Benchmark

A benchmark is the plain alternative a rule is compared with, such as holding the same assets and doing nothing. A rule's return only means something next to it.

How Benchmark is calculated

State the benchmark before the test. Tickfloor's benchmark for most tests is an equal-weight position in the same assets, rebalanced daily, with no costs charged against it. The comparison is the difference in annual return, with a p-value from a block bootstrap, then corrected for the number of tests run.

How it is read

A benchmark that has the same market exposure removes the part of the result that was just the market going up or down.

Common mistakes

• Comparing a rule with cash when it was in the market half the time.
• Picking the benchmark after seeing which makes the rule look best.
• Ignoring that the rule paid costs and the benchmark did not, or the reverse.

What Tickfloor tested

Tickfloor's research desk has backtested 678 strategies, net of modelled trading costs. After correcting for the 761 tests run, 0 passed. That does not prove no strategy works. These are historical diagnostics, not validation under Testing Standard v2: the backtest harness predates that standard and has not been re-run to meet it.

• Momentum under realistic costs (study)

Lessons that cover it

• Benchmarks and the dumb control
• Index funds versus stock pickers

• Short definition in the glossary

Related concepts

• Random-entry control
• Backtest
• Multiple testing
• Survivorship bias
• All concepts

General information only. It doesn't consider your objectives, finances or needs. Tickfloor holds no financial services licence and never places trades.

• See a full worked example (free, no account)
• Browse every course lesson by topic