Q60, Q75 and Q90 are percentiles of how far a quantum forecast's simulated paths REACH before the horizon ends — read as probabilities, not targets. Q90 is the level it estimates only a ~10% chance of the price ever trading up through, not a price to sell at.
Q60 is the level the model estimates a 40% chance the price trades up through at some point before the horizon ends; Q75 and Q90 are the same idea at higher, less likely levels (25% and 10%). They are measured on the touch rather than on the finish, because a limit order fills the moment price trades through a level — the chance of the price still sitting above it at the end is lower. Read them as a fan of outcomes, not a list of targets: the upper levels sketch the good-case tail, and how far they sit from spot is your uncertainty — which is what should drive position size, not any single percentile.
The most common beginner mistake is reading a high percentile like Q90 as a "sell here" target. It isn't one — it is an optimistic edge with only about a one-in-ten chance of being beaten. The R-multiple take-profit ladder (TP1/TP2/TP3), shown separately, is the actual exit plan.
The confidence band answers a different question, on a different distribution. Its edges (Q05, Q95 for the 90% band; Q25, Q75 for the 50% band) are percentiles of where a path FINISHES at the end of the horizon, which is what CI90 coverage measures. Q60 / Q75 / Q90 above are percentiles of how far a path REACHES along the way. A level a path reaches is always at least as far out as where it ends, so the two families of numbers are related but never interchangeable — a level 40% of paths touch is finished above by noticeably fewer than 40%.
Live example: AAPL last traded around $323.55, and the model's current 3-month Q95 confidence-band edge is $364.36 — the price the model estimates there is only about a 5% chance of finishing above at this horizon (a 95% chance of finishing at or below it). The Q90 the ticker page displays reads a little differently: it marks how far price could reach along the way, which sits further out than the finish level of the same probability. Both are statements about what the model thinks is possible — not a target to sell at and not a promise. See the full AAPL forecast for the current percentile band this attaches to.
On the committed TOP-20 backtest the 90% band (Q05-to-Q95) covered 78–95% of realized prices at 3 months, 71–92% at 6 months and 86–96% at 1 year, depending only on which quarter the forecast started in. We publish every start date rather than one average, because the start date moves the answer more than most readers expect. The cut-points are broadly honestly placed, but individual windows are far worse: NVDA started 31 Dec 2024 covered 2% at 1 year. Check the per-ticker forecast-reliability label before leaning on any single percentile.
No. Q90 is the level the model estimates only about a 10% chance of the price ever trading up through — an optimistic edge, not where it expects price to go. The R-multiple take-profit ladder (TP1/TP2/TP3), shown separately, is the actual exit plan.
Q60 is the level the model estimates a 40% chance the price trades up through at some point before the horizon ends — a cut-point on how far the forecast's simulated paths reach, not a prediction it will settle exactly there.
They come from the same simulated paths but measure different events. The confidence band spans where paths FINISH (Q05 to Q95 for the 90% band); Q60/Q75/Q90 above mark how far paths REACH along the way, so they sit further from spot than the finish percentiles of the same probability. On the committed TOP-20 backtest the 90% finish band covered between 78% and 95% of realized 3-month prices depending on the forecast start date, and per-ticker calibration varies more still.
AAPL analysis shows this metric in context, or browse all S&P 500 tickers.
Educational research only — not investment advice.