nvidia reports the person on the call is colette kress, who is now cfo and interim ceo at the same time ! Get it, girl ! Women on charge can be one the best things of the century.
jensen stepped down last week. $NVDA dropped about 13% in five days and options are pricing a 9.2% move on the print
street wants $108.8b revenue and $2.47 eps, roughly 90% growth year over year. market cap went from $5.77t on oct 6 to about $4.07t
49
micron's taiwan union voted 99% to authorize a strike over bonuses and honestly i get it
sk hynix puts 10% of operating profit into bonuses. samsung puts in 10.5%. micron reportedly sits at 2.2% with a $1t+ market cap
they already turned down an offer of up to 68 months of pay. the union wants a profit formula. it's only fair.
taiwan hosts a big share of micron's dram and hbm output so $MU holders have a labor line item now
1
89
SPREADSHEET EXPOSÉ
nvidia reports tuesday after the close with jensen huang out as ceo and colette kress running it as interim. most sheets have the same - i mean exactly the same - consensus number, about $109b in revenue, up roughly 91%. the cell that decides whether that number lands in the right place is the period label above it
nvidia calls the quarter fiscal q3 2027. plenty of data feeds and plenty of hand-typed headers call that same stretch q3 2026 or q4 2026. if your model pulls consensus with a vlookup on a text label like “Q3 26”, it can grab the neighboring quarter and never throw an error. a revenue number from the wrong quarter looks exactly like a revenue number from the right one
key every column on its period-end date instead of a label. type the quarter end as a real date, then pull with =XLOOKUP(date, dates, values, “no match”) so a mismatch shows up as words instead of a neighbor’s figure. nvidia’s fiscal year ends in january, so its q3 closes in late october. put that date in the row and let the sheet do the matching
went through my own earnings tracker this morning. nvidia was one column off. i built that lookup in august
1
32
the interest row deserves more screen space in your $CRWV model. CoreWeave reported $3.66b revenue yesterday. quarterly operating income was $207m. net interest expense was $846m.
put those last two numbers next to each other. operating income covered about 24% of net interest expense. revenue growth has a fairly expensive passenger.
build a separate debt schedule with maturity dates and fixed or floating rates. link the interest forecast to it. keep non-cash financing charges separate so the cash forecast stays readable.
then stress the refinancing assumptions. TA DA ! works like magic i promise!
38
SPREADSHEET EXPOSÉ
check the units before yesterday’s CPI gets pasted into your forecast. october prices rose 0.4% month over month, above the 0.3% consensus. that little “monthly” label is doing A LOT of work.
repeat 0.4% for 12 months and the compounded pace is about 4.9%. that’s a hypothetical run rate. label it before it wanders into a chart titled “annual inflation.”
give monthly changes and annual changes - separate columns, please. don't be a slop. for the annualized monthly pace, use =(1+monthly_rate)^12-1, with 0.4% entered as a percentage.
a cell formatted to two decimal places can still contain a very confident misunderstanding.
25
wall street paid 3% for gridlock today.
the dow rose 1,569 points, its absolutely best day since april 2025, after democrats took both chambers, and oil fell hard.
trump says he’ll veto everything, so gridlock is now the base case. we just wait.
74
Just worth a read after the rollercoaster we've been through over the past few weeks.
he gives us a very precise idea of the moment and what's to come in short term.
It's lovely to see a green day like this, especially considering how many people expected the market to dump if Democrats won the midterms.
That said, I think some people are getting a little too bullish over a 3–5% pump. Don't get me wrong, it's a welcome move, but it's still relatively small in the grand scheme of things.
I'm heavily allocated in the market myself, but I'm not entirely convinced that we're out of the woods just yet. I'd still keep some dry powder on the sidelines to take advantage of any potential dips. A lot can happen over the coming weeks.
I'm bullish, don't get me wrong, but the start of Q4 could still bring some nasty volatility. Stay optimistic, but don't get complacent.
Good night friends. Tired.
26
friday’s jobs report will move stocks more than today’s midterms. I bet!
september printed just 29,000 jobs, and cook and hammack already voted for a hike at the october meeting. one weak number and december gets repriced fast
52
microsoft’s $175b 2026 capex number is an accounting change. data centers moved to operating leases after the useful life went to 25 years from 15, and the old forecast was $190b with spending unchanged. read the cash flow statement before the margins. $MSFT
47
aws is still the biggest cloud at about $184b run rate and the slowest compounder on this chart, 29% cagr vs 44% for azure and 65% for google cloud. $AMZN $MSFT $GOOGL holders, do you pay for the run rate or the growth rate?
$440B in combined annualized revenue across AWS, Azure and Google Cloud.
Combined 2026 capex guidance for Amazon, Microsoft, Alphabet and Meta: about $735B, up 79% from $410B last year.
A $295B gap. Someone has to be right about the demand.
39
strategy’s average cost is $75.4k per btc as of the august filing. six figures would put the whole stack about a third in the green. $MSTR holders, what’s your plan if btc stalls under six figures? bitcoin:native
BREAKING: 🇺🇸 Michael Saylor's Strategy has bought $63 Million worth of Bitcoin.
He says Bitcoin is definitely moving back into six figures.
19
microsoft’s capex doubled to $39.4b and free cash flow covered 51% of net income vs 92% a year ago.
stock was up 4% after hours anyway because azure grew 48%. i’d look at the free cash flow line before the azure line
13
let's not challenge 2027. it can always get worse. LOL
Everything that could go wrong has happened in 2026
- US and Iran war
- High inflation
- Brutal crypto bear market
- Quantum FUD
- Exchange hacks
- DeFi exploits
- Clarity Act failure
- Saylor selling
- Cold wallets hack
- Fed rate hikes
- Stock market dump
Now, things can only get better from here.
If you have made it so far, don't give up now.
12
WILD day we're living
$HYPE tvl was $8.2b on defillama as of oct 25, down from $9.03b on sep 26. polymarket has a live market on whether october revenue clears $60m. that revenue line on defillama is the number i’m watching into month end
1
56
i can't wrap my mind around what's going on on the mkts today. It's kinda giving me a headache even
Wow.
$GOOGL down 12%.
It lost the most ever in market cap, hundreds of billions of dollars, in a day.
Market Tide is lopsided bearish all morning: Net call premium: -$298M and falling Net put premium: +$124M and rising
Traders are dumping calls and buying puts.
17




























