Research

The research behind the sentiment.

Every TradeIntel sentiment model is built on daily price data from 2015 to the present, and each one draws on well-documented market research. This is the thinking behind each read — the indicators it weighs and the literature it builds on.

2015–now
Daily price history behind every model
4 + macro
Markets read for sentiment
11
Indicators in the recession overlay
Start free →See pricing
Models

Four models, one discipline.

Trend-following where trends persist, mean-reversion where prices overshoot, and a macro overlay behind all of it. Each model is rule-based and transparent — you see the inputs, not just the conclusion.

JSE Top 40 · Equities
Confluence model
Multi-factor
Trend, momentum & volatility

The JSE read is a confluence model: rather than leaning on any single indicator, it scores several independent technical measures and only tilts bullish or bearish when enough of them agree.

The inputs span the long-term trend (200-day), momentum oscillators (RSI, MACD), trend strength (ADX), the prevailing volatility regime, and a cross-asset momentum rank. Agreement across uncorrelated measures is what firms up the read.

What it reads
Price relative to the 200-day moving average sets the directional lean.
RSI and MACD gauge momentum; ADX gauges whether a trend is actually present.
A volatility-regime check keeps the read in context with the backdrop.
Sentiment only shifts when the combined score clears a fixed threshold.
Supporting research
[1]Jegadeesh & Titman (1993) — momentum
[3]Faber (2007) — trend / moving-average timing
[4]Wilder (1978) — RSI & ADX
S&P 500 · US Index
High-conviction read
Rare
High-conviction alignments only

The high-conviction read is deliberately rare. It waits for an unusual confluence of oversold conditions, a supportive volatility regime, and trend structure before leaning strongly one way.

Because those conditions align infrequently, it favours precision over frequency — most weeks it simply reads neutral.

What it reads
A multi-factor oversold reading must line up with a constructive regime.
The VIX-based regime overlay must not be in a risk-off state.
The strength of the lean is read from price structure, not a fixed rule.
Nothing shifts unless every condition lines up — most weeks it stays neutral.
Supporting research
[5]Lehmann (1990) — short-term reversal
[7]Whaley (2000) — the VIX investor fear gauge
Nasdaq 100 & Silver
Mean reversion
Reversion
VIX spikes & the gold/silver ratio

Two mean-reversion reads. On the Nasdaq, sharp VIX spikes tend to overshoot — the model reads the panic as stretched and leans toward a snap-back. On Silver, extremes in the gold/silver ratio tend to normalise.

Mean reversion is the mirror image of trend-following: it reads when prices have over-extended away from a reference and are likely to revert. The edge is in defining the extreme and the turn precisely.

What it reads
Nasdaq: a VIX spike past a threshold flags a stretched, reversion-prone market.
Silver: a stretched gold/silver ratio flags a reversion candidate.
The read waits for confirmation of the turn, not just the extreme.
Volatility context keeps the read honest — extremes can persist.
Supporting research
[5]Lehmann (1990) — fads & reversal
[6]Jegadeesh (1990) — predictable short-horizon returns
[7]Whaley (2000) — VIX
Macro overlay · US
Recession early-warning
11 indicators
Validated against every US recession since 1971

Sitting behind every read is an 11-indicator recession model — six coincident indicators answering "are we in it now," and five leading indicators answering "is one coming." Each month it counts how many are flashing and reads the result on a single dial.

Back-tested against NBER recession history to 1971, the 5-indicator threshold flagged 93% of recession months out-of-sample with a single false alarm since 2000. US downturns drive global risk-off, so this reading gates sentiment across the platform.

What it reads
0–2 active indicators: healthy expansion.
3–4 active indicators: warning — a risk-off tilt.
5 or more: the threshold that has marked every US recession since the 1970s.
Built entirely from free, public Federal Reserve (FRED) data.
Supporting research
[8]Estrella & Mishkin (1998) — financial variables as recession predictors
[9]Sahm (2019) — the Sahm Rule recession indicator
Methodology

How we build the read.

Every model is built and checked on daily price data from 2015 to the present, out-of-sample where possible. The approach here is TradeIntel's own; the cited papers inform it but do not endorse it.

The strength of each read is drawn from price structure rather than arbitrary thresholds. A VIX-based regime overlay keeps every read in context with whether the market is risk-on or risk-off.

Each model is reviewed against price action over time, and how the read behaves is tracked openly rather than assumed. Nothing here is a recommendation to buy or sell.

Model Inputs
Data window
2015 – present
Daily OHLC price data
Core inputs
Trend · Momentum
Plus volatility & cross-asset
Regime overlay
VIX-based
Risk-On / Neutral / Risk-Off
Macro overlay
11 indicators
Recession early-warning
References

The literature.

The academic and practitioner work our models draw on. TradeIntel's specific implementations and parameters are proprietary; citation does not imply endorsement by the authors.

[1]
Jegadeesh, N. & Titman, S. (1993). Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency. Journal of Finance.
[2]
Moskowitz, T., Ooi, Y. H. & Pedersen, L. H. (2012). Time Series Momentum. Journal of Financial Economics.
[3]
Faber, M. (2007). A Quantitative Approach to Tactical Asset Allocation. Journal of Wealth Management.
[4]
Wilder, J. W. (1978). New Concepts in Technical Trading Systems. (RSI, ADX.)
[5]
Lehmann, B. (1990). Fads, Martingales, and Market Efficiency. Quarterly Journal of Economics.
[6]
Jegadeesh, N. (1990). Evidence of Predictable Behavior of Security Returns. Journal of Finance.
[7]
Whaley, R. (2000). The Investor Fear Gauge. Journal of Portfolio Management.
[8]
Estrella, A. & Mishkin, F. (1998). Predicting U.S. Recessions: Financial Variables as Leading Indicators. Review of Economics and Statistics.
[9]
Sahm, C. (2019). The Sahm Rule recession indicator. The Hamilton Project, Brookings.
[10]
Sharpe, W. F. (1994). The Sharpe Ratio. Journal of Portfolio Management.

Past performance does not guarantee future results. Trading and investing involve significant risk of loss. Nothing on this page is financial advice or a recommendation to buy or sell any financial product. See the full market sentiment inside the platform — start free.

See the sentiment

See the models run on live data.

The research is the backbone. The platform puts the market's sentiment in front of you — read across the JSE, US indices, and the macro backdrop. Start free, upgrade when you are ready.