The grade system: MEASURED, PLAUSIBLE, LORE

It is Monday, 08:55 ET (20:55 ICT, winter), half an hour before the NQ cash open. Three messages reach you in five minutes. A chat says: “Long the overnight low, it holds most of the time.” A video says option walls act as a ceiling and a floor. A textbook says a normal curve puts about 68% of outcomes within one standard deviation of the average. All three sound equally sure. On this site they carry three different badges. This lesson teaches you to read a badge in a second, and to grade an unbadged claim before you risk money on it.

Key terms

Grade (mức bằng chứng): a label for how well a claim about markets has been tested Definition. Four grades are coloured: Measured, Measured: no edge, Plausible and Lore.

Baseline (mức nền): how often the same outcome happens without the signal Definition. A claim only matters if it beats its baseline.

Scope chip (phạm vi): the market the evidence comes from Definition. It reads NQ, Nasdaq index, Other market or One trader’s study.

Terms you need: A placebo level is a random price at the same distance from the open as a real level (EV1.04). Dealer gamma describes how options dealers hedging customer options can dampen index moves when long gamma or amplify them when short gamma; a gamma-exposure chart maps that hedging across strikes (OPF course). R is one unit of initial trade risk, ATR is average daily range over 14 days, and pp means percentage points.

Chart first: how a claim earns its badge

Look at the two middle leaves. Neither beat its baseline. Our lab tested one with a script, so it is Measured: no edge. One trader tested the other, so it stays Lore.

How it works

The four coloured grades

• Measured MeasuredOther market: peer-reviewed, or replicated by a lab script on a stated sample, and the effect is there. Example: when dealers are short gamma, SPX tends to trend more into the close (Baltussen et al., 2021). The chip says Other market: SPX, so for NQ it is extrapolated.
• Measured: no edge Measured: no edgeNasdaq index: tested by our lab or in peer-reviewed work, and no better than its placebo. Example: pre-market levels on 721 sessions were respected at the placebo rate.
• Plausible PlausibleNasdaq index: a mechanism plus partial support, such as a working paper or one trader’s study. Example: an opening-range breakout on the Nasdaq-100 earned about +0.13R a trade before costs in one replication of a working paper, and about zero after costs.
• Lore Lore: untested, or tested and not supported. Lore means “no edge shown”, not “proven useless”. On this site Lore appears only inside a labelled box:

Myth / hypothesis: “Absorption at a level reverses price.” LoreOne trader's study In a public NQ study, bars with absorption reversed 51.1% of the time and bars without it 51.1%.

The scope chip: where the evidence lives

A Measured badge without its chip is half a sentence. NQ means the test used NQ futures. Nasdaq index means the Nasdaq-100 index or QQQ; the futures themselves were not tested. Other market means SPX, ES, stocks or another field. One trader’s study means one practitioner’s public, pre-registered research: nobody else has checked it, its intervals ignore clustering within sessions (multiple touches on the same day are not independent), and it caps at Plausible.

In our lab’s census of 66 concept notes, only 2 are Measured with no weaker part, both on SPX or stocks, none on NQ Definition.

The grey tags

Grey badges are not evidence. Definition Definition is a definition, contract fact or arithmetic (“one σ covers about 68% of a normal curve”). Model Model is the output of a stated model, such as a gamma-exposure chart. Context Context is a ruler for orientation, with no edge claimed. House rule House rule is an academy process rule, such as trading micro contracts only in the simulator. Firm rule Firm rule is a firm’s published term on a date:

A firm rule says nothing about where price goes, so it never carries an evidence colour.

The lab’s three rules

1. Working papers cap at Plausible, however strong they look. On 2026-11-06 our lab moved “high dealer gamma damps short-term volatility” from Measured to Plausible for this reason alone; nothing had refuted it.
2. A practitioner study never upgrades a claim on its own. One trader’s study can make a claim Plausible or keep it Lore. Only a lab replication or peer-reviewed study can make it Measured.
3. Lore is not useless. It is a hypothesis nobody has shown working yet. Its vocabulary still helps you describe the auction and place stops; you just do not size a trade on it.

