Synthetic Time Travel™
Watch the future play out. Then check the receipt.
Public forecasts, simulated as stories with a cast, a timeline and the odds, each sealed before the outcome is known. Then point the same engine at your next meeting.
See all forecasts → Preview your meeting →
Sealed forecasts
All forecasts →
Generated imageSimulated forecastResolves todayCentral banks · Simulated forecast
Will Tokyo raise in October? The simulation says 43%
The question Bank of Japan October 2026: does the Policy Board raise the policy rate at its Oct 29-30 meeting (decision Oct 30)?
Simulated by Claude Sonnet 4.5.
In the simulation's most likely path the Policy Board waits again on October 30 while the yen and JGBs hold their breath; a raise is 43%, uncalibrated.
Generated imageSimulated forecastResolves in 6 daysCentral banks · Simulated forecast
Will the Bank of England move Bank Rate? 38% simulated
The question Bank of England November 2026: does the MPC change Bank Rate on Nov 5?
Simulated by Claude Sonnet 4.5.
In the simulation's most likely path the MPC holds on November 5 after the last inflation prints; a change sits at 38%, uncalibrated.
Generated imageSimulated forecastResolves in 31 daysEnergy · Simulated forecast
Will OPEC+ change first-quarter output? Simulated odds 40%
The question OPEC+ ministerial meeting, late 2026: does the group announce a change to agreed production levels for Q1 2027?
Simulated by Claude Sonnet 4.5.
In the simulation's most likely path the ministers hold production levels for Q1 2027 and Brent slips; a change is 40%, uncalibrated.
Generated imageSimulated forecastResolves in 40 daysCentral banks · Simulated forecast
A December Fed cut? The simulation puts it at 40%
The question FOMC December 2026: does the FOMC lower the federal funds target range at its Dec 8-9 meeting?
Simulated by Claude Sonnet 4.5.
In the simulation's most likely path sticky core inflation keeps the Fed on hold in December, with a cut at 40%, uncalibrated.
Generated imageSimulated forecastResolves in 65 daysElections · Simulated forecast
A Republican Senate? Simulated odds of 73%
The question US midterms 2026: do Republicans hold a Senate majority (51 or more seats, or 50 plus the Vice President) once every race is called?
Simulated by Claude Sonnet 4.5.
In the simulation's most likely path Republicans reach a majority without needing the Vice President; the 73% is uncalibrated.
Generated imageSimulated forecastResolves in 65 daysElections · Simulated forecast
A Republican House? The simulation calls it 48%
The question US midterms 2026: do Republicans hold a majority of US House seats once every race is called?
Simulated by Claude Sonnet 4.5.
In the simulation twenty toss-up districts keep the House undecided for weeks; a Republican majority is 48%, uncalibrated.
Synthetic Time Travel™ — see the meeting before you attend it.
See your next meeting before it happens.
Paste an invite or connect a calendar. Timepoint plays the meeting out with stand-ins for everyone in the room, and returns the likely outcome and its odds, where each person stands, the minute it turns, and three moves toward the outcome you want.
Seed-round partner meeting
- Lead partnerInterested, sceptical on market size
- AssociateChampion. Will ask about burn.
- Turns atMinute 18: the market-size question goes unanswered
- Moves
- Open with the bottom-up market model
- Name the two design partners
- Ask for a partner-meeting date before leaving
Ask the same question two ways.
A general AI writes the pep talk.
I pitch the lead partner on Tuesday. How will it go?
You’ve got this! Lead with your passion, know your numbers, and tell a compelling story. Investors back great founders. Walk in confident.
Timepoint plays the meeting out.
What happens in the room
- 01
34% to a term sheet in two weeks. A range, not a promise: 22 to 47%.
- 02
It turns at minute 18. The market-size question goes unanswered.
- 03
Open with the bottom-up model. The first of three moves toward the outcome you want.
Fictional demo. A simulation with odds and reasons, not a transcript of real people.
Bring the meeting. Leave with a plan.
-
Bring the meeting
Connect Google Calendar, paste an invite, or have your agent send it over MCP. Say what you want out of it.
-
See it play out
The likely outcome and its odds, each attendee’s stance and objections, and the minute the meeting turns.
-
Rehearse it
Take your seat while the simulated attendees push back. Try the three moves before you need them.
-
Settle it
Afterwards, one tap on “What happened?” settles your private forecast, so you can see how the previews hold up for you.
Walk back from the future you want.
-
Pin an outcome with a date
A deal, a job, a grant, a life goal. Timepoint walks back from it and returns a route with milestones.
-
Or an honest “no route”
When it can’t find a way there, it says so, and gives you the nearest outcome you can reach, and why.
-
It keeps watch
Pinning sets a watch on the milestones, so you know when the route is holding and when it isn’t.
Your agent can call it too.
Add Timepoint to any MCP client. It signs in with OAuth in your browser; there’s no key to paste.
{ "mcpServers": { "timepoint": { "url": "https://app.timepointai.com/mcp" } } }
Your meetings, pins and records are private. Nothing about them appears on this site.
What we do publish: forecasts on public events, sealed with a timestamp before the outcome is known and settled in the open. News →
Pick the next meeting that matters. See it first.
Sign up with GitHub or email, buy credits, and preview your first meeting in a few minutes. Each run costs model cost plus a flat platform fee, shown before it runs.