When has it been a bad time to buy stocks?

In the past, the bad time to open a stock trade was the mildly stressed market, not the most fearful one. In a 2019 to 2025 test, trades opened in Mildly Off markets lost 3.15% on average, and every other regime gained.

Mildly Off was the only regime where the average trade lost money

Mildly Off is one of Regime Card's five market regimes: a little stressed, not alarming.

A trade is a stock that passed a fixed, rules-based screen, bought at the open of the week's first session. A session is one trading day. Each was sold at a stop 10% below its purchase price, at a target 20% above, or on a set date 3 to 10 weeks later. Costs are not counted.

One year carries most of this. 68.2% of the 2019 to 2025 Mildly Off trades were opened in 2022.

2019 to 2025, a test with its rules fixed in advance. 10,185 trades, grouped by the regime in force when each was opened.
2009 to 2018, the earlier period the idea came from. It is shown for comparison and is not a test. 5,848 trades.

Bar chart. In the 2019 to 2025 test, trades opened in Mildly Off markets lost 3.15% on average, the only loss among the five regimes. Risk-Off was the best, with a gain of 3.17%. Neutral gained 0.51%, Mildly On 2.27% and Risk-On 1.31%. In 2009 to 2018, the earlier period, Mildly Off was again the only loss, at 1.30%, and Risk-On was the best, at 1.43%.

Average result per trade, by the regime in force when the trade was opened
PeriodRegimeAverage result per tradeTradesNote
2019 to 2025Risk-Off+3.17%397best
2019 to 2025Mildly Off−3.15%1,897only loss
2019 to 2025Neutral+0.51%4,567
2019 to 2025Mildly On+2.27%2,200
2019 to 2025Risk-On+1.31%1,124
2009 to 2018, the earlier periodRisk-Off+0.75%651
2009 to 2018, the earlier periodMildly Off−1.30%1,454only loss
2009 to 2018, the earlier periodNeutral+1.13%2,608
2009 to 2018, the earlier periodMildly On+1.18%833
2009 to 2018, the earlier periodRisk-On+1.43%302best

See today's regime

1 in 2 Mildly Off trades hit its stop

A stop is a price, set in advance, at which a losing trade is sold.

Dot chart. In 2019 to 2025, 50.4% of trades opened in Mildly Off markets hit their stop, about 1 in 2. In the other four regimes the share was 23.0% to 32.6%.

Share of trades that hit their stop, by regime
RegimeHit their stop
Risk-Off25.4%
Mildly Off50.4%
Neutral28.1%
Mildly On23.0%
Risk-On32.6%
Share of trades that hit their stop, 2019 to 2025. Each row of dots stands for ten trades. In the other four regimes it was between 1 in 4 and 1 in 3.

About 2 in 3 Mildly Off episodes ended in a recovery

An episode is a run of Mildly Off days. It continues through up to 3 sessions at another label.

Dot chart. Of the 61 Mildly Off episodes that had ended, 40 recovered to Neutral or better and 21, or 34.4%, ended in Risk-Off.

How Mildly Off episodes ended
How it endedEpisodes
recovered to Neutral or better40
ended in Risk-Off21
The 61 Mildly Off episodes that had ended, 2009-07-02 to 2026-11-12. One dot is one episode. A 62nd was still running.

After 10 Mildly Off sessions, a fall of more than 10% in QQQ was more common

QQQ is a fund tracking the Nasdaq-100 index. The bars show how often it closed more than 10% below that day's close in the next 60 sessions.

2009 to 2018: 5 of the 17 episodes that reached 10 Mildly Off sessions, against all 2,390 sessions.
2019 to 2025: 5 of 10 episodes, against all 1,760 sessions. 4 of those 5 were in the 2021 to 2022 decline.

Bar chart. In 2009 to 2018, QQQ closed more than 10% lower within 60 sessions after 29.4% of the episodes that reached 10 Mildly Off sessions, against 11.5% of all sessions. In 2019 to 2025 it was 50.0% against 19.3%.

How often QQQ closed more than 10% lower within 60 sessions
PeriodStarting pointShare
2009 to 2018Any session11.5%
2009 to 2018After 10 Mildly Off sessions29.4%
2019 to 2025Any session19.3%
2019 to 2025After 10 Mildly Off sessions50.0%

Since 2009, two episodes ran longer than the one that began in September 2026

Bar chart. The three longest Mildly Off episodes since 2009: 62 sessions from 2018-01-25, 60 sessions from 2022-07-18, and 46 sessions from 2026-09-10 as of 2026-11-12.

The three longest Mildly Off episodes
EpisodeSessions
From 2018-01-2562
From 2022-07-1860
From 2026-09-10, as of 2026-11-1246
Length of the three longest Mildly Off episodes, in sessions, on labels from 2009-07-02. The 2026 episode is counted as of 2026-11-12, the latest close in the study.

What did not work: selling on Mildly Off

2019 to 2025Return a yearWorst fall
Held QQQ the whole time23.0%35.1%
Sold after the 3rd Mildly Off session, bought back on recovery or at the first Risk-Off reading20.8%33.3%
Selling did not beat holding. 2019 to 2025, 20 exits. Worst fall is the largest drop from a high to a later low.

How this was tested

Labels. Labels from before the site launched in 2026 are rebuilt from history, using only data available up to each day. They are not labels published at the time.

Not every year. In 2019, 2023 and 2024, Mildly Off trades averaged a gain.

Rules first. Each test's rules were written down before it ran.

Checked twice. A separate program recomputed every number.

This page describes the past. It is not investment advice or a forecast.

Common questions

Is risk-off a bad time to buy stocks?

In a 2019 to 2025 test, stock trades opened in Risk-Off markets gained 3.17% on average, while trades opened in Mildly Off markets lost 3.15%. 68.2% of those Mildly Off trades were in 2022.

How often do trades hit their stop loss?

In that test, 50.4% of trades opened in Mildly Off markets hit a stop set 10% below the purchase price. In the other four regimes the share was 23.0% to 32.6%.

Does mild market stress usually get worse?

More often it has eased. Of the 61 Mildly Off episodes that ended between 2009-07-02 and 2026-11-12, 40 recovered to Neutral or better and 21 (34.4%) ended in Risk-Off.

What happened after long stretches of market stress?

From 2009 to 2018, QQQ closed more than 10% lower within 60 sessions after 5 of the 17 episodes that reached 10 Mildly Off sessions (29.4%), against 11.5% of all sessions.

Does selling when the market looks stressed beat holding?

It did not in 2019 to 2025. Selling QQQ after the 3rd Mildly Off session returned 20.8% a year against 23.0% for holding, with a worst fall of 33.3% against 35.1%.

Related