Skip to main content
Track Mechanical trading system
2 / 22
Search the archiveLibrary
Library

1988issue C051-6

Pessimistic margin ranking for mechanical systems

Public mechanical-trading-system programs can be ranked on a short recent evaluation-window and on longer multi-year samples. The ranking rebuilds return on required margin after a pessimistic restatement of wins and losses, then converts that figure to an annual rate so windows of different length can be compared.

  • A short recent evaluation-window is used so vendors have less room for historical-fitting to the ranking sample.
  • Programs already released to users are treated as a live-window rather than as a reconstructed historical fit.
  • Pessimistic-return-on-margin reduces the win count and increases the loss count by the square root of each observed count, rebuilds net result from average win and average loss, and divides by required margin.
  • A disclosed-system and a black-box-system can share one list, but a score taken before public release is not treated as a post-release live result.
Entries in this reading1 entry
Current trackMechanical trading system322 readings

Ranking released mechanical systems

A mechanical-trading-system is a fully specified procedure that turns rule inputs, market state, and execution constraints into entry, exit, or stand-aside signals over a defined holding period. Public programs of that kind can be ranked on a recent short evaluation-window as well as on longer multi-year samples.

The ranking is presented as covering programs already released to users. The tabulated evaluation-window is therefore treated as a live-window rather than as a reconstructed historical fit.

A window that is harder to retune

A short recent evaluation-window is used so vendors have less room to tune rules to the ranking sample. Historical-fitting is the practice of tuning rules to a known sample so tabulated results look stronger than a later live test.

Disclosed rules and hidden engines

The same ranking list can include a disclosed-system, whose rules can be inspected and independently recoded for testing, and a black-box-system, whose internal rules are hidden from the user and from the ranker.

At least one black-box-system entry is flagged as having been scored before public release, so it is not treated as a post-release live-window result.

Pessimistic return on required margin

The ranking statistic is pessimistic-return-on-margin. It rebuilds return on required margin after subtracting the square root of the win count from wins and adding the square root of the loss count to losses, then multiplying those adjusted counts by average win and average loss.

In the worked example, 20 winning trades and 16 losing trades are restated as 16 wins and 20 losses after that square-root adjustment.

The adjusted net result is divided by required margin and then converted to an annual rate so evaluation-window samples of different length can be compared.

What the ranking does not claim

The compiler is described as not selling the listed systems. The write-up states that tabulated history does not guarantee later success.

Educational research material, not investment advice. Historical source context does not establish present-day performance.
2 of 22 in the Mechanical trading system track
19911-7 pp.Next on Mechanical trading systemFrom a differenced-average sketch to coded rulesDifferencing successive values of a 35-day simple moving average of closes put later inspection on a detrended path, and hand-drawn lines marked trend start and end rather than exact highs and lows.
All readings on this track · 22 readings
  1. 1985Treat a shared mechanical printout as a public script
  2. 1988Pessimistic margin ranking for mechanical systems
  3. 1991From a differenced-average sketch to coded rules
  4. 2003Commodity enthusiasm is not a complete system
  5. 2005Reader surveys as system evaluation checklists
  6. 2008Critiquing unchanged mechanical signals in evolving markets
  7. 2010Read a mechanical trading system catalog as a specification checklist
  8. 2011Specify mechanical system signals as one testable procedure
  9. 2012Treat a packaged mechanical system as one inspectable procedure
  10. 2013Mechanical trading systems as complete signal procedures
  11. 2013Testable cycle rules as one procedure
  12. 2014Evaluate mechanical trading systems as one procedure
  13. 2015Mechanical trading systems as one testable procedure
  14. 2016Evaluating mechanical trading systems as testable procedures
  15. 2016Reading a systems catalog as one mechanical procedure
  16. 2018Mechanical systems as testable signal procedures
  17. 2019Treat mechanical system listings as specification audits
  18. 2019Match the automation stack to mechanical system constraints
  19. 2019Bar size as a mechanical system construction constraint
  20. 2020Interrogating mechanical system listings as one procedure
  21. 2020Mechanical systems still need emotion rules
  22. 2020Comfort, not theory, keeps mechanical rules in use
By fieldSystem design203 readingsBy year1988116 readings
All 322 readings tagged Mechanical trading system
Also on Mechanical trading system5 readings
1982A bounded-risk entry separates a forecast from a trend1984Evaluating mechanical systems beyond net profit1985Treat a shared mechanical printout as a public script1987Evaluating a black-box pyramiding routine with adverse excursion1987A mechanical rank-rotation sleeve for monthly fund leaders