How Confidence is calculated

What Confidence measures, the thresholds the model uses for win rate and profit factor, and the exact formulas behind the four rows on the Confidence tab.

Same on every platform Updated 26 July 2026

What this does

Confidence asks one question: do you have an edge, and do you back it? It is the largest part of the VEXA Score, suggested at 40 per cent of the composite.

The Confidence sub-score and what feeds it — Web — The Confidence sub-score and what feeds it

What the model weighs

Confidence is produced by VEXA’s AI rather than by a fixed formula, and the scoring guidance it works from is specific.

Win rate, judged against your trading style. The same percentage means different things to different traders:

Style	Win rate the model reads as good
Scalping	40 to 55 per cent
Day trading	50 to 65 per cent
Swing trading	55 to 70 per cent

Profit factor, on a fixed scale: 2.0 and above is excellent, 1.5 to 2.0 is good, 1.0 to 1.5 is acceptable, and below 1.0 is treated as a problem.

Average P&L, where positive expectancy matters more than the size of any one day, and a rising trend counts for more than a flat one.

Consistency, meaning low variation in daily P&L. Steady beats spiky at the same total.

Direction, meaning whether the recent trading days are improving or deteriorating.

The model is told to be adaptive rather than mechanical. A trader with a 45 per cent win rate and a 3.0 profit factor is scored as highly confident, because the risk-to-reward is doing the work. A trader with a 70 per cent win rate and a 0.8 profit factor is not, because the losers are being allowed to run.

The four rows on the Confidence tab

The Confidence tab of the CDI breakdown shows four component rows. These are an explanation computed on your device over the last five trading days. They do not produce the score. The score comes from the AI. Expect the two to agree in direction and not always in level.

Row	Weight label shown	The value	Arrow up when	Arrow down when
WIN RATE	25% weight	`round(winRate)%`	60 or more	below 40
PROFIT FACTOR	35% weight	profit factor to 2 decimals	2.0 or more	below 1.0
AVG P&L	20% weight	`Σ P&L / trades`, rounded to the dollar	50 dollars or more	below 0
CONSISTENCY	20% weight	`round(consistency)%`	70 or more	below 50

Consistency has an actual formula:

CV          = standard deviation of trade P&L / |mean trade P&L|
consistency = (1 − min(CV, 2) / 2) × 100

A coefficient of variation of 0 gives 100 per cent. A coefficient of 2 or worse gives 0 per cent. Fewer than two trades in the window returns a flat 50, because there is nothing to measure variation against.

Where to find it

Open CDI Breakdown from the sidebar.
Choose the Confidence tab.
Read the ring, then the component rows underneath.

Open the Home tab, the house icon at the far left of the bar at the bottom.
Tap the ring labelled CONFIDENCE.
Read the ring, the contributing factors and the insight.

Open Dashboard from the sidebar.
Tap the ring labelled CONFIDENCE.
Read the ring, the contributing factors and the insight.

Open the Home tab, the house icon at the far left of the bar at the bottom.
Tap the ring labelled CONFIDENCE.
Read the ring, the contributing factors and the insight. Swipe right anywhere on the page to go back.

Open Dashboard from the sidebar.
Tap the ring labelled CONFIDENCE.
Read the ring, the contributing factors and the insight.

What you should see

A number to one decimal place, four component rows with arrows, a set of AI contributing factors each carrying a weight of critical, high, medium or low, and a written paragraph.

With no trades in the last five trading days, the component list is empty and only the ring and the AI text render.

If it doesn’t work

The rows disagree with the ring. They are meant to. The rows look at five trading days on your device. The ring comes from the AI, which reads a wider window, your baseline and your prior scores.
Profit factor shows a very large number. With no losing trades in the window there is nothing to divide by, and the surfaces handle that differently.
Confidence dropped after a profitable week. Check the consistency row. A week made by one outsized day scores worse than the same total spread evenly.

What the VEXA Score is The 0 to 100 headline number, the exact weights that build it from Confidence, Discipline and Intuition, and what feeds each part. Read the CDI breakdown The three tabs, what the component rows are actually measuring, and why they can disagree with the ring above them. How Discipline is calculated The four internal buckets behind Discipline, the four-day recency weighting, and why the Discipline tab uses a 500 dollar loss limit that is not the one you set.

Still stuck? Email support@vexatrade.ai.