The DAX climbs to a fresh high straight into its top alert, and the long yields are lining up behind the same warning

Stocks bounced on Friday, although the major indices still closed the week in the red. The S&P 500 gained 0.4%, the equal-weight version of the index 0.6%, the Dow jumped 1.0%, the Nasdaq added 0.4%, and the Russell 2000 rose 0.9%. The bounce came despite a further rise in yields, which climbed across the curve as investors increasingly dismissed the Treasury’s buyback program as too small against the forces pushing borrowing costs higher, and the odds of a September Fed hike moved up toward 40%. The August flash PMIs were bullish, with growth momentum strengthening while inflation cooled, but the growth signal itself fed the move in rates. Brent edged up to about $94.10, gold spiked 2.25% and silver jumped 1.9%. The week ahead brings the July PCE on Wednesday, Warsh’s Jackson Hole speech on Friday, and the Nvidia and Salesforce reports on Wednesday night.

Two stories from this letter’s recent coverage moved again on Friday, in opposite directions. The real-asset floor kept paying, and both legs of that call are on the record. The gold turn flagged on August 5 and the silver confirmation that followed on August 9 both delivered another up day on top of Wednesday’s surge. The DAX, which the August 15 review named the next candidate for a momentum crossover and which completed that crossover on Thursday, answered with a fresh high instead. The model has not blinked. The German index holds the bearish maximum for a second reading, the rate complex now escalates behind it with both long Treasury yields moving to the edge of the same extreme on the very day yields rose, and underneath, the Nasdaq 100 joins the Composite on the bottoming side, so the thin tech floor flagged on Thursday has broadened by one index.

Our daily analysis filters roughly 45 global markets and condenses every detected cycle into a single Consensus Score between -100 and +100, where +100 is the strongest reading for a possible upward turn and -100 for a possible downward turn. Values beyond ±60 mark the critical zone, the range where the cycle model raises the alarm for a possible trend change. Today the scan shows eleven markets in the critical bottoming zone, where declines could end, and eight in the critical topping zone, where rises could stall or reverse. Let’s take a closer look.

Bottoming cycles

Read more

The summer’s loudest floor call has traveled all the way to the ceiling as the BRICS ETF tops the warning list

Bulls won the week. The July inflation data cooled against June on both the consumer and producer side, following the soft July jobs report from the prior Friday, and expectations for a September Fed hike faded further. The AI boom supplied the rest. A heavy slate of tech earnings from Applied Materials, Cisco, CoreWeave, Nebius and others read bullish on the infrastructure buildout, Cisco spoke of a “networking supercycle”, and several operators emphasized improving returns on their AI capital spending. Workday jumped on Thursday after a report that Silver Lake is considering a buyout, a sign that value buyers are circling the legacy software names the market had written off. The counterweight came from the Middle East, where Brent spiked on Monday and Tuesday as Washington and Tehran drifted further apart before the pressure eased late in the week.

Into that risk-on close to the week, our cycle model turns its warning list international. The day’s story is the BRICS 50 ETF. In early July it carried the model’s strongest bottoming reading, meaning the market was in a time window where the decline could end, and a recovery of more than ten percent off the July low followed. Today the same ETF crosses to the topping side, a time window where the rise could stall or reverse, at the model’s strongest possible bearish reading, with India’s Nifty 50 and the German DAX standing close behind. On the other side of the table the real-asset complex, from gold and silver through the miners to uranium, consolidates the cyclical floor it confirmed in early August. Both legs of that call are on the record: the gold turn flagged on August 5 and the silver confirmation that followed on August 9. The rate complex leans exactly where the week’s inflation data pushed.

Our daily analysis filters roughly 45 global markets and condenses every detected cycle into a single Consensus Score between -100 and +100, where +100 is the highest score the model gives for a possible upward turn and -100 for a possible downward turn. Values beyond ±60 mark the critical zone, a critical reading at which the cycle model raises the alarm for a possible turn. Today the scan shows eleven markets in the critical bottoming zone and seven in the critical topping zone. Let’s take a closer look.

Bottoming cycles

Read more

Friday delivered gains across the board as investors read the weak July jobs report through a “bad is good” lens. The Nasdaq gained 1.30%, the Russell 2000 added 1.10%, and the S&P 500 rose 0.62%. Treasuries rallied alongside, with yields falling across the curve, and the market trimmed its Fed expectations to roughly a quarter point of rate increases for the rest of the year. The dollar sank 0.4%. The strongest moves came from the metals, where gold rose 2.5% and silver jumped 3.4%, and our cycle scan has a lot to say about exactly that corner of the market. Brent slipped 0.6% to around $82 after a Reuters report that the Iran-Oman deal to reopen Hormuz should be released soon.

The jobs report itself was weaker than the rally suggests. Establishment adds missed badly with negative revisions, household employment declined, and the drop in the unemployment rate came from a slump in participation, which carries hawkish implications. The odds of a September hike still stand near 44%, and Friday revived the Fed-independence fight with the threat to remove governor Lisa Cook back on the tape. Against that backdrop, the story of the day in our scan is a single sequence. The hard-asset complex has moved from cycle warning to a turn that price now backs, with silver and the miners leading higher, and the long end of the bond market is lining up as the next candidate, this time for a peak in yields. Wednesday’s CPI print is the first test of that second leg.

Our daily analysis filters roughly 45 markets through a cycle consensus engine. Each asset receives a Consensus Score from -100 to +100, where positive values mark potential cyclical bottoms and negative values potential cyclical tops. Readings beyond the ±60 threshold define the critical zone, a critical reading at which the cycle model raises the alarm for a possible turn. Let’s take a closer look.

Bottoming cycles

Read more

Elliott waves and time cycles both claim that markets repeat, and they claim it about different things. That distinction carried this episode of the FSC interview series, recorded with Elliott Prechter, President of Qualitative Analytics and developer of EWAVES, the software that automates Elliott wave analysis. The lineage behind him is three generations deep: R.N. Elliott catalogued the pattern in the 1930s and published “The Wave Principle” in 1938, A.J. Frost and Robert Prechter codified it in their 1978 book, and Robert’s son Elliott has spent the past decade turning it into executable code. His own path runs through MIT (2002), Microsoft (2006), a stint as CTO of an algorithmic hedge fund, and a guest piece in The Elliott Wave Theorist that recommended Bitcoin in 2010 at 6 cents.

