Ask any active retail trader what's making them money this year and they'll have an answer ready in under five seconds. "I do well on breakout setups on the 15-minute. Tuesdays and Wednesdays. Mostly indices." They'll tell you with confidence, because they've been there. They've felt the wins. They remember the feel of getting the breakout right on NQ at 9:42 AM on a Tuesday.
Now show them the actual data. Run a clean attribution on their 600 trades from the last six months. Group by setup type, day of week, instrument, time of day, position size, holding period. Compute net P&L by every dimension after spread and commission. Show them the table.
About 60% of the time, the answer they gave you is wrong.
What the data actually says
From observed data across thousands of trader accounts, here's the pattern we see repeatedly:
- The setups traders think make them money usually don't. The "breakout setup" they're certain works for them is, when measured, a slight money-loser after spread costs. The unglamorous mean-reversion entries they take when nothing else is happening are quietly producing most of their positive R-multiple.
- The instruments they think are their best aren't. They love trading NQ because it's exciting. The data says they make 2.4x more money per trade on the boring forex pair they ignore most weeks.
- The time-of-day they prefer is the time-of-day they lose money. They love the open. Their best risk-adjusted returns are at 11:15 AM, during the lull they always skip to eat lunch.
- The position sizes they're proud of are too big. The 2.5% risk trades produce dramatically worse outcomes than the 0.8% risk trades on the same setup types. They've been over-sizing the very thing they were trying to be aggressive about.
None of this is the trader's fault. Human memory weights recent wins higher than recent losses, dramatic outcomes higher than quiet ones, and emotionally salient trades higher than systematic ones. The mental model you build of your own trading from memory alone is, on average, more wrong than right. Not because traders are dishonest with themselves. Because the human mind doesn't track 600 trades accurately.
What Trade Analytics actually does
Trade Analytics is the post-mortem layer of the A-Trader platform. Every trade you take, live or paper, gets logged automatically with full context: instrument, timestamp, direction, size, entry/exit price, spread, commission, holding period, P&L. Then the analytics engine slices that history by every dimension that matters.
The headline view is the attribution table: net P&L by setup type, by instrument, by day-of-week, by hour-of-day, by holding-period bucket, by position size band. You don't tag your trades with setup type manually. Trade Analytics identifies the structural setup from your entry context (trend pull-back vs. breakout vs. mean reversion vs. news reaction). The tags are automatic. You just review them.
What you see in the first 90 seconds of looking at the table is usually upsetting. The setups you talk about most are quiet money-losers. The setups you'd be embarrassed to talk about (the ones that don't have cool names, the boring ones) are producing your edge. The trader's relationship with their own narrative gets ruptured. Almost everyone goes through a few uncomfortable days before they integrate what the data is telling them.
What changes after the honest mirror
The traders who survive that initial discomfort end up with materially different trading practices within 60 days. Specifically:
- They quit setups they thought worked. The "breakout on 15-minute" they were emotionally attached to gets dropped. Forty fewer trades per month. Higher average quality on what remains.
- They concentrate on what actually pays. If 70% of their net P&L came from one specific setup category, they double the screen time spent on that category. The setup itself doesn't change. The selectivity does.
- They trade smaller windows. If their best two hours produce 80% of their net P&L, the other six hours of the session quietly stop happening. They take more lunch breaks. Their cortisol levels drop. Their win rates rise.
- They size correctly. The data tells them which setup-and-regime combination justifies 1.5% sizing versus 0.7%. Standardizing this gets rid of one of the largest sources of P&L volatility.
None of these changes require a new strategy. They're refinements to the existing strategy based on what the data shows is actually working. Most traders don't run this analysis on themselves because they don't have the infrastructure to do it. Trade Analytics is that infrastructure.
How this affects your Trader Rating
The 14 behavioral signals composing your Trader Rating in Arizet | The Desk include several that respond directly to Trade Analytics-driven refinement:
- Signal 5: setup selectivity. When you stop taking setups that the data shows don't work, your selectivity score rises. Skilled traders show clear "waiting for setup" patterns; you become more like them.
- Signal 8: average R-multiple. Concentration on what pays raises your typical winner/loser ratio. R-multiple is what separates the 10% who make money from the 90% who don't.
- Signal 11: time-of-day discipline. Trading only the windows that produce results lifts this directly.
- Signal 13: strategy adherence. When your strategy actually matches what you do (because you've stripped out the parts that don't work) the gap between intent and execution shrinks. This is one of the strongest predictors of forward-looking trader success.
The composite effect across 100+ trades after a Trade Analytics-driven refinement: typical Trader Rating gains in the 500-1,000 point range within 90 days. Larger than what most strategy changes produce.
Where Trade Analytics fits in the career arc
Trade Analytics is part of the bundled tier on Elite (Trader Rating 4,500+) and included in Master. It's not in the Pro free-apps default because Pro-tier traders are usually still building enough trade volume to make the analytics meaningful. You want 200+ trades in your history before the attribution conclusions start being statistically defensible.
The trader who'll get the most value from Trade Analytics is one who's been trading consistently for 6-12 months and has been telling themselves a story about what works. The data will challenge that story. The good ones use the challenge to refine. The ones who can't accept the data go back to telling themselves the same story and stay stuck in Pro tier indefinitely.
That's why Trade Analytics is less of a "tool" and more of an "evaluation moment" in the career arc. The traders who pass the evaluation moment (who accept what the data shows about their actual trading versus their imagined trading) tend to advance to Elite within a few months. The traders who don't, don't.
The deeper case
Trading professionally is a measurement problem. The retail trading industry has historically sold signals and strategies because those are easy to package and sell. The thing actual professional traders use to improve, honest measurement of what they've been doing, has been almost completely absent from retail tooling.
Trade Analytics is the most boring version of professional infrastructure made available to retail. It doesn't predict anything. It doesn't recommend anything. It just shows you, ruthlessly, what's been happening in your own account. The traders who can handle being shown become better, faster, than the traders who can't.
That's the case. Not better signals. Not better predictions. The mirror that doesn't flatter you, so you can finally see what's quietly working and what's quietly not.