Never upgrade without new evidence, and say “not yet measured” when no test exists, never “no effect”.

What the data says

Worked example: grading “long the overnight low”

Take the chat message from the hook and run it through the tree (illustrative prices, not a real session):

1. Tested against a baseline? Yes, in our lab’s level-families test.
2. Beat the baseline? No. The overnight low was respected 18% of the time vs 20% for its placebo, a lift of −1.7 pp [−4.9, +1.6] over 554 touches Measured: no edgeNasdaq index.
3. Who ran the test? Our lab, with a script, on QQQ and SPY proxies. Badge: Measured: no edge, scope Nasdaq index.

Now the trade. Say the overnight low is 20,412.50 and you plan a long there with a stop at 20,405.00: 7.50 points (30 ticks of 0.25). At $5 a tick one NQ risks $150; one MNQ ($0.50 a tick) risks $15, before costs. The badge does not forbid the trade. Remember that the 18% badge measures how often price bounced from the level (respect rate), not how often a trade wins. Size assuming the level gives no edge, require a written confirmation rule (M4.09), and measure your own trade win rate in your journal.

The mirror and the failure

The mirror. “Short the overnight high” grades the same way: respected 18% vs 19% for the placebo, −1.1 pp [−3.8, +1.9] over 683 touches Measured: no edgeNasdaq index. A short at 20,487.50 with a stop at 20,495.00 risks the same 30 ticks: the level is a place to look, not a reason to act.

The failure: reading the colour without the chip. “Dealer short gamma goes with stronger late-day momentum” MeasuredOther market reads “Measured”, so you lean on it in NQ into the close. But the evidence is SPX, 1996 to 2020. On NQ, one trader’s test (+5.9 pp [−0.1, +11.8]) and our lab’s 45-session check were both inconclusive.

Sort the claims yourself

Each card below states a common market hypothesis or definition to classify—guess its grade first, then flip it to see the numbers and the source.

Do it on the platform

Prop-firm note. A daily loss limit or a maximum loss limit is a Firm rule, never evidence. It caps what a placebo-rate trade can cost you on a funded or evaluation account; the PROP course covers each rule with its dated card.

Common mistakes

Common mistake: treating a high hit rate as an edge. “Revisited 85%” means nothing until you see the baseline; if random prices at the same distance score 84%, the level added about one to two percentage points (+1.9 pp [−0.5, +4.3]), and the interval includes zero.

Common mistake: reading “Measured” and dropping the chip. An SPX result applied to NQ is extrapolated until someone tests it on NQ.

What we don’t know yet

• Whether any pre-market level family beats its placebo on a long multi-year NQ futures tick sample: not yet measured (a 46-session 5-minute check on NQ and ES futures found levels respected no more than placebo, −6.1 pp).
• Whether dealer gamma helps on NQ’s late-day moves: inconclusive so far, because the samples are too small.

Recap

• Coloured badges grade evidence; grey badges label definitions, models, context, house rules and firm rules.
• Measured, Measured: no edge and Plausible always carry a scope chip; read the chip with the colour.
• The same null result is Measured: no edge when our lab or a journal ran it, and stays Lore when only one trader did.
• Lore means no edge shown, and it appears only inside a labelled box.

Further reading

• Baltussen, Da, Lammers and Martens (2021), “Hedging demand and market intraday momentum”, Journal of Financial Economics.
• Cont, Kukanov and Stoikov (2014), “The Price Impact of Order Book Events”, Journal of Financial Econometrics.
• Order Flow, Measured: All claims. A public, pre-registered practitioner study on NQ, 2018–2026; read it with the three caveats in the scope-chip section.
• Our lab: pre-market level families vs placebo (721 sessions, 226 tests); the one-day implied band on the Nasdaq-100 (2001–2026); and the grade census of the lab’s 66 concept notes.

Prerequisite: EV1.02 (reading a hit rate against its baseline). Next: EV1.04 shows you how to build the placebo that turns a hit rate into a test, so you can grade a level yourself.

Educational content only. Not financial advice. Futures trading involves substantial risk of loss.