Tennis · Research roadmap
Research roadmap, not a current capability
Tennis is structurally different from every team sport on this list: continuous scoring, a huge number of events, and per-point data that supports modelling approaches the other sports do not. Nothing described here is running.
Current status
Nothing here is running
- Data collection
- None
- Model
- None
- Published record
- None
- Timeline
- Not committed
Roadmap means roadmap
VigPulse does not run tennis. There is no model, no data pipeline, no graded entry, and no committed date. The only sport currently covered is MLB, that coverage is moneyline only, and it is paper-only.
The record
What a published record would have to contain
| Field | MLB | Tennis |
|---|---|---|
| Graded entries | 33 | 0 |
| Record | 9W-24L | None |
| Paper units | +2,417.5u | None |
| Pre-match CLV sample | 6 | 0 |
| Stake type | Paper only | Nothing staked |
MLB figures are read from the published ledger. Tennis columns are empty because no tennis entry has ever been logged, not because the data is withheld.
Market structure
How tennis differs
Point-level structure enables different models
A tennis match is a sequence of near-independent points with a well-defined scoring structure. That means a match can be simulated from serve and return probabilities, an approach with no clean equivalent in team sports.
Surface is a first-order effect
The same two players have materially different win probabilities on clay, grass, and hard court. Any model that pools results across surfaces will be systematically wrong about specialists.
Volume is very high
Professional tennis runs almost year-round across multiple tours, producing a large number of gradeable events. That is attractive for the same reason baseball is.
Liquidity varies by an order of magnitude
A grand slam final and a lower-tier qualifier are not the same market. Limits, margins, and the number of books pricing an event vary so much that a single approach across all tiers would be a mistake.
Method
What would have to change in the method
The sequence on the methodology page was written for team sports whose outcomes cannot be simulated from first principles. Tennis can be, which changes the modelling step more than any other sport here. Below is what each step would have to become. It is a description of unsolved work.
Six steps, rewritten for an individual sport
not implementedMarket
Match winner is a clean two-way market with no draw, so the de-vigging arithmetic is the simplest of any sport here. The complication is not the market shape, it is that a grand slam and a lower-tier qualifier are priced by entirely different sets of books.
Consensus
Consensus has to be computed per tier rather than globally. Pooling a deeply priced final with a thinly priced early-round match produces an average that describes neither, and the error is invisible unless tier is carried through the pipeline.
Movement
Tennis lines move on withdrawal, injury, and fitness news that has no mandatory reporting standard behind it. Attribution is therefore harder than in baseball, where the lineup release is scheduled and public.
Model
This is where tennis genuinely differs. A match decomposes into a sequence of near-independent points with a fixed scoring structure, so a match can be simulated forward from serve and return probabilities. No team sport on this list supports that approach cleanly.
Opportunity
Surface has to be first-class. The same two players carry materially different win probabilities on clay, grass, and hard court, and any estimate pooled across surfaces will be confidently wrong about specialists in both directions.
Risk
Settlement is the exposure. Retirements and walkovers are graded differently by different books, so an entry can be a win at one book and a void at another. A tier policy is also needed, because integrity risk is not uniform across the tour.
The tier problem
StructuralEvery other sport on this site has roughly uniform market depth across its schedule. Tennis does not. Limits, margins, and the number of books pricing an event vary by an order of magnitude between a grand slam final and a lower-tier qualifier.
A single approach applied across all tiers would be calibrated for none of them, so a tier policy is a prerequisite rather than a refinement.
The settlement problem
StructuralRetirements and walkovers are common enough to matter and are graded differently by different books. A record that does not pin down its settlement rules per book is not reproducible, and a record that is not reproducible is not proof of anything.
Since VigPulse does not record which book an entry came from even in baseball, this gate is further away than it looks.
Difficulties
What makes tennis hard
Known problems
3 listed- Retirements and walkovers create settlement rules that differ between books and complicate grading.
- Injury and fatigue information is far less standardised than in leagues with mandatory reporting.
- Lower-tier events have thin markets, wide margins, and elevated integrity risk.
What building this would require
3 listed- Surface-aware modelling rather than pooled historical records.
- Explicit handling of retirement and walkover settlement rules per book.
- A tier policy determining which events are worth pricing at all.
Readiness
Gates before tennis would be anything but a research note
| Gate | Status | Notes |
|---|---|---|
| Surface-aware ratings rather than pooled historical records | Not started | Not started. Pooling across surfaces is the fastest way to be wrong about specialists. |
| A point-level simulation validated against actual match outcomes | Not started | Not started. The approach is well known; having it working and checked is a different thing. |
| Explicit retirement and walkover settlement rules, per book | Not started | Not started. Without this, a graded record would not be reproducible. |
| A tier policy deciding which events are worth pricing at all | Not started | Not started. Liquidity and integrity risk both vary by an order of magnitude across the tour. |
| Publish a graded record before describing tennis as a capability | Not started | Not started. No tennis entry has ever been logged or graded. |
Comparison
Where tennis sits against the rest
| Sport | Status | Cadence | Primary market | Liquidity |
|---|---|---|---|---|
| MLB | Current · paper | Daily, roughly 15 games | Moneyline | High on major markets |
| NFL | Roadmap | Weekly, roughly 16 games | Point spread | Highest of any US sport |
| NBA | Roadmap | Daily, roughly 8 games | Point spread and total | High |
| WNBA | Roadmap | Seasonal, several games per week | Point spread and total | Lower than NBA |
| Tennis | Roadmap | Near-continuous, year-round | Match winner | Varies enormously by event tier |