First-innings win-probability — HSH Cricket Intelligence

How to buy

Two ways in — pick yours.

For people & teams

Buy credits with a card — no wallet, no crypto.
Get an API key emailed to you instantly.
Call this product with your key — credits debit per call.
Check your balance anytime via Check balance.

Buy credits with card →

For AI agents

Point your agent at the x402 endpoint, or describe the need in plain language.
Pay per call in USDC (Base or Solana) — no signup.
Receive the signal inline, settled on-chain.

View agent catalog →

What you get · 3 layers

The signal, in full.

Pass the match state; receive a complete, model-backed read.

Setting-team win probability

The chance the team batting first goes on to win, from their current first-innings state and match context — honestly scored at AUC 0.76.

setting_team_win_prob

Venue-aware

A venue batting-index feature tells the model whether this is a high-scoring ground or a bowler's pitch — worth six AUC points over match-state alone.

venue_bat_index

Projected-total context

A naive projected total is returned alongside as a reference point for the setting side.

projected_total_naive

Where it comes from

Modeled on thousands of first innings.

Trained on 4,683 real T20 matches from Cricsheet — free, legal, ball-by-ball — then validated on unseen matches. No fragile live-feed dependency; the intelligence is in the model.

Real first-innings states

Trained on the first innings of thousands of completed T20 matches, labelled with the eventual result.

trained

Venue batting-index

A per-venue scoring index — the single biggest lift, from 0.70 to 0.76 AUC.

feature-engineered

Team-strength Elo

Optional Elo inputs sharpen the read when team quality is known.

proprietary methodology

Out-of-sample AUC 0.76

Measured on held-out matches — an honest figure for a genuinely hard problem.

out-of-sample

Who calls it

Use cases.

For autonomous agents

Betting bot

Price the innings break

Have a win-probability read before the chase even starts.

Trading agent

Pre-chase positioning

Take a view on the match while the first innings is still being built.

Dashboard bot

First-innings %

Show a live win read during the first innings, not just the chase.

For people & teams

Bettor

Get in early

Form a model-backed view before the second-innings markets sharpen.

Broadcaster

First-innings graphic

Show a win read while the setting team bats.

Analyst

Study setting scores

Quantify how score, wickets, and venue shape first-innings win odds.

How to call it

Discover, pay, receive.

01 · Discover

Find the tool

Via MCP at dod.hshintelligence.com/mcp or the /x402/catalog.

02 · Pay

Pay per call

Send the x402 payment in USDC on Base. Fixed deterministic price.

03 · Receive

Get the signal

The signal returns inline — ready for your model, lineup, or strategy.

# call the first-winprob tool
curl -H "X-Payment: <x402-usdc-base-payload>" \
"https://dod.hshintelligence.com/mcp" \
-d '{"tool":"hsh-cricket-first-winprob","input":{...}}'

Questions

FAQ.

Why is AUC lower than the chase model?

Because predicting the winner while a team is still setting a total is genuinely harder — the chase hasn't happened yet. 0.76 is honest and useful; the chase model's 0.88 has a fixed target to work from.

What lifted it from 0.70 to 0.76?

Adding a venue batting-index and team-strength Elo — knowing the ground and the sides is worth six AUC points.

What inputs?

Balls, wickets, runs, and optionally venue and team Elo for the sharpest read.

When should I use this vs the chase product?

Use this during the first innings; switch to the chase win-probability product once the chase begins.

First-innings win-probability.