Weekly Market Recap: Gold, Forex and Crypto — 5-11 Oct 2026

Gold rose while silver slipped, and Bitcoin, Ethereum, and Solana all ended the seven-day crypto window lower. Yet neither the dollar nor Treasury yields supplied a simple explanation: US yields eased across the curve, DXY was almost unchanged, and selected precious-metals fund flows moved against the direction of spot prices.

The defining feature of 5–11 October was divergence. Central-bank communications added context rather than a fresh policy decision, US activity data offered mixed signals, and crypto recovered somewhat over the weekend without reversing its completed-period losses.

For continuity, see HFR's previous weekly market recap for 28 September–4 October 2026.

Week at a glance: gold was the exception

Period	Instrument	First price	Last price	Key context
5–9 Oct	DXY	102.17	102.21	+0.04%
5–9 Oct	EUR/USD	1.1223	1.1202	−0.19%
5–9 Oct	GBP/USD	1.3223	1.3235	+0.09%
5–9 Oct	USD/JPY	157.92	158.34	+0.27%
5–9 Oct	Gold (spot, $/oz)	4,139.89	4,194.46	+1.32%
5–9 Oct	Silver (spot, $/oz)	61.1000	60.8302	−0.44%
5–11 Oct	Bitcoin ($)	85,771	83,691	−2.43%
5–11 Oct	Ethereum ($)	2,710.42	2,534.63	−6.49%
5–11 Oct	Solana ($)	120.76	111.25	−7.88%

These are first-to-last eligible provider-dated price comparisons, calculated as (last ÷ first − 1) × 100. Forex and metals cover 5–9 October; crypto covers 5–11 October, including Saturday and Sunday. They are not conventional previous-Friday-to-Friday returns, official exchange settlements or live 12 October quotations.

Market context: different assets, different messages

There was no single cross-asset direction to explain this week. Gold gained 1.32% even as estimates suggested redemptions from two prominent US gold funds. Silver declined 0.44% despite a positive estimate for the largest silver fund examined. Bitcoin, Ethereum and Solana fell by 2.43%, 6.49% and 7.88%, respectively, although their daily closing-price series showed a partial recovery after Friday.

The policy backdrop needs equally careful dating. The Federal Reserve's 16 September decision had already established a 3.75–4.00% target range before the week began. The minutes published on 7 October described that earlier meeting; they were not a second interest-rate decision. Likewise, the European Central Bank's October account revisited its September meeting, and the Bank of England's 8 October speech did not change Bank Rate. In Japan, the BoJ's official decision register records no new policy-rate decision during 5–9 October.

Fed, US data, Treasury yields and the dollar

Services held up, but price pressures remained prominent

The September ISM Services PMI, published on 6 October, eased to 54.9 from 55.4 in August. Readings above 50 indicate an expanding survey sector, so the decline points to slower expansion rather than contraction. The Prices Index increased to 74.0 from 72.6, while New Orders eased to 59.8 from 60.9. That combination points to slower services growth alongside firmer reported price pressures.

The August US international trade release, published on 6 October, showed the nominal goods-and-services deficit widening to $105.6 billion from a revised $92.8 billion in July. Exports were $315.2 billion, up 1.4%, and imports $420.8 billion, up 4.3%. These nominal totals should not be read as a direct calculation of trade's real-GDP contribution.

The 8 October weekly unemployment-claims report recorded 197,000 initial claims for the week ended 3 October, down 2,000 from a revised 199,000. Separately, August wholesale trade showed sales of $817.5 billion, up 1.8% month on month, and inventories of $964.2 billion, up 0.5%. Neither report was a September payrolls release. The relevant September jobs report was published before the coverage window, while September CPI and PPI were scheduled for the following week.

Governor Christopher Waller's 8 October remarks on the economic outlook and the Summary of Economic Projections added one policymaker's perspective, not collective guidance from the Federal Open Market Committee. No exact historical FedWatch percentage is quoted because a time-matched archived snapshot was not independently verified.