Form, not period

A cycle repeats in time. It has a length, the length drifts, and the analyst’s job is to measure it and project the next trough. A wave repeats in form. Elliott’s observation was that markets advance in five waves and correct in three, and that this five-three structure is self-similar at every degree, from the intraday wiggle to the multi-decade advance. From an analytical perspective the two schools operate in different domains: cycle analysis lives in the frequency domain and asks when, wave analysis lives in the pattern domain and asks where in the structure you stand. Both are recurrence claims about the same tape, made along different dimensions, which is precisely why the comparison is worth an hour of conversation rather than a dismissal from either side.

Teaching the machine to count

The standard objection to wave analysis has always been the analyst: ask five Ellioticians for a count and you receive five counts. Prechter’s answer was not to train a neural network on past charts but to build symbolic AI, encoding the wave model’s own rules and guidelines into software. EWAVES generates candidate counts and grades every wave with an elliotticity score, an absolute measure from 0 to 100 percent of how well the structure conforms to the model, and its pattern recognition is scale-invariant by construction, mirroring the fractal claim at the heart of the theory.

Two results from this work deserve attention beyond the wave community. First, Prechter and the mathematician Sacha Sardo-Infirri published research showing the model can distinguish real market returns from randomized versions of the same data across the Dow, the Nasdaq, gold and Bitcoin. That is a falsifiable test, and wave analysis spent decades being accused of not offering any. Second, his stability argument: a form-based model is stationary across all of market history, while parameter-fitted quantitative models morph with each regime and burn out when the regime changes. Whatever one thinks of wave counts, that is a serious methodological position.

Waves out of cycles

The bridge between the schools is older than most practitioners on either side know. J.M. Hurst made the connection in 1970, arguing that summed cycles of different lengths generate exactly the wave-like structures Elliott catalogued: a long rising component with a shorter cycle riding on it produces five waves up and three down without any pattern being imposed. My own research extends the point beyond Elliott counts. Overlay the right composite of cycles and almost any classical formation falls out, head-and-shoulders included. The patterns are not the cause of anything; they are what cycle summation looks like on a chart.

As a result I treat the two models as complementary projections of the same object rather than rivals. The cycle model explains where the form comes from and supplies the clock, since a dominant cycle can be measured and its next trough projected. The wave model reads the form directly in the here and now, without measuring a single period, and tells you where inside the structure the market stands. In practice the combination is stronger than either input: a completed five-wave decline arriving in the time window of a projected cycle trough is the kind of agreement that separates a setup from a statistic, the same conditions-first logic Larry Williams argued two episodes ago from an entirely different toolbox.

The full conversation is on the FSC channel:

Elliott Prechter: Teaching a Machine to Count Elliott Waves

EWAVES: https://ewaves.com

Elliott Wave International: https://www.elliottwave.com

Foundation for the Study of Cycles: https://cycles.org

US stocks came into Wednesday on the back of a broad rally. Yesterday the S&P 500 jumped 1.79%, the Dow climbed 1.71%, the Nasdaq surged 2.59%, and the Russell 2000 added 1.85%. Treasuries gained as yields fell across the curve, helped by Brent’s 5.25% slide to around $79.40 on reports that a deal to reopen the Strait of Hormuz could arrive within days. Gold climbed 0.55%, silver rose 2.4%, and Bitcoin advanced 0.8%.

The rally had two earnings drivers, Caterpillar’s blow-out quarter and Palantir’s huge beat-and-raise, layered on top of the Hormuz headlines and a market that has decided Big Tech’s Q2 reports answered the AI bears for now. The tone has turned into aggressive chasing as the S&P sprints toward fresh all-time highs, with the jobs report on Friday and the CPI next week as the nearest tests. For our cycle work, the day belongs to a single story reaching its payoff. Three weeks ago we wrote that seven cycles were building a floor beneath a falling gold price. Today nine cycles point to the same low, the deepest alignment this scan has ever shown, price has begun to move higher behind the signal, and the entire mining-and-uranium block reaches the model’s strongest bullish reading on the same day. That is the day’s thesis, and everything below serves it.

Each day we filter roughly 45 global markets (equity indices and sectors, metals, energy, agriculture, crypto, FX, and rates) through our cycle consensus engine. The key reading is the Consensus Score, which ranges from -100 to +100 and measures how strongly the dominant cycles of a market agree on a potential turning point. Values at or beyond ±60 mark the critical zone, the range where the cycle model raises the alarm for a possible turn and a market earns a place in the tables below. Let’s take a closer look.

Bottoming cycles

Read more

A price index can rise while most of its stocks fall. The McClellan Oscillator was built in 1969 to expose exactly that condition, and for this episode of the FSC interview series both generations behind it joined me: Sherman McClellan, who created the Oscillator and Summation Index with his late wife Marian, a mathematician, and their son Tom McClellan, who edits The McClellan Market Report and still works on it with Sherman daily. The conversation covered an origin story that is better than most people know, and the claim that gives this article its title.

Rocket science and hand-drawn charts

In the summer of 1969 every data point was computed by hand. Marian McClellan calculated the values, Sherman posted them to charts, and the mathematical machinery came from an unlikely source: P.N. “Pete” Haurlan, a genuine rocket scientist at the Jet Propulsion Laboratory who had used exponential smoothing for missile and satellite tracking and applied the same technique to market data. Haurlan’s terminology survives inside the Oscillator to this day. His “10% Trend” and “5% Trend” are exponential moving averages with smoothing constants of 0.10 and 0.05, which correspond to lengths of 19 and 39 days (the EMA multiplier is 2 divided by length plus 1). Gene Morgan, host of the “Charting The Market” television show in Los Angeles, put the new indicator on air in 1969 and attached the family’s name to it, and Haurlan’s Trade Levels firm published the McClellans’ book “Patterns for Profit” in 1970.

The construction itself is compact. Take the NYSE’s daily net advances (advancing issues minus declining issues), compute the 19-day and 39-day EMAs of that series, and subtract the slow from the fast: that difference is the McClellan Oscillator. The running total of the Oscillator is the Summation Index, the slower gauge for the larger swings.

What breadth sees that price hides

Cap-weighted indexes hide participation. A handful of mega-caps can carry an index higher while the average stock rolls over, and today’s concentration makes the effect stronger than at any point in the indicator’s history. Breadth counts every issue equally, and as a result it measures something closer to the availability of money than to the opinion of the index. When liquidity is abundant, the buying reaches even the weakest issues and breadth is broad. When liquidity drains, money concentrates in the largest names first, the advance-decline statistics deteriorate before the headline index does, and the divergence appears at major tops with remarkable consistency.

