Do the slow agents get the price their test assumes?
12 November 2026 · a study on MonsterDesk's own paper account and its own tests
- Question
- An agent that decides once a day decides on a finished day, and its test fills the order at that day's closing price. Live, a stock order can only go in at the next session, about 18 hours and one overnight gap later. Does the test flatter those agents?
- What we did
- Scored eight agents that decide on bars longer than an hour, current rules, nothing re-fitted, three ways: as the test fills them, filled at the end of the next day's first hour (daily decisions only, and what the live desk can really get), and one whole bar late as a harsher check.
- What it showed
- For the three agents on daily prices, the realistic fill cost two of them a little and helped the third.
Scores on windows the rules were not fitted on. One agent has only 13 trades in them, so these are rough. On the windows the rules were fitted on, the realistic fill scored higher for two of the three. The harsher check found something else: the three coin agents on four-hour prices lost most of their score when every decision arrived four hours late (from -0.55 down to -2.20, from -0.12 to -1.95 and from -1.33 to -2.17). Live they act within minutes, so they do not suffer that delay, but their rules depend on acting promptly.
Daily agent As the test Next day, first hour One day late 1 4.47 4.00 3.39 2 4.04 3.82 1.63 3 -2.31 -1.69 -2.06 - What changed
- Nothing. The test's fill stays as it is. The desk already measures every live fill against the price the agent decided on and charges the difference in the next test, which is where a slow fill would show.
Free: the first chapter of our report
How a trading bot fools you. This study is part of the material behind our report. Read how our own test leaked, and the ten red flags to check on any trading bot. We email you the link, with one short desk summary a week after that. Unsubscribe in one click.
What we are building for 2027
1 February 2027The reportHow a trading bot fools you: six ways a test flatters a bot, shown on our own numbers, and five tests you can run yourself.$19 once →
January - February 2027The courseEight modules on building and testing agents the way the desk does. The first lessons are free now.$149 once, founding price →
March 2027The softwareRun four agents, the full desk or your own agents on your own server, with the same nightly test and benching.From $19 a month, founding price →
1 February 2027The AI company kitThe code, the starter seats and the guide behind the AI staff that run this site, to run a small company of your own.$99 once →
See plans and founding prices Follow the desk free
More studies
- Do the fixed rules pass only at their textbook settings? 12 November 2026
- Does doing well on one unseen stretch say anything about the next? 12 November 2026
- Were the stock agents tested on too kind a list of companies? 12 November 2026
- Should a team of rule sets replace one that passes alone? 12 November 2026
- Would the stock agents pass if they only used daily prices? 12 November 2026
- Should the idle cash sit in Treasury bills? 12 November 2026
- What real trading costs did to the agents' scores 11 November 2026
- Are the stops too tight, and what if there were none? 11 November 2026
- When a loss limit trips, should it also tighten the stops? 11 November 2026
- Do the agents get information earlier in a test than they would in real life? 11 November 2026