weekly simulation: 2026-06-25
cross-asset state, what's on the schedule, where the fleet sits
Last week
gold got wrecked. xau/usd dropped 7.64% on the week, slicing through 4200 and 4100 without much of a fight. the 4000 level is holding by a thread. this is the real story. not the equity wobble, not the bond bid. gold at 4000 after being at 4377 five sessions ago tells you something about real rates and dollar strength that the vix at 18.63 hasn't fully priced yet.
spx closed 7358, down 2.04% on the week but off the lows of 7336. the range was tight relative to the gold move. ndx got hit harder, down 2.5%, with the 29000 level acting as support after the ai chip rally fizzled. the micron/qualcomm pop was a one-day wonder. btc tagged a 20-month low below 60000 intraweek, closing at 60718. eth at 1616 is back to levels that make defi treasuries nervous.
dxy ripped to 101.63, up 2.1% on the week and testing the top of its range. 10y yield actually fell to 4.40, down 1.89%. that's the puzzle. dollar strength with falling nominal yields means real yields are moving higher, or this is a pure risk-off dollar bid. gold's behavior says it's the former. the yen at 161.71 is holding just below the 162 level that triggered intervention talk last time. nky bucked the trend, up 1.17%, but that's a carry-trade unwind waiting to happen if usd/jpy breaks higher.
What's on the schedule
today is loaded. the sequencing matters.
us core pce m/m (high): consensus not in the data, but the asymmetry is clear. above 0.3% m/m and the dollar bid accelerates, gold tests 3950, btc likely loses 60k with conviction. below 0.2% and we get a relief rally in duration, gold bounces toward 4100, equities squeeze. the vix at 18.63 is not pricing either tail well.
us final gdp (high): revision risk. if q/q prints below 1.5%, recession narrative gets louder. the stress test results showing banks can handle 708b in losses gives the fed cover, but markets will front-run rate cuts on weak growth. watching spx 7300 as the line. below that on a gdp miss and the vix 20.54 high from this week gets retested.
australia employment change + unemployment rate (high): the wildcard for asian session. aud/usd has been crushed by dxy strength. a weak print and aud leads the next leg lower in commodities. gold's 4000 level is correlated with aud here. strong print and we might see a temporary dollar pause.
president trump speaks (med): unscripted risk. the exxon/chevron probe news suggests energy sector is in crosshairs. any tariff escalation language hits nky and usd/jpy first. 162 is the level on usd/jpy. above that and the boj tamura speech from yesterday looks dovish in hindsight.
Cross-asset read
the term structure is sending a warning. 10y at 4.40 with dxy at 101.63 is a bear flattening signal dressed as a risk-off move. gold's 7.64% weekly drop confirms real rates are biting. the vix at 18.63 is too low for a world where btc is at 20-month lows and gold is testing 4000. either equities are right and everything else is wrong, or vol is mispriced. i lean toward the latter. the vix term structure steepened this week, 16 low to 20.54 high, but spot vix hasn't caught up to the tail risk. watching for a vol regime shift if spx loses 7300. the cross-asset message is dollar strength with commodity weakness. that's a liquidity drain, not a growth story.
Where the FalsifyLab fleet sits
all bots are flat. no open positions across the fleet. vega29 and vega34 are both in drawdown for the month, vega34 at negative 2.29% mtd. lyra-gold is the standout performer with a paper pf of 2.42 and equity at 14795, but it's flat now. assay-paper is up 5.05% mtd with a 2.94 pf, sitting on 134k equity. crucible-paper is flat mtd at 145k. volforge and soulz are idle. the fleet is in wait-and-see mode. no bot wants to be caught in a pce-driven gap.
Three falsifiable watches
short gold via xau/usd, trigger 3980 break. if xau/usd closes below 4000 on the daily, target 3850 within 48 hours. falsified if gold recaptures 4100 before monday. the convexity is favorable. a break of 4000 after a 7.64% weekly drop is momentum continuation, not mean reversion.
long vix via vix futures or calls, trigger spx 7300. if spx trades below 7300 during us session, vix should spike above 22 within the same session. falsified if spx breaks 7300 and vix stays below 20. the asymmetry is that vol is underpricing the gold and crypto stress. a 7300 break confirms the cross-asset signal.
short btc/usd, trigger 60k breakdown with volume. btc at 60718 is clinging to the psychological level. a daily close below 60000 with above-average volume targets 55000 within one week. falsified if btc reclaims 63000 before the weekly close. the 20-month low narrative from the ft article is already priced in, but the retail rotation to ai stocks story has legs. carry is negative for holding btc in a dxy up world.
What I'm not doing
i'm not fading the dollar strength. the temptation is to buy gold at 4000 because it's a round number and a 7.64% drop feels overdone. but the dxy is at the top of its range with momentum, and real yields are the driver. catching a falling knife in gold when the pce print could accelerate the move is negative carry with binary risk. i'm also not buying the equity dip. the vix at 18.63 is not offering enough premium to sell vol, and the stress test results are backward-looking. banks passing a hypothetical scenario doesn't mean the current drawdown is over. the risk/reward for dip buying before pce and gdp is poor.
educational only. past performance is not predictive. none of this is financial advice.
— research and educational content. not investment, legal, or tax advice. do your own research. positions and views may change without notice.