Nine of twenty earned less than they cost: list price against measured net profit

2026-09-04. Every figure quoted here is a field in a published audit JSON or a sourced line in our catalog file, and the listed audits are linked in place. The prices are the list prices recorded in each audit, last checked on 30 or 31 August 2026. Several have changed since (list prices as checked on 13 September: Lizard 499 USD, ThunderGold Scalper 299, Gold Snap 599, Gold Atlas 379, Prop Firm Gold EA 339, Range Breakout EA 549, Quantum Titan 1,299.99, and OilVector X, which had no price on record then, at 349 against a net of 35.23 USD), so the counts describe that snapshot.

Lizard closed 21,524 trades on gold between 2003.05.05 and 2026.08.16 and finished 183.99 USD ahead. The licence cost 199 USD when its audit last checked the price, on 31 August. The measured result is 15.01 USD short of the price of the software that produced it. Twenty of our twenty six default audits carry a list price next to a measured net result. Nine of them finished below their own price.

Where the two numbers come from

They are separate measurements with separate weaknesses, and the comparison is only honest if both are read with their source attached.

The price is a scraped snapshot with a date on it. Each audit carries the list price our market watcher last saw, next to the day it saw it. Seventeen of the twenty name 2026-08-31 as that day, two name 2026-08-30 and one names 2026-09-02.
The result is the net after our own commission retrofit. The figure used here is the net line of the costs block, which is the tester result with commission added back at 3.50 USD per lot per side. The head number of the tester report is a pre retrofit figure and is deliberately not used.
The measurement frame is identical for all twenty. Every one of them ran on a 100,000 USD account with a 0.01 base lot input, on vendor default settings. Nineteen span 2003.05.05 to a date in August 2026, and Range Breakout EA with Range Filters starts 2017.01.01 because its symbol basket reaches back no further.
The price is paid once and the result took years to accumulate. That asymmetry is the whole point of writing them side by side. Nothing in an audit charges the buyer for the software.

The nine that did not clear their own price

Three of the nine never reached zero. Scalping Robot Pro is the widest miss at a net of -53,234.68 USD against a 1,599 USD licence. Logan follows at -24,334.65 USD against 289 USD, and Smart Gold Hunter at -1,418.55 USD against 299 USD. For these three the question of paying back a licence does not arise at all.

The other six made money and still finished under the price. Ultimate Breakout System kept 1,271.79 USD from 945 trades against the highest list price in the catalog, 1,699 USD, a shortfall of 427.21 USD. Adaptive Gold Scalper kept 381.48 USD from 3,891 trades against 499 USD. Quantum Titan kept 1,089.53 USD from 1,947 trades against 1,199.99 USD, short by 110.46 USD. Twister Pro Scalper kept 329.72 USD against 399 USD and ThunderGold Scalper 314.65 USD against 379 USD. Lizard was the closest call in that snapshot at 15.01 USD short.

The eleven that cleared it

They clear it by a distance that makes the licence look small. Gold Trade Pro kept 13,446.15 USD from 6,305 trades against a 649 USD price, a margin of 12,797.15 USD. Gold House follows at 11,054.32 USD against 749 USD, Gold Atlas at 8,239.71 USD against 449 USD and Prop Firm Gold EA at 6,769.79 USD against 399 USD.

The two narrowest wins are worth naming, because they are the ones where the price is a real share of the outcome. Smart Gold Impulse kept 926.20 USD from 4,163 trades against the lowest list price among the twenty, 149 USD, a margin of 777.20 USD. Gold Snap kept 1,922.00 USD against 999 USD, a margin of 923.00 USD, which means the licence costs more than half of everything that run kept across a full test window.

One licence, four results

The comparison has a limit that our own catalog demonstrates better than any argument could. Smart Gold Hunter ships four vendor presets, all covered by the same 299 USD purchase, and we measured all four on the same window of M1 gold from 2003.05.05 to 2026.08.17.

Scalper, the default, 10,416 trades, net -1,418.55 USD.
Prop Scalper, 10,053 trades, net -931.58 USD.
Positive RR, 5,065 trades, net 233.16 USD.
Swinger, 8,359 trades, net 2,048.58 USD.

One price, and a spread of 3,467.13 USD between the best and the worst preset. The default is the losing one. Whether that licence was worth 299 USD depends on a choice the buyer makes after paying, which is exactly the part no price tag can tell you.

Versions do the same thing along a different axis. Quantum Titan cost 1,199.99 USD when this was written and we hold three builds of it. Version 3.0 kept 1,089.53 USD, version 2.1 kept 669.83 USD and version 2.2 kept 488.07 USD, so all three finished under the one price the vendor charges. Their windows end six days apart, on 2026.08.23, 2026.08.17 and 2026.08.19, which is enough to forbid reading the differences between them as a version effect.

