Linder Horizon track record ยท Crypto
TIA forecast accuracy on Linder Horizon
Every Horizon forecast for TIA in the last 30 days, checked against the real candle once it closed.
Direction calls, by how early they were made
Bullish or bearish calls, scored right when the candle closed on the called side.
| When the call was made | Calls | Right | Hit rate |
|---|---|---|---|
| In the first 10% of the candle | 0 | 0 | Not enough calls yet |
| Between 10% and a third of the way inRead this row | 15 | 8 | Not enough calls yet |
| Between a third and two thirdsLate in the candle | 17 | 9 | Not enough calls yet |
| In the last third of the candleLate in the candle | 4 | 4 | Not enough calls yet |
How to read this table
- "Between 10% and a third of the way in" is the figure that describes Horizon. Most of the candle was still to come when these calls were made, so there was something left to forecast.
- The two later rows are high because of when the call was made. Two thirds of the way into a candle, the price has usually already moved away from its open, and saying the candle will close on that side is easy. In the last third it is close to reading the result off the chart. A high figure there says little about the forecast, and by then most of the move is over.
- We publish those rows anyway because every reading is recorded, and a track record that hides rows cannot be checked. They are there for completeness, not as a figure to rely on.
- The first 10% is the hardest call, because the candle has barely started. That row usually sits closest to what always guessing "up" would have scored.
- When Horizon is not confident it makes no call. Those readings are counted and shown below. They are scored neither right nor wrong, so the hit rate describes only the moments when Horizon picked a side.
For comparison, 42.9% of these candles closed up, which is what always guessing "up" would have scored. Horizon made no call on 3 readings where it was not confident enough.
Price range
The close landed inside Horizon's forecast range 77.1% of the time (54 of 70 forecasts, the one made at the open included). The range is built to hold about 80% of the time, so a figure near 80% means the ranges are honest, not too wide and not too narrow.
By timeframe
Each row is a checkpoint: how long after the candle opened the call was made. Highlighted rows fall between 10% and a third of the way into the candle. Greyed rows come later, when the candle has mostly formed.
15m candles
Close in range: 80.6% (29 / 36)
1h candles
Close in range: 68.2% (15 / 22)
4h candles
Close in range: 83.3% (10 / 12)
Latest calls and how they turned out
| Candle | Timeframe | After open | Call | Result | Candle move |
|---|---|---|---|---|---|
| 17 Oct, 14:45 | 15m | +10 min | Bullish | Wrong | -0.20% |
| 17 Oct, 14:45 | 15m | +5 min | Bullish | Wrong | -0.20% |
| 17 Oct, 14:30 | 15m | +10 min | Bearish | Right | -1.37% |
| 17 Oct, 14:30 | 15m | +5 min | Bearish | Right | -1.37% |
| 17 Oct, 14:00 | 1h | +30 min | Bearish | Right | -2.78% |
| 17 Oct, 14:00 | 1h | +15 min | Bearish | Right | -2.78% |
| 17 Oct, 13:45 | 15m | +5 min | Bearish | Wrong | +0.68% |
| 17 Oct, 13:30 | 15m | +10 min | Bearish | Wrong | +0.34% |
| 17 Oct, 12:00 | 4h | +2h | Bullish | Wrong | -1.69% |
| 17 Oct, 07:00 | 1h | +45 min | Bearish | Right | -1.22% |
| 17 Oct, 07:00 | 1h | +40 min | Bearish | Right | -1.22% |
| 17 Oct, 07:00 | 1h | +30 min | Bearish | Right | -1.22% |
How this is scored
- A reading is bullish when Horizon gives the candle more than a 65% chance of closing above its open, bearish below 35%, and no call in between.
- A call is right when the candle closed on the called side. A candle that closed exactly at its open counts neither way.
- The forecast made at the open makes no direction call, so it is scored for its price range only.
- The figures cover forecasts that were run on the site, so busy markets weigh more. They are recalculated about once an hour.
Updated: 17 Oct, 16:41 UTC. Past results do not guarantee future results. Linder explains what the data says so you can learn to read it; it is not financial advice, and the decision and the risk stay with you.