In terms of this interview series, the McClellans complete a measurement chain that Michael Howell opened in episode one: Howell measures liquidity at its source, the balance-sheet capacity of the financial system, while breadth measures its arrival at the level of the individual stock.

A band-pass filter before the term was fashionable

From an analytical perspective, subtracting a slow EMA from a fast EMA is a band-pass filter. The fast average removes the day-to-day noise, the subtracted slow average removes the trend, and what remains is the vibration between the two smoothing horizons, roughly the swings that play out over several weeks to a few months. The Oscillator was therefore a cycle instrument from birth, tuned by its fixed 19/39 pair. That fixed pair is the same design decision John Bollinger made with his fixed 20-period bands, and the same adaptation applies: measure the dominant cycle first, then set the EMA pair relative to it, so the passband stays centered as the market’s rhythm morphs. In practice, breadth data rewards this treatment more than price data does, because removing the mega-cap distortion leaves a cleaner carrier. Running dominant-cycle detection on the cumulative advance-decline line regularly produces sharper spectra than the cap-weighted index itself.

Repetition as a family business

The other thread through six decades of McClellan work is repetition. Tom’s research lives on lead-lag relationships and recurring patterns:

• commercial traders’ net positions in eurodollar futures as a roughly one-year leading indication for the S&P 500 (a relationship he introduced publicly in 2011, and one he is candid about),

• gold leading interest rates by about 20.5 months, and an 8-year cycle in gold itself.

Sherman has watched enough complete market cycles. Even the founding dataset was framed this way: before publishing in 1970, the McClellans assembled advance-decline data covering two full 4-year market cycles, because one cycle proves nothing.

The full conversation is on the FSC channel:

The McClellan Oscillator: Two Generations of Market Timing

McClellan Financial Publications: https://www.mcoscillator.com

Foundation for the Study of Cycles: https://cycles.org

Larry Williams calls himself a conditional trader, and the phrase carries more analytical weight than it first appears. Williams has been trading since 1962, created the Williams %R indicator in 1966, and won the 1987 World Cup Championship of Futures Trading by turning USD 10,000 into more than USD 1.1 million in 12 months of real-money trading, a result that still stands as the record. For this episode of the FSC interview series we talked about the part of his work that rarely makes the headline: the recurring structures underneath six decades of results.

Six decades of the same market

Williams’s continuity claim is simple. The instruments, the speed and the participants have morphed beyond recognition since 1962; the behavior has not. Fear and greed produce recurring structure, and as a result, patterns he traded in the 1960s still appear on today’s charts. For a cycle analyst this is the foundational assumption made flesh: recurrence is the raw material of the discipline, and Williams is a sixty-year sample of evidence that the material exists. The working question is never whether markets repeat but which repetitions carry information, and his answer is a hierarchy: conditions first, timing second.

The commercials know their market

The first condition is positioning. The CFTC’s weekly Commitments of Traders report splits open interest into commercials (the hedgers who produce or consume the underlying), large speculators, and small traders. Williams began reading this report in the early 1970s, decades before it became fashionable, on the logic that the commercials hold the best fundamental information about their own market.

His COT Index turns the raw report into a bounded oscillator: take the commercials’ current net position, subtract the lowest net position of the past 3 years, divide by the 3-year range, and multiply by 100. Readings above 75 mean the commercials are unusually long (a bullish condition), readings below 25 mean they are unusually short. The discipline sits in what the index is not: it is weekly data with a reporting lag, so it cannot time an entry. It tells you which direction deserves your timing tools, nothing more.

Seasonality as the calendar cycle

The second condition is the calendar. Williams built his first seasonal indexes in 1973, mapping the recurring pressure of harvests, inventory cycles, tax flows and demand patterns onto price. From an analytical perspective, a seasonal is a cycle with the length locked at 12 months and the phase locked to the calendar. The dominant cycles I measure are the free-running counterpart: length and phase estimated from the data, allowed to drift. The two are complementary rather than competing. Fixed calendar cycles supply the stable scaffold, measured cycles capture the vibration that drifts around it, and my own second book (Part II of “Decoding the Hidden Market Rhythm”, devoted to Metonic calendar cycles) argues the fixed-length case at lengths well beyond 12 months, so Williams’s seasonal conviction landed on prepared ground in this conversation.

In practice: the checklist

The framework operationalizes cleanly.

• Condition one: the commercials’ COT Index above 75 on the weekly report, which the CFTC publishes free every Friday.

• Condition two: a supportive seasonal window.

• Condition three: a projected dominant cycle trough due in the same zone.

When at least two of the three agree, the timing tools take over, for Williams typically %R turning up out of its oversold zone (below minus 80) or a volatility breakout off the open. A cycle projection standing alone is a statistic; the same projection inside agreeing conditions is a trade. That division of labor is the transferable lesson of his six decades, and it applies to every timing method in this publication, not only his.

The full conversation is on the FSC channel:

Larry Williams: Six Decades of Trading Market Cycles

Larry Williams: https://www.ireallytrade.com

Foundation for the Study of Cycles: https://cycles.org

Price is not particularly cyclical, but volatility is. That claim comes from John Bollinger, and for a publication devoted to market cycles it lands as a challenge worth taking seriously. Bollinger is the creator of Bollinger Bands, founder of Bollinger Capital Management and author of “Bollinger on Bollinger Bands”. For this episode of the FSC interview series we deliberately skipped the standard Bands tutorial and went to the intellectual roots of his work, which run straight through cycle analysis.

From Hurst’s envelopes to statistical bands

In the opening chapter of his book, Bollinger surveys the history of plotting bands around price, and J.M. Hurst sits prominently in that lineage. Hurst (”The Profit Magic of Stock Transaction Timing”, 1970) drew constant-width channels around price, positioned and nested according to his cyclic analysis. The approach was conceptually elegant and practically fragile: the width was fixed and the construction was manual.

Bollinger’s contribution in the early 1980s was to let the market set the width. The middle band is a 20-period simple moving average, the outer bands sit 2 standard deviations above and below, and because the standard deviation is recalculated on every bar, the envelope morphs with the market’s vibration. Width expands when volatility rises and contracts when it falls, with no hand-drawing required. As a result, Bollinger Bands are a child of the cycle-analysis tradition with a statistical spine, and Bollinger chose statistics over cycle-anchored envelopes for a pragmatic reason: cycles in price were too unstable to carry a reliable trading tool, while volatility could be measured directly.

