MOEasymmetry← All articles
Lessons · 2026-06-10 · 6 min read

5 Signals That Tell You When the Market Is About to Turn — And Why None Works Alone

Track. Study. Wait. Strike.
English อ่านภาษาไทย (Thai)

Source: Financial Wisdom "5 Signals That Tell You When the Market Is About to Turn" (2026-04-21)


The S&P 500 has spent more than 70% of the past 100 years in an uptrend. Yet most traders fail to benefit from it. The reason: they fight hostile market conditions, trading through corrections and getting stopped out repeatedly.

This article summarizes 5 tools for identifying when markets are about to exit hostile territory and return to an uptrend, with connections to the framework MOEasymmetry uses in practice.


Understand the 3 Phases of a Market Bottom First

Before looking at signals, understand the structure of a bottom:

Phase 1 — Capitulation: Sellers dominate. Price falls hard. Bad news is abundant. Fear peaks.

Phase 2 — Accumulation: Sellers begin to exhaust. Buyers start emerging. But buyers don't yet outnumber sellers. Uncertainty remains. Price moves sideways.

Phase 3 — Markup: Bad news diminishes. Buyers take control. Large green candles on elevated volume — this is the real start of the new uptrend.


Signal 1: RSI Divergence at the Low

During a downtrend, when price makes lower lows, RSI normally confirms with its own lower lows.

But near a bottom, divergence often appears: price makes a new low, but RSI doesn't follow (stays flat or rises).

This signals weakening downside momentum — sellers are running out of steam. Note: RSI divergence works more reliably on weekly charts than daily.


Signal 2: Follow-Through Day (William O'Neil / IBD)

Developed by William O'Neil, founder of Investor's Business Daily — the same system MOEasymmetry uses as its market regime gate.

How to count: 1. Day 1: the first day a major index closes above the prior day's high after a correction low 2. The rally attempt remains valid as long as the index doesn't undercut Day 1's intraday low 3. Follow-Through Day = index gains ≥1.5% on volume above the prior day, between Day 4 and Day 7

Critical: Follow-Through Days don't work well in isolation. Sometimes they appear and the market continues lower. The correct approach is using them alongside other signals.

In MOEasymmetry: Follow-Through Day is one component of our distribution-day machine, not the sole signal.


Signal 3: Monthly Reversal Candle

This technique works on monthly charts — suitable for deep corrections.

Conditions: - Price marks the month's low (lowest point of that month) - But closes strong (typically +5% or more) - Volume exceeds the prior month

The signal says: even though sellers pushed price to extreme lows during the month, buyers came back powerfully and closed near the high. A genuine turning point.

Note: Works well on S&P 500 but less reliably on Nasdaq 100. However, both indices tend to bottom at roughly the same time.


Signal 4: 50-Month Moving Average

In deep corrections (not routine corrections), the S&P 500 often finds support at the 50-month MA.

From 1980 to present, when price broke below the 50-month MA and stayed there, it was a major bear market (dot-com crash, 2008 financial crisis). When it just touched and bounced, it was a deep correction within an ongoing bull market.


Signal 5: 10/20 EMA Crossover (Weekly)

Plot the 10-week EMA and 20-week EMA on weekly charts:

This signal is slower than the Follow-Through Day but more reliable because it filters out noise.


The Key: Confluence, Not a Single Signal

The central conclusion of this research aligns with MOEasymmetry's approach:

"The optimal approach is combining other tools with follow-through days."

No single signal works alone 100% of the time. Follow-Through Days sometimes appear before the market continues lower. Monthly reversal candles can be temporary bounces. RSI divergence can persist far longer than expected.

In deep corrections: when 3 or more signals align during the bottoming process, that's when to increase position sizes and trade more aggressively.

In routine corrections: use Follow-Through Day + 10/20 EMA primarily, and be ready to reduce exposure quickly if the market weakens again.


These Tools Work on SET Too

These 5 signals were validated primarily on the S&P 500, but the same logic applies to SET because:

What's different: SET is smaller, so a single large foreign fund flow can override technical signals in the short term — making confluence even more important than in US markets.


What Our Data Shows

We tested all 5 signals against 40+ years of SET daily data (1982–2026, 10,875 bars).

Signal 1: RSI divergence (weekly) When SET weekly price hit a 13-week low but RSI did NOT confirm (bullish divergence, RSI < 45), the 60-day forward return was +4.31% mean / +0.81% median with a 53.2% win rate — versus +0.63% mean / −0.15% median for confirmed lows without divergence. Directional edge is real, but noisy. Use as a first alert, not a standalone trigger.

Signal 2: Follow-Through Day Already embedded in our live regime gate (IBD distribution-day machine). When the market is in Distribution or Correction, no new entries. This rule alone filters out the worst environments.

Signal 3: Monthly reversal candle on SET We identified 34 months (1982–2026) where SET hit a new 3-month intraday low but closed in the top 30% of the monthly range. Forward returns compared to all other months:

PeriodSignal monthsOrdinary monthsWin rate: signal vs ordinary
30 days+1.91% median+0.83% median64.7% vs 54.5%
60 days+2.65% median+1.28% median64.7% vs 56.0%

The +8.7pp win rate boost is real. When sellers pushed SET to a multi-month low but buyers closed it strongly, the 60-day follow-through was significantly better. Not a standalone trade trigger, but a meaningful regime confirmation.

Signal 4: 50-month MA on SET The 50-month MA separated routine corrections from true bear markets: - Above 50-month MA (335 months): 60.3% win rate, +2.73% median 6-month return - Below 50-month MA (142 months): 56.3% win rate, higher but more volatile returns

The 14 cross-below events in our data include every major SET bear market: the 1996 Asia bubble peak (6m forward: −20.2%), the 2008 GFC (−35.4%), and the 2019 pre-COVID break (−15.9%). When SET crossed back above the 50-month MA, the average 6-month forward return was positive. The MA doesn't predict exact bottoms — it identifies when you're in a structurally hostile environment.

Signal 5: 10/20 EMA weekly crossover Tested on SET weekly closes since 1982. The bearish cross is the stronger signal:

Signal60d mean60d medianWin rate
Baseline (all weeks)+2.42%+1.41%
Bullish cross (10w > 20w)+3.35%+2.17%65.9%
Bearish cross (10w < 20w)−0.44%+0.03%50.0%

After a bearish cross, expected 60-day return drops from baseline +1.41% to essentially zero. After a bullish cross, win rate lifts to 65.9%. Use the bearish cross as a "reduce exposure" signal and the bullish cross as a "regime improving" signal — consistent with the article's recommendation to reduce quickly if the market weakens after a Follow-Through Day.


This article summarizes lessons from Financial Wisdom's YouTube channel for educational purposes only — not investment advice. Every signal fails sometimes. Past performance does not guarantee future results.

Get new research by email
Tested across decades. Failures published. Real money.
Subscribe — free
📊 See the live dashboards, the breakout scanner, and the real track record at the MOEasymmetry hub — research, not advice.
← Previous
7 Lessons From 35 Years of Trading — Brian Shannon
งานวิจัยและบันทึกการเทรดส่วนบุคคล ไม่ใช่คำแนะนำการลงทุน · Personal research & trading journal — not investment advice. The author does not provide licensed advisory services.
Home · Articles · Methodology · Track record