Skip to content
Prediction mode

A dated, falsifiable call. With a confidence score and the reasons.

Ask a yes/no or which-way question with a real deadline — a match, an election, an interview outcome, a launch date. Prediction mode validates the question, casts the relevant chart + transits, then runs a bull case and a bear case in parallel before an arbiter weighs them into a single dated verdict with a confidence level. Every prediction is logged so you can hold it to account later.

Model: Adversarial sides → arbiterLatency: ~30 secondsMax tier

Best for

Time-bound yes/no or which-way questions — sports, elections, a dated decision — where you want a committed call, not a hedge.

What goes in. What comes out.

What goes in

  • A specific question with a clear resolution date (e.g. "Will I get the offer by 30 June?").
  • Your chart context, or the relevant event/entity charts for sports & elections.
  • The live transits + running dasha for the window in question.

What comes out

  • A single committed verdict — yes / no / which-side — with a confidence level.
  • The bull case and the bear case, each argued from the texts, shown side by side.
  • The arbiter’s reasoning for landing where it did.
  • A logged, dated entry in your Prediction Log so the call can be checked when it resolves.

Under the hood

The exact pipeline this mode runs.

ValidationRejects un-falsifiable / dateless questions up front
Chart castReal ephemeris — event/entity charts + transits
Adversarial sidesBull vs bear briefs argued in parallel
ArbiterWeighs both → dated verdict + confidence, logged

Three worked examples

The kind of question this mode is built for, and the kind of reply you would actually get.

Question

Will I clear my visa interview scheduled for 12 March 2027?

Reply

Verdict: likely yes. Confidence: moderate-high. The bull case leans on Jupiter transiting your 9th (foreign travel / sanction) through the interview window and a running Guru sub-period. The bear case flags Mercury combust on the date — a communication-snag risk in the interview itself. The arbiter lands on yes because the 9th-house support is structural while the Mercury risk is a single-day transit; mitigation — over-prepare the documentation. Resolves 12 Mar 2027.

BPHS 26.5Phaladeepika 12.7

Question

India vs Australia ODI on 2 Feb 2027 — who wins?

Reply

Verdict: India, narrowly. Confidence: moderate. Cast from the match-start chart for the venue. Bull-for-India: the Lagna lord is strong and the 10th (victory) is well-aspected at toss time. Bear: the 6th/8th axis carries a mid-innings collapse risk. The arbiter weights the ascendant strength above the volatility flag — but the moderate confidence reflects how transit-sensitive a single match is. Resolves at match end, 2 Feb 2027.

Prasna Marga 16BPHS 7.32

Question

Should I launch my product on 15 April or wait for the next muhurta?

Reply

Verdict: wait. Confidence: high. Bull-for-15-Apr: the date is inside your strong 11th-lord sub-period. Bear: 15 Apr falls in a Tara-dosha window and the Muhurta agent finds no clean launch lagna that day. The arbiter sides with waiting — the dasha support persists into May, so you lose nothing by taking the cleaner 3 May muhurta. Resolves on launch outcome.

MC 4.12BPHS 87.2

When to use Prediction

  • Your question has a clear yes/no (or which-side) answer and a real deadline.
  • You want a committed call you can check later, not a hedge.
  • You are weighing a dated decision — a launch, a fixture, a vote.

When NOT to use Prediction

  • Your question is open-ended or has no resolution date (use Deep Research).
  • You want a gentle interpretive reading, not a committed call (use Normal).
  • It is a major life decision needing every school’s view (use Max Council).

The other three modes

Same chat. Different depth.

FAQ

Why does it refuse some questions?

A prediction has to be falsifiable. If a question has no clear resolution event or date — "will I ever be happy?" — the mode asks you to reframe it into something that can actually be checked against an outcome.

What does the confidence level mean?

It reflects how strongly the chart structure favours the verdict versus how much hangs on volatile single-day transits. High confidence = structural support; moderate/low = transit-sensitive, treat as a lean not a certainty.

Are predictions saved?

Yes — every prediction is logged with its date and verdict in your Prediction Log, so when it resolves you can see how the call held up. Honest track-keeping is the point.

Which tier do I need?

Prediction is a Max-tier mode. Free and Pro users can buy a one-off prediction credit to try it without upgrading.

Try Prediction on your own chart

Free Kundli first, then ask Panditji anything in Prediction mode.