Treasury yields declined unevenly

Against that mixed data backdrop, the US Treasury's daily par-yield series shows the 2-year yield falling 4 basis points to 4.80%, the 5-year falling 4 basis points to 5.02%, the 10-year falling 7 basis points to 5.24%, and the 30-year falling 6 basis points to 5.60%, from 5 to 9 October. The 10-year minus 2-year spread narrowed from 47 to 44 basis points: a modest three-basis-point flattening, rather than a parallel curve move.

Even as yields declined, the US Dollar Index historical series barely moved, rising 0.04%, from 102.17 to 102.21 on the same endpoint convention. Falling nominal yields alongside a marginally firmer DXY are observed together; the data do not establish that one caused the other.

Forex market reaction: the dollar was not uniformly stronger

The nearly flat dollar index masked different outcomes for individual pairs. The euro eased: EUR/USD moved from 1.1223 to 1.1202, or −0.19%. The ECB's October account is useful context for the policy debate, but it did not announce a new rate cut or increase. Separately, Eurostat reported that August euro-area retail volume rose 0.1% month on month, following a 0.6% fall in July.

Sterling did not follow the euro lower. GBP/USD rose 0.09%, from 1.3223 to 1.3235, despite the small gain in DXY. The Bank of England had set Bank Rate at 3.75% in September; Governor Bailey's 8 October speech addressed resilience and shocks rather than a fresh Monetary Policy Committee decision.

The yen also weakened against the dollar, with USD/JPY rising 0.27%, from 157.92 to 158.34. No new Bank of Japan policy decision occurred during the completed traditional-market week. These three currency pairs underline why a near-flat broad-dollar index cannot stand in for each bilateral exchange rate.

Cryptocurrency market reaction: losses despite a weekend recovery

CoinGecko's Bitcoin history shows Bitcoin moving from $85,771 on 5 October to $83,691 on 11 October, a 2.43% decline. The provider-dated series fell as low as an $81,686 closing observation on 8 October before recovering on the next three dates.

Ethereum declined 6.49%, from $2,710.42 to $2,534.63; its 8 October daily reference was $2,472.73. Solana was weaker still, declining 7.88%, from $120.76 to $111.25. Its 9 October reference was $109.07 before the weekend improvement.

All three percentages use the same CoinGecko provider-dated closing-price convention. They are not Coinbase exchange-native candles, and this news recap does not use intraday technical levels to interpret those moves.

Two developments were relevant to the wider crypto ecosystem but require precise labels. On 8 October, the Thai SEC issued crypto-ETF notifications with an effective date of 16 October, initially covering bitcoin and ether; that was not proof of a fund launch on the announcement date. The same day, Securitize announced on-chain trading of US stocks with security entitlements on Solana. That is an issuer-announced product development, not a verified measure of network usage or SOL purchases.

Crypto ETF and product flows: redemptions across all three groups

Fund flows provide a separate perspective on those crypto losses. The Farside Bitcoin ETF series recorded $678.9 million of net outflows across the five US fund sessions from 5 to 9 October. The Ethereum spot ETF series showed $542.2 million of net outflows, while the Solana-linked product series showed $25.0 million of net outflows.

Within that series, Bitcoin's largest daily outflow was $484.9 million on 7 October, followed by $244.1 million on 8 October; 9 October then recorded a modest $21.1 million inflow. Ethereum and the Farside-tracked Solana-linked group were negative in each of the five sessions. These are net flows through the specified US-listed product universes, not a record of worldwide crypto investment or a direct measurement of spot-token selling.

Solana-linked vehicles are not all legally identical to the spot Bitcoin and Ethereum ETF categories. Product histories can also include seed-related conversions that should not be mistaken for new customer subscriptions. ETF flows can provide useful context, but timing and product mechanics prevent a simple claim that redemptions caused the week's token-price declines.

Weekend crypto recovery: gains did not erase weekly losses

Crypto continued trading on 10–11 October after traditional US fund sessions had finished. From Friday's CoinGecko observation to Sunday's, bitcoin recovered 1.37% ($82,564 to $83,691). Ethereum gained 1.97% ($2,485.70 to $2,534.63), and Solana rose 2.00% ($109.07 to $111.25).