The cycle lives in volatility

That pragmatic choice contains the deeper thesis of the conversation. The recurring structure in markets, in Bollinger’s view, lives in the second moment of the return distribution, not the first. Periods of low volatility are followed by periods of high volatility, and high by low. The Squeeze is the measurable expression of this: BandWidth (upper band minus lower band, divided by the middle band) falling to a six-month low (roughly 125 trading days on a daily chart) marks the point of maximum contraction, and expansion follows. The signal says nothing about direction; it is a statement about energy, not sign.

From an analytical perspective this is a genuine cycle claim, only relocated. Econometrics files the same phenomenon under volatility clustering (the ARCH and GARCH family); Bollinger read it off the chart with a tool any trader can plot. For cycle analysts the implication is uncomfortable and useful in equal measure: phase measured in volatility can be steadier than phase measured in price, and a volatility trough is a projectable event even when the price cycle is drifting.

Fixed 20 periods or adaptive to the dominant cycle

Bollinger’s defaults were never doctrine, and his own published rules say so. The defaults (20 periods, 2 standard deviations) are labeled as exactly that, defaults, and his containment guidance moves the multiplier with the length: 2.0 standard deviations at 20 periods, 2.1 at 50 periods, 1.9 at 10 periods. The parameters flex, which invites the obvious next step.

In the conversation I put the cycle analyst’s alternative on the table: measure the dominant cycle first, then set the band length to half of it. A market vibrating on a 40-bar dominant cycle gets 20-period bands, a 60-bar cycle gets 30. In practice this is the same adaptation I apply to the cyclic RSI, where the lookback follows half the dominant cycle instead of Wilder’s fixed 14. The fixed 20 has survived four decades because it sits near half of the roughly 40-bar rhythm that many daily charts oscillate around; when the measured dominant cycle drifts away from that length, the fixed setting loses containment and the adaptive length earns its keep.

Bands, patterns and timing

The bands also act as a reference frame that makes classic formations jump off the chart: W-bottoms and M-tops become readable shapes instead of vague impressions. Cycle analysis adds the clock. A W-bottom completing where a projected cycle trough is due carries more weight than either signal alone, and that combination is exactly what an analyst needs on the right side of the chart, which is the only side that pays.

The full conversation is on the FSC channel:

Beyond the Bands: John Bollinger on Volatility Cycles

John Bollinger: https://www.bollingerbands.com

Foundation for the Study of Cycles: https://cycles.org

Asset prices follow liquidity, not earnings. Michael Howell has built a three-decade research career on that single claim, first at Salomon Brothers, then at CrossBorder Capital, and in his book “Capital Wars: The Rise of Global Liquidity”. For the opening episode of the FSC interview series I spoke with him about how he measures global liquidity, why it moves in a cycle, and what that cycle implies for the here and now.

A refinancing system, not a capital-raising system

Howell’s starting point is an uncomfortable piece of arithmetic. The world economy carries roughly USD 350 trillion of debt with an average maturity of about five years. As a result, financial markets must roll over approximately USD 70 trillion of debt every year just to stand still. In Howell’s words, debt is never repaid, it is only rolled over.

The financial system has therefore morphed from a capital-raising mechanism into a debt-refinancing mechanism, with refinancing transactions outweighing new capital raising by roughly seven to one. In this world the binding constraint is not the price of money (interest rates) but the availability of money: the balance-sheet capacity of banks, shadow banks and cross-border investors to absorb the rollover. That capacity is what Howell defines as global liquidity. It is not M1 or M2, and it behaves differently, because around 77 percent of global lending is collateral-based (a World Bank figure). Collateral values drive lending capacity, lending capacity drives asset values, and asset values feed back into collateral. The multiplier that matters today is a collateral multiplier, not the textbook deposit multiplier.

The 65-month cycle

Because the refinancing schedule is set by past issuance, liquidity does not drift randomly. It moves in a repeatable cycle that Howell measures at roughly 65 months (about 5.4 years) across the CrossBorder Capital indexes. From an analytical perspective this is exactly the kind of vibration a cycle analyst looks for: a dominant cycle in a driver series rather than in price itself. Price is the shadow, liquidity is the object casting it. In practice the phase of the liquidity cycle carries more information than its absolute level, and it can be projected forward the same way any dominant cycle can, with the usual caveat that length and amplitude drift over time.

What it means for asset classes

The asset implications follow directly. Rising liquidity expands balance-sheet capacity beyond what the rollover absorbs, and the surplus flows into risk assets, equities first. Falling liquidity forces the system to fund the rollover out of a shrinking pool, and financial assets get sold to do it. Gold and Bitcoin sit in a separate category: Howell treats both as hedges against monetary inflation, the debasement that a permanently growing debt pile makes unavoidable, which is why he reads them as liquidity assets rather than CPI assets. Bonds carry the burden itself; a market asked to absorb USD 70 trillion of rollover per year plus net new issuance is structurally heavy.

Where the cycle stands

Howell’s indexes put the latest liquidity upswing, the one that began in late 2022, at or just past its peak around the turn of 2025/2026. Ahead of it sits a heavy refinancing calendar: a large share of the debt issued in the Covid years falls due between 2026 and 2030. The tension he describes is between a Federal Reserve that is not accommodating the rollover and a US Treasury that partially substitutes for it through heavy short-dated bill issuance, a form of fiscal quasi-QE. However that tension resolves, the refinancing requirement itself is not negotiable, and that is precisely what makes this cycle projectable rather than merely observable.

The full conversation is on the FSC channel:

Michael Howell on Global Liquidity: How Money Flows Move Asset Prices

Michael Howell / CrossBorder Capital: https://www.crossbordercapital.com

Foundation for the Study of Cycles: https://cycles.org

The euro turns up right on schedule while rising yields push back on the model’s rates call

Amazon’s blow-out AWS quarter did the heavy lifting on Friday. The S&P 500, the Dow and the Nasdaq finished higher on the back of it, and the report lifted the other hyperscalers with it, even though yields spiked further and several earnings reactions elsewhere disappointed. The strength was narrower than the headline indices suggested, since the equal-weight S&P and the Russell 2000 both lagged. Apple was hit hard after its Thursday report, with pressure in services revenue and September guidance below plan.