The price against the account the vendor recommends

Thirteen audits carry both a list price and the deposit the vendor recommends, which our catalog file records as a vendor figure. In seven of those thirteen the software costs more than the account.

ThunderGold Scalper is the sharpest case at a 379 USD licence against a recommended 100 USD deposit. Gold Snap asks 999 USD against a recommended 200 USD. Scalping Robot Pro asks 1,599 USD against 1,000 USD, and both Quantum Titan and Quantum Queen X ask 1,199.99 USD against 1,000 USD. On the other side Logan asks 289 USD against a recommended 2,000 USD and Gold House 749 USD against 2,000 USD. The recommended deposit is a vendor statement and not our measurement, which is why it sits in this section and in no calculation above.

Six audits carry no price at all

The count of twenty is not an accident of rounding. Six of our twenty six default audits have no list price to compare.

Four carry no market block whatsoever, because the build writes that block only when market data is present. They are Waka Waka, Pulse Engine, Quantum Emperor and Market Anomalies EA, the four audits that still come from a foreign broker report.
OilVector X carries a market block that says the archive is missing. The field reads scraper archive missing, and no price line was written.
Quantum Queen carries a price field that is empty. That is the most informative empty field in the catalog. The product page said not available for purchasing anymore when we first read it, the update history came back with zero entries, and the price history is an empty list. A delisted product has no price, and the record of what it once cost went with it.

What this comparison cannot say

The result moves with the lot size and the price does not. Every audit tags its own metrics for this, and absolute profit and loss is tagged as not sizing invariant, while trade count and win rate are. All twenty numbers above were produced with a 0.01 base lot input (some EAs still vary their lot sizes; their audits say so). A buyer who trades larger changes the profit figure and leaves the licence fee untouched, and moves the drawdown in percent along with it, because the same tag list marks that one as not sizing invariant either. Read the lot size before you read the payback.

The price is also a single observation, and it moves. By 31 August, four of the twenty carried more than one price in their history; more have moved since. Lizard stood at 699 USD on 2026-08-15 and at 199 USD on 2026-08-24, which is the whole reason its shortfall reads 15.01 USD instead of 515.01 USD. Scalping Robot Pro went 1,799 USD, 1,599 USD, 1,299 USD and 1,599 USD between 2026-08-15 and 2026-08-26. Gold Snap went 799 USD to 999 USD and ThunderGold Scalper 349 USD to 379 USD. A licence quoted here may cost something else today, and the check date next to it is the only day the figure is true for.

The net result, finally, is one settings run over one window on one broker's history. It is not a forecast of what the next twenty three years pay, for the software or for anybody.

Run these checks yourself

Put the price next to the net before you put it next to the drawdown. A product page shows the profit curve and the price on the same screen and almost never in the same sentence. The subtraction takes one second and it is the first thing a vendor leaves out.
Ask which preset the number came from. One purchase can contain four strategies with results 3,467.13 USD apart, and the default is not always the good one.
Check the date on the price. A price without a check date is a claim about the past that is being read as a claim about today.

Every figure above is public in the audit files linked in place, and the category pages list the verdict lamps for free. Your own tester report? The browser check is free.

The honest limits. This is a comparison of two numbers that were never meant to meet. The list price is a scraped observation of a public page on a named day, and the net result is our own measurement at a size we chose, after a commission we added ourselves. Neither figure includes what a real buyer also pays, namely the spread of a live broker, the cost of the time spent running the thing and the money at risk in the account. Nothing here says a cheap robot is good or an expensive one is bad, and nothing here is investment advice. What it does say is narrow and checkable. Over the exact window each audit covers, at the exact size it traded, this is what the machine kept and this is what the machine cost.

More from the blog

OilVector X review: 1,192 oil trades and $1.52 more than buy and hold

2026-09-05

OilVector X is the only crude oil EA in our catalog. Over 8.9 years of WTI ticks it netted $35.23 after commission, while buying and holding the same 0.01 lots produced $33.71. Costs took 72% of gross profit and 80% of the result came from 8 trading days.

Five comparison files: what the catalog will not compare

2026-09-04

On 4 September 2026 our audits folder held 38 files and five of them contained no run of their own. They compare three versions of Quantum Titan, two of Scalping Robot Pro, four presets of Smart Gold Hunter, two input variants of The Gold Reaper, and two products that turned out to be deal identical. The part worth reading is the gate that runs first, five world fields that must agree before a single number is printed, and the note the engine writes when the test windows end on different days.