Fundamentals set the level, technicals set the timing

This article was submitted by Michael Stark, financial content lead at Exness.

Markets go into this week with one number left to resolve a two-week argument. August’s payrolls came in at 162,000 against a consensus near 55,000, which killed the ‘too weak to hike’ case, but average hourly earnings at 3.1% annually gave the Fed no wage-inflation case either. Odds of a hike in September are sitting somewhere around 45%, and Wednesday’s inflation data decides the rest. In this week’s episode I’m discussing with [guest name, role at organisation] how fundamental and technical analysis actually fit together for traders — not as rival camps, but as answers to two different questions. This article covers some of the ground we went over, but for the full discussion please watch or listen to the podcast if you haven’t yet done so.

link to/embed podcast

Two questions, not two camps

The framing of ‘fundamental versus technical’ has always seemed to me like a category error. Fundamentals tell you roughly where an asset ought to trade and what would have to change for that to be wrong. Technicals tell you where other participants are currently positioned and what it costs to be early. Those are different questions, and answering one of them well doesn’t answer the other at all.

Put concretely: fundamental work on the Fed’s reaction function might convince you that a hike this month is underpriced. It won’t tell you what price to buy the dollar at, how much room to give the trade, or what invalidates it. Conversely, a clean chart pattern will tell you where the risk is defined but nothing at all about whether the event calendar is about to make that level irrelevant.

August’s payrolls: the right read, the wrong trade

Last Friday is a decent illustration. Payrolls beat every estimate in the Bloomberg survey, with May and June revised up by a combined 55,000 and unemployment holding at 4.1% even as participation rose to 61.6%. On a purely fundamental reading, that’s a hawkish print and it should have been good for the dollar and awkward for equities.

Equities did give back part of the previous session’s gain, but yields finished the week roughly where they started, and gold only eased. Anyone who traded the headline aggressively got the economics right and the outcome close to flat, because the market had already spent two weeks pricing a hawkish Fed after Jackson Hole. The information was new; the positioning wasn’t.

I’d put it this way as a rough rule of thumb: the reaction to data is a function of the surprise minus what’s already in the price. Fundamentals give you the surprise. Only the chart and positioning data give you the second term.

Charts as a record of what’s already priced

XAUUSD compressing between $4,450 and $4,550 ahead of Wednesday’s inflation data. Source: Exness MT5

Gold is a fairly clean current example. The price has been holding above the $4,450 shelf since the Jackson Hole flush but has failed twice at $4,550, and the daily range has been contracting into the 20-day moving average. That structure isn’t telling you what inflation will be. What it is telling you is that positioning has been flushed out and that the market genuinely doesn’t know — which is usually the setup for a sharper move once the number lands, in either direction.

The Nasdaq looks like the mirror image. USTEC has now failed at the 26,600–26,800 band on every attempt since the AI de-rating in July, including last Wednesday’s bounce on Waller’s comments. Repeated failures at the same level are a positioning fact, not a forecast, but I do think they shift the balance of probabilities: it takes progressively better news to clear a level that’s already rejected buyers three times.

Where the two disagree, and what to do about it

NVDA gapped up nearly $15 on its late-August earnings and hasn’t filled it. Source: Exness MT5

Nvidia’s late-August report is the case I find most interesting, because the two frameworks pointed in slightly different directions. Fundamentally the report was excellent — revenue roughly doubled year on year to around $96 billion, earnings per share beat at $2.22 — yet the reaction was muted relative to the numbers, and 12 of the top 20 institutional holders have either trimmed or added less than 1% this year. Technically, though, the gap from 25–27 August held, and the moving averages stayed constructive.

My own resolution in cases like that is to let the fundamentals decide whether I want exposure at all and let the chart decide the entry, the size and the invalidation. On Nvidia that meant staying long-biased while the price stayed out of the gap area, with a test of the highs around $235 as the reasonable objective rather than a target derived from any valuation model. A price-to-earnings ratio near 30 doesn’t generate an entry level, and a gap doesn’t tell you whether demand for accelerators is accelerating.

Into Wednesday

Consensus for August points to +0.2% monthly on the headline and +0.3% on core. A hot print pushes hike odds back above 60%, and I’d expect that to cap gold at range highs and send equities back toward the August lows. A soft print validates the hold, and the more interesting question then becomes whether the Nasdaq can finally clear that resistance band on the fourth attempt.

Either way, the fundamental question — will they hike — and the trading question — where do I get in and what proves me wrong — have different answers, and I think a lot of avoidable losses come from using one to answer the other. There’s a good deal more in the episode on how each of us actually sequences the two in practice, so don’t miss this week’s Trading Talks.

This is my personal opinion, not the opinion of Exness. This is not a recommendation to trade.