The AI debate sharpened. Amazon raised its capex budget for the year to $220 billion from $200 billion and made the growth case for AI compute, while the bears point at the same report’s weak free cash flow as the clearest illustration of the industry’s cash question. The three hawkish Fed dissenters argued their case on Friday morning, a reminder that Wednesday’s decision to hold was contested, and the coming week brings the ISMs, a heavy earnings slate and the July jobs report. For our cycle work, the Fed window we described on Wednesday has resolved, and it resolved in two directions at once. That is the story of the day. The euro completed its projected turn at the model’s strongest possible bottoming reading while the dollar reached the strongest possible topping reading on the same session, and the rise in yields pushed back against the rates side of the call exactly along the invalidation line we had named.

Each day we filter roughly 45 global markets (equity indices and sectors, metals, energy, agriculture, crypto, FX, and rates) through our cycle consensus engine. The key reading is the Consensus Score, which ranges from -100 to +100 and measures how strongly the dominant cycles of a market agree on a potential turning point. Values at or beyond ±60 mark the critical zone, the range where the cycle model raises the alarm for a possible turn and a market earns a place in the tables below. Let’s take a closer look.

Bottoming cycles

Read more

Tuesday extended the rotation. The S&P 500 gained 0.22% while the Nasdaq slipped 0.22%, and the strength sat everywhere else, with the Dow climbing 1.03%, the equal-weight S&P jumping 1.14% and the Russell 2000 adding 0.19%. Brent slumped 5.2% lower to around $83.75 on further signs of diplomatic progress in the Middle East, gold fell 1.2%, silver dropped 1.95%, and Bitcoin sank 1.85% to about $63,700.

The selling in chips and AI infrastructure continued through the morning before easing into the afternoon, and the money leaving that corner rotates into software and the non-tech sectors, which is exactly what the equal-weight strength shows. Earnings season remains strong against a high bar, and the week’s heaviest reports are still ahead, with Meta, Microsoft and Qualcomm after Wednesday’s close and Apple and Amazon on Thursday. Above all of it sits the Fed decision on Wednesday afternoon, with the market giving roughly a one-in-three chance to a hike.

For our cycle work, the timing is the story. Five rates-and-currency instruments, the dollar index, both Treasury yields, the long bond and the euro, have moved into the model’s deepest late-stage readings simultaneously on the very day the Fed decides, and yields already slipped on Tuesday in the direction the model has been projecting.

Each day we filter roughly 45 global markets (equity indices and sectors, metals, energy, agriculture, crypto, FX, and rates) through our cycle consensus engine. The key reading is the Consensus Score, which ranges from -100 to +100 and measures how strongly the dominant cycles of a market agree on a potential turning point. Values at or beyond ±60 mark the critical zone, the range where the cycle model raises the alarm for a possible turn and a market earns a place in the tables below. Let’s take a closer look.

Read more

The S&P 500 finished about flat, while the Nasdaq and the Russell 2000 slipped a little. The other side of the tape told a different story, because the Dow climbed 0.45%. So the average stock had a good day while the big technology names weighed on the cap-weighted indices. Silver rose 1.1% and Bitcoin fell 1.5%.

The driver behind the tech weakness sits away from geopolitics. Investors are reassessing how sustainable the data-center build-out is when capital spending outruns the cash those businesses generate.

For our cycle work, Friday is the day a two-week projection completed itself. The story of the day is a triple arrival, three markets reach the model’s strongest possible reading for an upward turn at the same time, each with a broken downswing behind it, and price has started to move their way.

Each day we filter roughly 45 global markets (equity indices and sectors, metals, energy, agriculture, crypto, FX, and rates) through our cycle consensus engine. The key reading is the Consensus Score, which ranges from -100 to +100 and measures how strongly the dominant cycles of a market agree on a potential turning point. Values at or beyond ±60 mark the critical zone, a critical reading at which the cycle model raises the alarm for a possible turn, and a market at that level earns a place in the tables below.

Let’s take a closer look.

Read more

US stocks drifted through Wednesday without direction. The S&P 500 slipped 0.14%, the equal-weight index and the Dow finished about unchanged, the Nasdaq fell 0.57%, and the Russell 2000 lost 0.92%. The stronger moves happened away from equities. Brent jumped 3.3% higher to around $94 as the situation in the Middle East deteriorated further, gold climbed 1.3%, silver rose 1.4%, and Treasury yields finished higher across the curve while the market shifted its Fed expectations in a hawkish direction. Bitcoin slipped 0.75%.

Earnings delivered strong order numbers from the AI build-out. For our cycle work, the story of the day sits in the currency pair and the hard assets behind it. The model hands its strongest topping signal (the market is in a time window where the rise could stall or reverse) to the US dollar on the very day the euro future enters the bottoming table (a time window where the decline could end), both readings driven by the same 306-day cycle, while the entire metals complex behind gold steps toward a turn in unison and gold’s price has begun to move higher behind the call.

Each day we filter roughly 45 global markets (equity indices and sectors, metals, energy, agriculture, crypto, FX, and rates) through our cycle consensus engine. The key reading is the Consensus Score, which ranges from -100 to +100 and measures how strongly the dominant cycles of a market agree on a potential turning point. Values at or beyond ±60 mark the critical zone, the range where the cycle model raises the alarm for a possible turn and a market earns a place in the tables below. Let’s take a closer look.

Read more

I have spent more than two decades measuring cycles in financial markets, and the same objection still lands in my inbox every week: cycle analysis is witchcraft and doesnt work. I understand where that comes from. Most of what circulates under the label of cycle analysis deserves the criticism. However, the criticism almost always targets one of nine misconceptions rather than the method itself.

The core statement comes first because everything else follows from it: cycles never predict prices. A cycle projection points to a turning window on the time axis. It tells you when the market is likely to turn. It stays silent about where price will stand at that moment and about how far the move carries afterwards.

What cycles actually are

1. Cycles predict prices
Everything downstream breaks when this part is wrong. A dominant cycle of 120 bars tells you that a trough window opens roughly 120 bars after the last confirmed trough. That is a statement about time, and nothing in it concerns the price axis. Even if you plot a sine-wave, you can not read the max. or min. turns of the plotted sine to read a price target from.

