About MarketMithra
Understand momentum, sector strength, participation, risk, and signal quality through transparent market intelligence. We fuse seven transparent NSE-oriented inputs into one structured setup read. Every weight is published. Every Agree / Disagree vote is logged with the read and date for the public record.
Why we built it
Most screeners for Indian retail investors either dump raw data or hide the logic behind a black box. Neither helps you review evidence before you act.
We wanted a tool that shows its work. Not a headline like “must buy”, but a structured read: relative strength, delivery trend, EMA stack, momentum, volume context, news tone, and mean reversion, each scored and weighted into one fusion probability. You review the inputs, then decide.
Agree / Disagree votes are logged with the setup read and date so the community record can be reviewed over time.
Using MarketMithra safely
Short guides for Indian retail investors, integrated with the product, not generic market theory.
How to read Bullish Setup, Watch Zone, and Risk Weakening
These labels describe evidence alignment across seven fused inputs, not buy, sell, or hold commands. Bullish Setup means evidence currently leans positive. Watch Zone means evidence is mixed or not strong enough yet. Risk Weakening means the risk structure is weakening. A label can change when inputs shift, so always check freshness.
How to use the fusion score safely
Fusion score measures how much the seven inputs align in one direction. It is evidence alignment, not win probability or a return forecast. A 70% score does not mean a 70% chance the stock goes up. Use it to see whether evidence is bullish, mixed, or weakening, then apply your own thesis and risk process.
Why data freshness matters
MarketMithra uses NSE-oriented end-of-day data with cached snapshots (typically refreshed every ~30 minutes). A “cached snapshot” or “warming up” state means you may be seeing slightly older reads. Opening Canvas or retrying later gives the latest fused read, so do not treat stale data like live prices.
Why this is not a trading recommendation engine
MarketMithra is decision support for research, not personalised investment advice, not a tip service, and not broker execution. We show weights, drivers, timestamps, and uncertainty so you can challenge the read. Alerts and setup labels are probabilistic signals, not trade instructions.
How to combine with your own research
Use MarketMithra as one layer: check the company, sector, valuation, and your portfolio risk separately. If the setup label conflicts with your thesis, treat the conflict as information, not as a signal to ignore your process. Consult a SEBI-registered professional for advice tailored to you.
Educational research only · not investment advice · see also fusion methodology and the timestamped track record.
The seven signals
Weights sum to 100%. Each indicator scores from −1 (fully bearish) to +1 (fully bullish). The fusion score is P = 0.5 + (weighted_sum / 2). P ≥ 0.55 → Bullish Setup; P ≤ 0.45 → Risk Weakening; otherwise Watch Zone.
IBD-style RS compares the stock to the broader Nifty 500 over 1 year and 3 months, blended 60/40. Names that lag the index score lower. This filters a lot of index-relative noise.
The fraction of traded shares that settle as delivery versus intraday square-off. High delivery into strength often reads as accumulation. Thin delivery spikes can read as speculation.
Full bull alignment (20 > 50 > 200) scores bullish. Inverted stacks score bearish. Mixed stacks stay neutral when the trend structure is unclear.
Price change over 20 sessions, clamped and normalised to [−1, +1]. Simple, observable participation in the prevailing move.
Pairs price versus a 20-day rolling VWAP with volume context. Sponsorship shows up when price holds above VWAP with participation.
Claude Haiku reads recent headlines for the name and scores tone from −1 to +1. Underweighted on purpose: headlines can be late, loud, or promotional.
Contrarian input using 20-day SMA ± 2σ bands. Extended toward the upper band scores bearish. Toward the lower band scores bullish. Counterweights trend-heavy legs.
Data sources
- →Price & volume history: NSE-first via bhavcopy, with yfinance fallback (.NS tickers)
- →Delivery %: NSE delivery regime data
- →News sentiment: recent headlines scored by Claude Haiku
- →Model synthesis: Claude Haiku narrative with all seven indicator scores as context
Data is cached for 30 minutes. All prices are in INR. End-of-day snapshots refresh on the MarketMithra cache schedule, not streaming ticks.
Disclaimer
MarketMithra is an educational research tool. It is not investment advice, a recommendation, or a solicitation to buy or sell any security. The operator is not a SEBI-registered investment adviser or research analyst. Do your own due diligence, consult a SEBI-registered professional before making investment decisions, and understand that past performance does not guarantee future results. Markets are volatile. You can lose money.
Decision support only. Not investment advice.