how often teams get to attack cleanly.
In-system digs. Freeballs received. Pass rating. Three of the four most influential metrics are doing the same job: deciding whether the offense operates in-system or just survives.
They're about how often teams get to attack cleanly.
In-system digs. Freeballs received. Pass rating. Three of the four most influential metrics are doing the same job: deciding whether the offense operates in-system or just survives.
What if standards told us how changes in performance affect the probability of winning?
That's a different class of question. And it requires a different tool: logistic regression.
Not fixed thresholds → probabilities.
Not one-size-fits-all → adapts to how teams actually win.
That's a different class of question. And it requires a different tool: logistic regression.
Not fixed thresholds → probabilities.
Not one-size-fits-all → adapts to how teams actually win.
January 13, 2026 at 3:13 PM
What if standards told us how changes in performance affect the probability of winning?
That's a different class of question. And it requires a different tool: logistic regression.
Not fixed thresholds → probabilities.
Not one-size-fits-all → adapts to how teams actually win.
That's a different class of question. And it requires a different tool: logistic regression.
Not fixed thresholds → probabilities.
Not one-size-fits-all → adapts to how teams actually win.