2. Price is the cycle
What you see on the chart is trend, cycles and noise in superposition, and the cyclic part is just one out of the three. I have watched analysts read raw price as if it were a clean wave. They find cycles that are nothing but the trend bending, and they miss real ones buried under noise. In practice this means detrending comes first, always. Once the trend is removed, the vibration of the market becomes measurable. Same otherway around: You can not plot a cycle on top of the raw price series to predict or forcast. You always need to remodel the trend and noise components if you want to get a somewhat related price curve-prediction.

3. One cycle explains the market
There is never just one. Several cycles of different lengths are active at any moment, and their combined wave shapes the swings you trade. As a result, a single-cycle overlay fits one stretch of history and falls apart in the next, because the additional dominant cycles moved the turning points. The dominant cycle operates alongside the others.

The detection trap

4. A peak in the spectrum proves a cycle
Run a spectrum analysis on pure random noise and you will get peaks, some of them impressive. A peak is a candidate. The spectrum tells you where to look, and the validation stage decides what you actually found. Trading the tallest bar in a scanner output means trading noise with conviction. Noise can generate peaks in a spectrum. So you never know if a peak in a spectrum is a valid, real cycle or just due to mathematical noise in the dataseries.

5. Detection is validation
Genuine cycles are separated from noise by testing, and two features do this work. The Bartels score measures the statistical significance of a candidate (the probability that the wave is a real periodic component rather than a chance pattern).
The Stability Score checks whether the cycle held its phase and length across the analysis window. A candidate that fails these tests remains noise, however tall its peak looked in the spectrum. My workflow has read the same way for years: detrend, detect, validate, then project. Skip the validate step and the projection is worthless.

6. More data means better detection
This one feels rigorous and fails in practice. Cycles morph. A detection window that reaches back too far averages the current cycle with its ancestors. The result is a length that fits the whole history while describing the average historical fit. At at the right side of the chart, the average fit over a long history is useless for prediction. The relevant question is which window captures the vibration in terms of phase and length that is active now, and the amount of available data says little about that. Even worse, you more data you throw at the analysis the more you average and loose precision for the current point in time where everything matters.

Cycles rhyme rather than repeat

7. Cycles are fixed clocks
The most common message I get after a turning window reads: “Lars, the cycle projection did not work.” In many of these cases the projection worked and the expectation was wrong, because the reader measured the market against a perfectly symmetric sine wave. Cycles in financial markets carry a skew. The downward phase runs shorter than the upswing. This asymmetry has been documented in the business-cycle literature for decades. From Keynes’s remark that downturns arrive suddenly and violently while recoveries build slowly, to the formal statistical work on asymmetric expansions and contractions. Hold a simplified sine next to a skewed real-world cycle and the projection looks broken; account for the skew and the same projection lines up. Skew is one example among several: length, amplitude and phase all morph as the market digests new conditions. Therefore cycle analysis is the continuous re-measurement of a living rhythm. The fixed-date folklore (anniversary dates, rigid multi-year counts applied without re-detection) fails in live markets for exactly this reason, while adaptive detection keeps working. Have you every applied a skew to you analysis? If not, start doing.

8. A cycle that worked will keep working
Cycles fade in and fade out. A dominant cycle can lose its dominance within a few swings when a longer or shorter vibration takes over, and a validation score from last year protects nothing today. As a result, the full pipeline runs on every update, and I have retired more than one cycle I was fond of.

The analyst owns the forecast

9. The tool makes the forecast
This is the misconception I push back on hardest, because it is the comfortable one. A fool with a tool still stays a fool. The scanner delivers validated turning windows and their statistical quality. The analyst weighs those windows against trend, liquidity and risk, and the analyst owns the forecast decision. A cycle projection is one measured input into a human judgement. Treat it as an oracle and you have converted a measuring instrument into an alibi. ;-)

The takeaway

Strip these nine away and cycle analysis becomes a sober craft. Decompose price into trend, cycles and noise. Detect candidates in the detrended series. Validate with the Bartels score and the Stability Score. Project turning windows on the time axis, and re-measure continuously, because cycles rhyme rather than repeat. The prediction at the end of that chain belongs to the analyst, and the tools for the earlier steps have never been better.

Always remember, a fool with a tool still stays a fool. I have been fooled myself on this subject many times. So I will keep writing to help others to overcome these issues.

Yours,
Lars

Companion article to the Market Cycles Report video of July 20, 2026. Watch the full episode here:

Twelve weeks of alerts. One ledger. Think of the model as a wind report for 45 harbors: it tells you where the wind is turning, not how to sail the boat. When several independent cycles converge on the same turning point, the Cycle Consensus Score crosses ±60, the critical zone: the cyclical tailwind under a market, or the headwind against it, has become strong enough to matter. Setting the sails, and deciding when to come home, is the sailor’s job. The model does not know what the Fed will say next, and it holds no opinion in the AI debate. It reads cycles.

Three months of daily articles flag the same market many times while a signal builds, so this review counts opportunities the way a reader experiences them: as episodes. An episode is one continuous stay on the alert list, one asset in one direction, from the day it first crosses into the critical zone until it drops off. Natural gas sitting on the bottoming list from April 24 through May 20 is one episode; its return on the topping side in late June is a second one. Each episode is then measured between its first alert and July 17: did price move in the signaled direction at all, how far did it run at its best close, and how quickly? No adjustments after the fact. What follows is the full record.

Summary

Start with what the model is built for. The Cycle Consensus Score is an alerting layer: it flags the markets where the cyclical conditions for a turn are present, early, across 45 instruments. It is not a complete trading system. It issues no exits and does not manage positions. Measured on its own terms, the quarter reads like this: 62 of the 72 episodes with post-alert data, 86%, moved into profit in the signaled direction at some point after the alert. Half reached at least 3% at their best close, 36% reached 5%, and nearly one in five ran into double digits. Only 10 episodes, 14%, never closed positive for a single session. The average episode offered a peak move of 5.3% (median 3.1%), and among the winners the peak arrived 23 days after the alert on average.

The top of the ledger shows what the lead time is worth. Natural gas offered 32.5% between the April 24 bottoming alert (article) and its June 25 peak. The April topping alert on crude (article) offered 27.4% to the July 1 low, and the June 2 re-entry offered 26.9% inside four weeks. The April silver topping alert (article) printed its best close on July 17, the final day of this review, 27% below the alert price, twelve weeks after the model first flagged the top. The May gold-miner longs (article) offered 12% within two weeks before the sector rolled over, which is the other face of the same number: the score opens the opportunity window and tells you when to be at the table. What you carry out of it depends on the exit, and the exit is yours.