These are Friday-to-Sunday dated-price comparisons, not Saturday-opening-to-Sunday-closing returns. The rebounds reduced the size of the weekly losses but did not reverse them. The price data show a weekend recovery, but they do not by themselves identify a single market-wide catalyst.

Gold, silver and ETF demand: price and fund flows diverged

The divergence extended to precious metals. Spot gold's XAU/USD series advanced 1.32%, from $4,139.89 on 5 October to $4,194.46 on 9 October. Spot silver's XAG/USD series moved from $61.1000 to $60.8302, a 0.44% decline. These are Investing.com historical spot prices, not official COMEX futures settlements; metals are sensitive to real-rate expectations, currency changes and their own distinct demand conditions, but those potential influences are not independently measured as causes here.

The estimates for individual metals funds tell a different story from their spot prices. ETFIQ's GLD estimate showed approximately $331 million of five-session net outflows; its IAU estimate showed approximately $42.3 million of net inflows. Together, the selected GLD and IAU estimates amount to around $288.7 million of net outflows. For silver, SLV showed an estimated $33.0 million of net inflows despite the spot-price decline.

There is no contradiction in a metal's price moving differently from creations or redemptions in a selected exchange-traded product. ETFIQ calculates daily net flows from changes in issuer-reported shares outstanding and the previous day's net asset value. Those figures are not issuer-confirmed cash subscriptions, global ETF flows, or direct measurements of bullion-holdings changes. The gold figures here cover GLD and IAU only, while the silver figure covers SLV only. The World Gold Council publishes global monthly data, so it should not be treated as a weekly 5–9 October flow total.

What to watch next: 12–16 October

The week ahead starts with a holiday-adjusted session. Monday 12 October is a US federal holiday, and SIFMA recommends a full close for US dollar-denominated fixed-income trading. It is also Japan's Sports Day: regular JPX cash trading is closed, while Osaka Exchange and TOCOM derivatives operate holiday trading. This does not mean every equity, currency, metal or crypto market is closed.

The major US events are September CPI and real earnings on Wednesday 14 October, followed by the Federal Reserve Beige Book; September PPI and September retail sales on Thursday 15 October, alongside the scheduled August business-inventories report; and import/export prices plus industrial production and capacity utilisation on Friday 16 October, according to the BLS, Census and Fed calendars.

For sterling, the ONS monthly GDP, Index of Production, Index of Services and UK Trade releases for August are each scheduled for Thursday 15 October at 07:00 UK BST. The Thai SEC crypto-ETF regulatory notifications are due to take effect on Friday 16 October; effectiveness is not the same as an actual ETF launch. These are forthcoming events, not results from the completed week.

Weekly takeaway

Gold's gain, silver's decline, easing Treasury yields, an almost flat dollar index and broad crypto losses do not add up to a single market narrative. They show why policy communications must be dated precisely and why spot-price performance should be assessed separately from selected ETF-flow estimates. The next week's releases may change that balance, but the 5–11 October record is one of divergence, with a modest weekend crypto recovery that left the full-period losses intact.

Risk note

Market and fund-flow observations are provided for education and information, not as personalised investment advice or a recommendation to buy, sell or hold an asset. Spot prices, currencies and cryptocurrencies can move rapidly; fund flows, liquidity changes and leverage may amplify risk, and historical comparisons do not establish future returns.

Sources and methodology

This recap was rechecked on 12 October 2026. Macro and policy claims use official central-bank and government releases where available. Traditional-market price comparisons use consistent Investing.com historical observations for 5 and 9 October, while the crypto comparison uses CoinGecko provider-dated observations for 5 and 11 October. Treasury changes use the US Treasury daily par-yield curve. Farside figures cover the specified US-listed crypto-product universes, and ETFIQ figures are selected-fund flow calculations rather than global precious-metals flows.

The return calculation is (last eligible provider-dated price ÷ first eligible provider-dated price − 1) × 100. Traditional-market observations stop on 9 October and crypto observations stop on 11 October; publication on 12 October does not add a further completed session. Farside and ETFIQ datasets can be revised after initial publication, so their live tables remain the final reference for later checks.