Skip to main content

Sports analytics with the data out in the open.

Sparvi publishes analysis meant to change a decision, not just describe a game. Every finding comes with the data behind it, in an explorer you can filter and search, so you can check it yourself or come to a different conclusion.

BaseballFootball

Latest

NFLNov 13, 2026
Is Jacksonville the best team in football?
42%chance Jacksonville is the NFL's single best team so far
The Jaguars lead the NFL in opponent-adjusted efficiency through week 10, but barely. How sure that is, and what history says about the week-10 leader.

College footballNov 13, 2026
Ranked vs. actually good
76.8%of games picked right by the power rating
When the AP poll and an efficiency-based power rating disagree about a game, the power rating has been right more often. Which ranked teams it doubts now.

BaseballNov 13, 2026
Which Triple-A numbers travel to the majors
r = 0.85exit velocity (EV90), Triple-A to MLB
Swing decisions, whiffs, and exit velocity come up from Triple-A almost intact. Triple-A results don't. For pitchers, Stuff+ travels; run value doesn't.

BaseballNov 13, 2026
Which farm systems make players better?
r = 0.24an org's Stuff+ gains, one year to the next
How much each organization's Triple-A pitchers and hitters improved during the season, and whether the same orgs came out on top the next year.

BaseballNov 12, 2026
Stuff or command? Splitting pitching into what the ball does and where it goes
r = 0.70Location+ from one season to the next
Location+ grades where a pitch crosses the plate. It's a steady trait, but it barely predicts next season, and adding it to Stuff+ doesn't help.

BaseballNov 11, 2026
The Rule 5 board: unprotected Triple-A players whose numbers travel
211unprotected Triple-A pitchers graded
Which Triple-A players clubs left exposed to the Rule 5 Draft, ranked on what carries to the majors: stuff for pitchers, swing decisions for hitters.

BaseballNov 10, 2026
Two weeks of swing decisions tells you something. Two weeks of results doesn't.
r = 0.48two weeks of chase rate vs. the next two weeks
Which short-window Triple-A numbers carry over to the next two weeks, and what that means for building a player development report coaches can trust.

BaseballOct 31, 2026
Midseason pitch changes mostly stick. Lost fastball velocity mostly doesn't.
40%of a lost fastball velocity was still lost the next spring
When a pitch's velocity or movement changes midseason, how much is still there next spring? Movement changes hold; lost fastball velocity mostly returns.

BaseballOct 21, 2026
Stuff predicts next season at least as well as results do
r = 0.30Stuff+ vs. next season's run value
A pitch model built only on velocity, movement, and release predicts next season at least as well as run value, settles sooner, and carries from Triple-A.
No matching items

Tools

Baseball · Tool

Pitch-grade leaderboard

Every MLB and Triple-A pitcher and pitch in 2026, graded on stuff, location, and both together.

 Baseball · Tool

Triple-A explorer

Swing decisions, contact quality, and pitching profiles for every Triple-A player, with league percentiles and recent changes.

 Baseball · Tool

Rule 5 board

Triple-A players likely eligible for the Rule 5 Draft and left off 40-man rosters, ranked on stuff, swing decisions, and contact quality.

 Baseball · Tool

Farm-system scores

Every org's in-season Triple-A improvement on six measures, 2025 and 2026, with intervals, sample sizes, and each org's players.

 College Football · Tool

Team ratings

Every FBS team's power rating and opponent-adjusted offense and defense, week by week, next to its AP and CFP rank.

 NFL · Tool

Team efficiency

Every NFL offense and defense in EPA per play, opponent-adjusted or raw, week by week since 2016.

 Reference

Glossary

What every number on the site means and how it's calculated.