The first half of 2026 produced significant tournament moments — NBA Finals, NCAA tournaments, NFL Draft, World Cup — and equally significant methodological lessons for sports analytics. The value of a mid-year methods retrospective isn’t the highlight reel of what happened; it’s the honest accounting of what our metrics got right, what they missed, and what changes for the second half. Five lessons matter for the rest of 2026. Here they are, without the vocabulary inflation.
Quick read: H1 2026 lessons in 60 seconds
- Lesson 1: Possession value metrics matured into mainstream coverage — EPV and expected threat cited in-flow during World Cup broadcasts without requiring the explainer.
- Lesson 2: Pressure rate continued validating as an NFL playoff predictor; teams above the median in pressure rate through Week 12 went 8-2 in the divisional round.
- Lesson 3: Closing-five lineup data outperformed regular-season net rating as a Finals predictor by a wide margin — a pattern we should have been more confident about heading in.
- Lesson 4: WNBA analytics infrastructure caught up to NBA standards; public tracking-style data is now available and usable at a depth that was unthinkable in 2022.
- Lesson 5: Coaching continuity remained underweighted in major public models; the 2026 college football preseason projections that undervalued it are still paying for it.
The five lessons in detail
Each lesson is a specific update to how we should be reading the second half of 2026. The point isn’t to celebrate what worked; it’s to name the specific coverage shifts that follow.
| Lesson | What we learned | What changes |
|---|---|---|
| Possession value matured | EPV and xT cited in mainstream coverage; broadcast graphics used possession-value tint | Soccer coverage adopts both metrics as baseline vocabulary; the disclaimer era ends |
| Pressure rate validated | Super Bowl LIX and the divisional round showed the metric identifying playoff quality with unusual precision | NFL coverage increasingly cites it; expect it in mainstream Sunday panels by December |
| Closing-five lineup data | Better Finals predictor than overall net rating — and the gap was larger than most public models assumed | NBA coverage shifts toward it; the smart writers already have |
| WNBA infrastructure | Public analytics now matches NBA-style depth on lineups, on/off, tracking-style physical metrics | WNBA coverage catches up in analytical maturity; the gap between men’s and women’s analytics coverage narrows |
| Coaching continuity | Still underweighted in major models; the 2025 season produced multiple examples where teams with stable staffs outperformed talent projections | Need for manual adjustment continues; the public models haven’t fixed it and won’t by August |
Why these five and not others
Five lessons is not the full list of things that happened in H1 2026 analytics. There were dozens of smaller stories — refinements to particular metrics, individual teams that beat their projections, tournament-specific quirks — and any retrospective could stretch to twenty items without much effort. The reason we cut to five is that these are the ones that change how we cover H2. A lesson that doesn’t change future coverage isn’t a lesson; it’s a footnote.
The other candidate lessons we considered and cut: the mid-major college football story (real but too narrow to reshape coverage), the NBA Draft class evaluation (the class is good but we already knew that), the WNBA expansion analytics (interesting but too new to draw stable conclusions from). Those five will get their own pieces when the samples are honest; they’re not in this retrospective.
A reading framework for H2 2026
The seven questions below are the audit we’re going to apply against the second half of the year. Every major H2 tournament and season milestone — NFL playoffs, CFP, WNBA Finals, college basketball non-conference stretch — will get evaluated against them.
| Question | What to do |
|---|---|
| Which metrics validated in H1? | Lean on them in H2; the burden of proof shifts to the skeptic |
| Which metrics underperformed? | Discount them; the burden of proof stays with the supporter |
| Which gaps persist? | Areas for manual adjustment; the models won’t cover them for us |
| What new metrics emerged? | Worth tracking in H2 with modest confidence, not full weighting |
| How has coverage evolved? | Vocabulary maturity check — where has the disclaimer disappeared, and where does it still need to |
| What does the analytical community converge on? | Consensus indicators — the moment a metric earns baseline status |
| What did mainstream coverage adopt? | Cultural penetration check — separate from analytical validity |
The companion read on which metrics earn their place across seasons lives in our durability piece. The vocabulary underlying the retrospective is documented in our field guide.
Frequently asked questions
What was the single biggest H1 2026 analytical story?
Possession value metrics reaching mainstream coverage during the World Cup. Nothing else this year produced the same shift in what a general audience can be expected to absorb without a disclaimer.
What metric most failed in H1?
Public pressing data trailed proprietary tracking by a larger margin than expected during the World Cup. The failure wasn’t methodological; it was availability. Public models had the right theory and the wrong sample size, which is a fixable problem for the next Cup cycle.
What changes for H2 coverage?
More confident use of possession value (soccer), closing-five lineup data (NBA/WNBA), and pressure-rate analytics (NFL). Continued manual adjustment for coaching continuity in college football. Everything else is roughly business as usual.
Where can I read serious methodological coverage?
Our field guide and the related Methods pieces published throughout the year. For broader public data, Basketball Reference, Pro Football Reference, and FBref remain the baseline sources most analysts build on.
The takeaway, in one paragraph
H1 2026 validated several public metrics in real competitive contexts while exposing others as still trailing the proprietary frontier. The five lessons above are the ones that actually change how we should be covering H2 — not the ones that produced the best headlines. The framework we use to evaluate them is the version we apply to any methodological retrospective. For the broader vocabulary, our sports analytics field guide is the natural companion read.
