Loading market data…

What We Don't Promise

Honest about what M2M can and cannot do. Every limitation listed here is a design decision, not a bug to be fixed.

Explicit limitations

We do not promise winners

M2M is a disciplined analysis system, not a crystal ball. A 75%-confident card that loses is one data point in a distribution, not a failure. The system is designed to be right over many trades, not on any single trade.

We do not give financial advice

M2M is informational and educational. We present analysis, not recommendations. Position sizing, portfolio allocation, and the decision to trade are entirely yours.

Bear path is in shadow mode

M2M evaluates both bullish and bearish setups under the same gates, but as of 2026-05-15 the bear direction is paper-traded only — pending a re-backtest on the 2024-05 → 2026-05 hold-out (CG1 threshold PF ≥ 1.3). Bear-direction cards are visible and tracked; they are not live-executed by the auto-trader until the gate clears. In any market, the regime gates can suppress setups in either direction if conditions are unsuitable — that is the gates working correctly.

We cannot predict black swans

Gap risk, liquidity withdrawal, flash crashes, geopolitical events — these are irreducible uncertainties that no retail-data analysis eliminates. The 88% confidence cap exists precisely because of this irreducible uncertainty.

We do not manage your portfolio

Taking five M2M cards in the same sector is effectively one correlated bet. M2M does not track or manage portfolio-level risk, correlation, or position sizing across your holdings.

Past performance does not guarantee future results

Backtest results, shadow mode outcomes, and historical calibration data describe the past. Markets change. Edges erode. Regimes shift. The calibration audit exists to detect this, but detection is not prevention.

What M2M sees vs. what it doesn't

Understanding the boundaries of our data is as important as the analysis itself.

Open interest is not positions

OI reflects the number of open contracts, not who holds them or why. A spike in call OI could be a bullish bet, a hedged collar, or a market maker's inventory adjustment. M2M infers likely intent from context (sweep vs block, bid vs ask side, unusual volume relative to OI) but these are inferences, not observations.

T+1 clearing settlement lag

Open interest data settles on a T+1 basis through the clearing house. Friday's reported OI reflects Thursday's trading activity. When M2M shows “today's OI,” it is actually yesterday's settled figure. Intraday volume is real-time, but OI is always one day behind.

No dealer identity

M2M infers dealer-side positioning from customer-side convention (when a customer buys, a dealer sells, and vice versa). But we do not know which specific dealer holds which position. Dealer positioning analysis is hypothesis, not observation.

Exchange-aggregated vs clearing-house OI

Options trade on multiple exchanges. Exchange-reported OI and clearing-house-reported OI can disagree due to settlement timing and cross-exchange position transfers. M2M uses exchange-aggregated data from our market-data feed, which may differ from clearing-house figures by small amounts.

Dark pool data is incomplete

Dark pool prints are reported after execution, not before. M2M sees the print (price, size, venue) but not the intent, the full order size, or whether it was the beginning or end of a larger program. Classifying prints as “accumulation” vs “distribution” requires inference from price context and sequence patterns.

Inference vs measurement

M2M measures prices, volumes, and reported OI. Everything else — dealer positioning, institutional intent, squeeze probability, smart money direction — is inference based on measured data. We label inferred values distinctly from measured values throughout the platform so you can tell the difference at a glance.