A strategy fact sheet is one page with a large number at the top and a rising curve underneath it. It looks like a report. Very often it is a sales asset wearing a report's clothes, and what separates the two is not the layout. It is which lines are on the page, and which ones are quietly absent.
Reading one properly takes five minutes and five questions. Ask them in order, and stop as soon as a sheet cannot answer one. Below they are worked through on the numbers this site publishes for the Algomyth engine, the first of its the system, so a complete answer and a dodge sit next to each other.
Everything else is commentary: the Sharpe ratio, the monthly heat map, the smooth curve, the badge that says "verified". Commentary is fine. But if those five answers are missing, the commentary is describing nothing.
A fact sheet is only as trustworthy as its worst-looking line. Drawdown is the one number that never flatters anybody, so if a sheet does not show one, that is not an oversight. It tells you what kind of document you are holding.
This changes the meaning of every other number on the page, and it is the answer most often blurred. A backtest is a simulation over historical data, built with full knowledge of what happened next. A live record is money that actually moved. Both are useful. Only one faced real fills.
Watch for words that dodge the distinction: "results", "performance", "verified track record". An honest simulated sheet says it is simulated, usually with a hypothetical performance disclaimer, because in the US that disclaimer is expected of anyone showing simulated figures. A live sheet names the broker, the account and the dates. A sheet that does neither is choosing not to tell you. Everything published on this site is a backtest, and what that can and cannot prove is in what a backtest actually is.
A result built on 40 trades is a story. A result built on several thousand is closer to evidence. In a small sample, luck and edge look identical, and no statistical vocabulary on the page fixes that.
Two tricks show up here. One is a period quoted instead of a count: "three years of results" sounds substantial until you notice the system took nine trades a year. The other is a count with no period, so you cannot tell whether 2,000 trades came from a decade or three frantic weeks. Our sheet shows 1,164 trades between January 2022 and September 2026, roughly one trading day at a time across more than four and a half years. What makes a sample real is worked out in how many trades a backtest needs.
Gross is the number the platform prints. Net is what survives commissions, exchange fees and slippage, and on a strategy that trades daily the gap is not a rounding error. Most sheets show gross without using the word. One profit figure and no cost assumption means gross.
Our sheet answers this by not having a gross line at all. Every trade is charged $1.90 a round turn plus one tick of slippage before it is counted, about $2.40 all in on a Micro Nasdaq contract, so the only profit figure published is the one that already survived the costs. That is the harder number to publish and the harder one to dress up, which is the entire reason for doing it that way. The assumption is stated so you can substitute your broker's rate.
| Line | How it is derived | Figure |
|---|---|---|
| Period | Day by day, historical data | Jan 2022 – Sep 2026 |
| Account modeled | Mid-size, Medium sizing level | $50,000 |
| Total trades | Strategy Analyzer report | 1,164 |
| Reward to risk | Target ÷ stop, both placed at entry | 3:1 |
| Break-even win rate | 1 ÷ (1 + 3) | 25% |
| Profit factor | Wins ÷ losses, after costs | 1.46 |
| Worst drawdown | Deepest peak to trough, whole period | $5,273 |
| Costs | $1.90 a round turn plus one tick | ≈ $2.40 per trade |
| Net profit | What is left once costs are out | $87,359 |
Hypothetical performance. Every figure in the table comes from a backtest of the Algomyth engine over historical data, not from a live account. Simulated results are prepared with hindsight, carry no financial risk, and cannot fully reflect real execution, slippage or liquidity. Past performance, real or simulated, does not guarantee future results.
A dollar profit with no denominator is not information. That same $87,359 is an excellent four-and-a-half-year result on a $50,000 account and an unremarkable one on $500,000. Futures results also scale with position size, so any profit figure can be doubled by trading twice the contracts, and so can the loss figure.
The line to find is therefore the sizing, not the profit. Ours is published at three levels, Low, Medium and High, and the figures quoted here are the Medium level on a mid-size account, one trade a day at 9:30 AM ET, never held overnight. A sheet that will not say what size produced its number cannot be compared to anything.
Maximum drawdown is the deepest fall from an equity high to the low that follows. It is the closest a fact sheet comes to describing what holding the system would have felt like, which is exactly why it is the line most often left off.
Ours is $5,273, the worst drawdown of the whole period, on the account the report models. Read it next to the profit, not instead of it: it tells you whether you would still have been running the system on the day it printed. Most people who abandon a working system do it inside a drawdown the system later recovers from, which is the subject of maximum versus average drawdown and of our risk page.
The profit line tells you what the system did. The drawdown line tells you what you would have had to sit through to collect it.
The full report is published with the live signal log that archives every new signal with its outcome, and the calculator lets you swap in your own account size, contract count and commission rate.
A sheet that leads with a win rate and nothing else is steering you. A 40% win rate looks poor alone. Set against an average winner three times the size of the average loser, it is the shape of a working system: it loses more often than it wins, and the winners are bigger. Our own sheet leads with the geometry rather than the win rate. The target sits at three times the risk, so break-even arrives at a 25% win rate: right once in four and the system is flat.
The trick runs the other way too. An 85% win rate with an average loser five times the average winner is a system quietly building a bomb. Whenever you see one of these numbers, find the other, and if only one is shown you are looking at the tell described in why very high win rates lie. The figure that folds both together is profit factor, total wins divided by total losses. Ours is 1.46, so $1.46 came in for every $1.00 that went out, after costs, in simulation. Near 1.0 is noise; above about 3.0 on a small sample usually means the test was fitted to its own data, as profit factor explained sets out.
Then check for absences. These are the ones that matter.
The last one is the real dividing line. A fact sheet you can reproduce is a claim. A fact sheet you can only look at is an advertisement. The checklist to hand any seller, us included, is in how to choose a trading bot, and the faster version for sheets that fail in thirty seconds is in the red flags that give away a scam.
Yes, provided it is labeled as one and you can reproduce it. A long backtest with thousands of trades shows how a rule set would have behaved across conditions that have not occurred recently, which six months of live results cannot. What it cannot demonstrate is real fills, real slippage or real liquidity. Treat it as evidence about the rules, not as a track record.
Because the average is only collected by someone who stays in the seat, and the drawdown decides whether you do. An average of a few thousand dollars a month is a total divided by the number of months, not a paycheck. The drawdown describes the worst stretch you would have had to hold through to reach it, so size your account and your expectations around that figure instead.
Check that the start date is not suspiciously convenient and that the sheet covers the hard periods for its market rather than skipping them. Then compare the headline window with the full history: a sheet showing its best seven months in large type and the whole period in small type is telling you which number it prefers. Length and continuity are what protect you, because any short window can be chosen to look good.
Only once you know what was verified. Some services confirm that trades really occurred in a connected account, which is meaningful. Others simply host a chart the seller uploaded, which is not. Read what the badge certifies, check whether the underlying account is live or simulated, and keep asking the same five questions, because a verified sheet with no drawdown line is still a sheet with no drawdown line.
Everything we write comes back to one idea: verify, then decide. The live terminal exists for exactly that.
Unlock Algomyth Live