Eighteen daily articles cover the window from April 24 to July 17, 2026: 316 critical-zone signal rows, grouped into 77 distinct episodes, 72 of them with at least one session of post-alert data. The five newest entries from July 17 have no tape to score yet and roll into the next review. Everything else is in the scorecard, the chart, and the full episode table below.

The opportunity ledger

The chart shows the ten largest peak moves of the window, measured from the alert close to the best close in the signaled direction; the day count behind each bar is how long that move took to arrive. All ten crossed 10%. They split into two speeds: fast pops that peaked inside three weeks (the gold miners, Nvidia, Solana) and slow burns that needed two months and more (natural gas, crude, silver). Whether any of it survived to July 17 varied: the miners gave all of it back, gas and crude kept most.

Two markets in that chart deserve a closer look, natural gas first. It is on the list twice, in opposite directions: the April 24 bottoming alert (article) ran 32.5% to its June 25 peak, and the June 26 topping alert (article) caught the other side for 11.5% by mid-July. One market, one quarter, both turns flagged in advance. The two charts show each entry as it stood on the alert day: analysis cutoff at the alert, the light price section is what followed.

The long entry: natural gas, daily. The dominant-cycle composite (62- and 215-day cycles) bottoms at the April 24 analysis cutoff; the light price section is what followed. (Original alert)

The short entry: the same market two months later. The composite (67-, 77- and 131-day cycles) tops at the June 26 analysis cutoff; the light price section is what followed. (Original alert)

Crude is on the chart twice as well: the April topping alert (article) needed ten weeks to reach its 27.4% peak, while the June 2 re-entry, a single quiet appearance at -60, delivered 26.9% inside four weeks, from $93.76 at the alert to $69.23 at the low.

How fast the peak arrives

Among the 62 episodes that moved into profit, the best close arrived 23 days after the alert on average. The distribution is the useful part: 23 peaked within ten days, 20 between day 11 and day 28, and 19 needed longer than four weeks. In practice, roughly a third of the opportunities are quick pops that demand a fast hand, a third mature inside the classic two-to-four-week turn window, and a third are slow burns that reward patience, silver’s twelve-week run being the extreme case.

What conviction added

The score’s extremes added less on this yardstick than expected. Episodes that reached the full ±100 delivered a 3% opportunity slightly more often than the rest, 56% against 48%, with an average peak of 6.1% against 5.0%. But three of the four largest opportunities of the quarter came from scores below the maximum, and the June 2 crude alert, the single best four-week move, entered the list with one appearance at exactly -60 and a neutral cRSI. On the opportunity ledger, breadth beats loudness: being at the table matters more than how hard the table was pounded.

The ten that never turned

Honesty requires the other end of the table. Ten episodes never closed positive for a single session: the June gold-complex bottoming calls (gold itself, both gold-miner ETFs, and the silver miners, all flagged June 17 (article) into a falling market), four topping calls that leaned against running trends (the semiconductor index, the Dow, the dollar, the 2-year yield), the June BRICS entry, and Ethereum. Most of them sit in the three trends that dominated the quarter: a US large-cap rally that would not break, rising yields, and falling precious metals. A high score is a necessary condition for most cyclical turns, not a sufficient one, and these ten are what that sentence looks like in practice. Several of the underlying signals are still live, gold’s seven-cycle floor at +93 (article) first among them; they belong to the next review.

The full episode table

Every episode, sorted by what it offered. Peak move is the best close in the signaled direction between the alert and July 17, a price peak for the long side and a price bottom for the short side; days to peak/bottom shows how long the opportunity took to develop; the last column links to the daily article that issued the alert.

_{The daily alert articles behind the table:
Apr 24 · Apr 28 · May 1 · May 7 · May 9 · May 13 · May 15 · May 20 · Jun 8 · Jun 17 · Jun 26 · Jul 11 · Jul 17}

What this record says

The Cycle Consensus Score identifies a market condition, the point where enough independent cycles align that the cyclical tailwind under a market, or the headwind against it, becomes statistically meaningful. It is not a trading signal. None of the numbers above are returns anyone booked; they are the size of the window each alert opened. Turning a window into a result takes individual timing, position sizing, and an entry and exit plan, and those belong to the trader, not to the model.

What the record does show is an edge in knowing where to look. When the score crossed into the critical zone this quarter, six times out of seven the market moved in the signaled direction at some point afterward, half the time by 3% or more, in a quarter where three major trends refused to turn. That is the claim, the whole claim: when an asset enters the critical zone, it is time to take a closer look. This ledger is what that claim looks like when it is tested.

If you made it this far, thank you. A review like this one only has value if the ledger is complete, and this ledger is: every critical-zone alert from twelve weeks of daily scans, measured as opportunity against the tape, including the ten alerts that never went green. The daily newsletter is the front end of that ledger: each morning, a fresh scan across 45 global markets, and an alert the moment the Cycle Consensus Score crosses ±60. The alert is where the work starts. The exit is where it pays.

The gold episode, the yield complex, and five fresh July signals are already running. If you want to watch them resolve in real time instead of reading the verdict in October, subscribe below.

© 2026 Lars von Thienen, All Rights Reserved.

Information contained on this site is obtained from sources believed to be reliable, but its accuracy cannot be guaranteed. Lars von Thienen (”lars.cycles.org”) is a publisher of scientific cycle analysis results for global markets and is not a registered investment advisor. The published analysis is not designed to meet your personal circumstances. We are not financial advisors and do not give personalized financial advice. The opinions and statements contained herein are the sole opinion of the author and are subject to change without notice. They may become outdated and there is no obligation to update any such information.

Any projections, market outlooks, or estimates herein are forward-looking statements and are inherently unreliable. They are based upon certain assumptions and should not be construed to be indicative of the events that will occur. Past performance is not indicative of future results.

Investments should be made only after consulting with your financial advisor and only after reviewing the prospectus or financial statements of the company or companies in question. You should not make any decision based solely on what you read here.

Neither the publisher, the author, nor any of its affiliates accepts any liability whatsoever for any direct or consequential loss howsoever arising, directly or indirectly, from any use of the information contained herein.

Gold builds its floor on seven cycles while the long yields show their first signs of tiring

Yesterday the S&P 500 fell 0.51% to 7,533 and the Nasdaq dropped 1.47%, but the damage was narrower than the headline suggests. The equal-weight S&P gained about 1%, carried by regional banks, insurance, real estate, healthcare, retail, and energy, as money rotated out of the AI complex and into value and cyclical groups. The driver was the deteriorating AI narrative, from missing free cash flow at the hyperscalers to fresh worries about data-center overcapacity, with TSMC’s solid-but-not-enough report the latest disappointment. Iran stayed in the headlines but was not the force behind the tape. Brent slipped 0.70% to about $84.35, Treasury yields edged slightly higher, and the metals had a rough session, gold down 2% and silver off 3.65%.

The cycle model reads the day differently than the tape traded it. The strongest reading on the board is gold, where the cycle score climbs to +93, a high level of agreement on a possible turn up, now resting on seven converging cycles (several market rhythms that agree and pull in the same direction), the deepest stack in the table. At the same time the 10-year and 20-year yields both reach -93 with bull fatigue (the bulls are losing interest, the upward move is getting tired), the first momentum signals on the yield side in this sequence. The bond ETF TLT mirrors that from below with a +87 floor reading. Yesterday’s price action ran against both signals, which is exactly what makes them worth watching.

The cycle model analyzes roughly 45 global markets across equities, rates, commodities, and currencies. The Consensus Score runs from -100 to +100 and works like a school grade for how strongly the cycles agree that a turn is approaching, positive for a possible turn up and negative for a possible turn down. A reading at or beyond 60 marks the critical zone (a reading at which the cycle model raises the alarm for a possible turn). Let’s take a closer look.

Read more

US stocks came into Friday on the back of a quiet session. The S&P 500 added 0.42%, the equal-weight S&P gained 0.37%, the Dow Jones Industrial climbed 0.29%, and the Nasdaq rose 0.29%, while the Russell 2000 fell 0.49% against the broader tape. Stocks dipped briefly after renewed statements that the Middle East ceasefire was over, then recovered just as quickly, since negotiations continue and a diplomatic outcome remains the most plausible path. Treasuries saw mild pressure, with yields finishing flat to slightly higher. Brent eased 0.3% lower to around $76, gold and silver each lost roughly half a percent, and Bitcoin advanced 1.1% to about $63,900.

The real story of the day sits in the cycle model’s configuration ahead of next week. The S&P 500 energy sector enters the bottoming table at the model’s bullish maximum of +100 with bear exit confirmation, the bears throw in the towel and the downswing appears broken, just as Brent sits near $76 and the Strait of Hormuz supply question remains unresolved. On the opposite side, the Dow Jones Industrial deepens to a bull exit at -89, the first confirmed momentum reversal among the large US benchmarks while price still grinds higher. The rates complex lines up on both dashboards at once, bond prices bottoming and yields topping, ahead of Tuesday’s CPI. Delta posted the only earnings report of the day and beat expectations with favorable guidance, yet the stock failed to rally, a sign of how much is already priced in ahead of the reporting season that begins next week.

Each day we filter roughly 45 global markets through our cycle consensus engine. The key reading is the Consensus Score, which ranges from -100 to +100 and measures how strongly the dominant cycles of a market agree on a potential turning point. Values at or beyond ±60 mark the critical zone, a reading at which the cycle model raises the alarm for a possible turn. Let’s take a closer look.

Read more

Palladium hits the model’s bullish maximum with a confirmed turn as the metals floor upgrades while small caps deliver the first topping confirmation

US markets went into the long Independence Day weekend on a split tape. In Thursday’s session, the last before the holiday close, the S&P 500 ended about flat while the Nasdaq fell 0.80% and the Russell 2000 lost 0.55%. The equal-weight S&P gained 0.70% and the Dow advanced 1.14%, so the selling was concentrated in the crowded chip and hardware complex while insurance, software, healthcare, staples, utilities, and materials carried the broad market. The June jobs report came in soft at +57K payrolls, yet two-year yields dipped only slightly, a muted reaction that speaks to lingering inflation concerns. The dollar index slipped 0.50% and Brent ended about flat near $71.60.

The story of the day is confirmation stacking up on the metals floor while the equity topping side adds its first momentum signal. Palladium reaches the model’s bullish maximum at +100 with bear exit, the bears throw in the towel and the downswing appears broken, and silver upgrades to the same confirmed state at +79. Thursday’s metals rally puts price behind both signals. On the other side, the Russell 2000 moves to bull fatigue at -81, the first US equity index in this sequence to pair its topping score with a momentum reading.

The cycle model analyzes roughly 45 global markets across equities, rates, commodities, and currencies. The Consensus Score runs from -100 to +100 and works like a school grade for how strongly the cycles agree that a turn is approaching, positive for a possible turn up and negative for a possible turn down. A reading at or beyond 60 marks the critical zone, where the model raises the alarm for a possible turn. Let’s take a closer look.

Read more

Thirteen markets sit on the bottoming side, the most in weeks, and they are led by commodities, precious metals, and Asian equities while US tech keeps running higher without confirmation

US equity futures were red on Friday morning, with the S&P down 0.53% and the Nasdaq off 1.16% as the selling in chips and AI names picked up again. The trigger was a report that OpenAI may push its stock-market debut into 2027, which would slow the spending that has carried the whole sector. Asia took it harder overnight, with Korea down 5.81% and several chip names off 6 to 11 percent. Oil fell about 3.7% to roughly $72.50 after Saudi Arabia resumed loadings at a key Gulf terminal, Treasuries rallied, and gold edged up 0.30%. The mood on the surface is risk-off, but the cycle scan underneath points somewhere the headlines are not looking.

The story today is a broad floor forming away from the crowded tech trade. Thirteen markets sit on the bottoming side, the most in weeks, and they are led by commodities, precious metals, and Asian and emerging-market equities. Corn Futures reach the model’s maximum at +100 with bear exit (the bears throw in the towel and the downswing has broken). Behind corn stands a deep cluster of gold, platinum, palladium, and silver, with Hang Seng, the BRICS ETF, and India’s Nifty 50 broadening it into the neglected corners of the market. The topping side, by contrast, is all US equities and yields, eleven of them, but every single one without momentum confirmation.

The cycle model analyzes roughly 45 global markets across equities, rates, commodities, and currencies. The Consensus Score runs from -100 to +100 and works like a school grade for how strongly the cycles agree that a turn is coming, plus for a possible turn up, minus for a turn down. A reading at or beyond 60 marks the critical zone (a reading at which the cycle model raises the alarm for a possible turn). Let’s take a closer look.

Read more