All Four Hinge on 2027

I’ve just updated SECSift, my EDGAR reading tool, so that all free accounts now get access to all premium features for NVIDIA (NVDA), Alibaba (BABA), and Celsius (CELH).

Premium features include exporting filings as PDF, creating watchlist and keyword alerts, merging and charting financial tables, and more!

There’s also an API and upcoming Excel add-on, which means you can import exact tables from any SEC EDGAR filing!

🌾 Welcome to StableBread's Newsletter!

It’s been another packed earnings week (and a rough week for the market overall)!

This post breaks down my quarterly analysis for four names I’m still following from the write-up archive, in the order they filed: Toast (TOST) and Fuel Tech (FTEK), Newbury Street II (NTWO), the SPAC taking FORT Robotics public, and Root Insurance (ROOT).

P.S. If you missed it, check out Tuesday’s macro update. I talked about how to read the macro beyond the headlines and which positions I’ve added to.

Toast (NYSE: TOST), the restaurant software and payments company, reported its most profitable quarter yet on Tuesday, in the same 8-K that announced President and CFO Elena Gomez is stepping down.

I bought Toast as a long-term hold when I wrote it up on June 2 at $24/share. It closed Friday at $34.67, up 44.5% since then, and 3.4% above its $33.53 close before the report.

The Quarter

Here’s what stood out in the Q3 report:

Toast also raised its full-year adjusted EBITDA guide for the third time in 2026, to $840-850M, and now expects recurring gross profit to grow 26% (up from 23-25%):

Safe to say the core business is running ahead of my thesis.

Two Executives Leaving

Gomez, who took Toast public in 2021, gave notice on October 28, and the board named SVP of Finance Michael Senno to succeed her on January 1, 2027. She stays on as a non-executive advisor through July 2, 2027.

Senno joined Toast in 2022 after running investor relations at Coupang, the South Korean e-commerce company, and earlier held treasury and investor relations roles at Expedia and Time Warner.

Chief Revenue Officer Jonathan Vassil, who leads Toast’s sales team, gave notice on September 28 that he’s stepping down on the same December 31 date, and no successor has been named.

Both 8-Ks say the exits weren’t due to any disagreement with the company, and both executives stay on as employees into 2027, so I read them as planned rather than forced.

But Toast heads into 2027 with a first-time CFO and no named head of sales, in the same year Gomez expects memory costs to hit harder (per the Q1 call) and with the FTC suit (discussed below) still pending.

Senno will own the 2027 numbers, and he already set the frame on the Q3 call. He said he’s “confident in the 20%+ framework” for recurring gross profit growth (Q4 is guided to 22-24%), but expects “very modest margin progression next year” after a ~36% adjusted EBITDA margin in 2026, since Toast wants room to keep investing in its newer markets.

Those newer markets (enterprise, international, and retail) are nearing $200M of ARR and doubling for a second straight year, per Narang, and Senno said the core business already earns margins over 40%.

FTC Lawsuit

The FTC and eight states sued Toast on October 29, the day after Gomez gave notice, in federal court in Northern California (California’s case is brought by Los Angeles County and San Francisco).

The complaint alleges Toast:

The 10-Q says Toast will “vigorously defend its interests” and can’t yet estimate a possible loss.

CEO Aman Narang argued on the Q3 call that ROSCA (the Restore Online Shoppers’ Confidence Act, a federal law on online subscriptions) was never meant for negotiated business contracts:

He also said Toast had been cooperating with the FTC “over the past year” before it sued, and that “how we settle the suit” won’t change Toast’s long-term path.

With $1.89B of cash and marketable securities and no debt, Toast can afford a fine, so the bigger risk to me is an injunction that could possibly change how it sells, prices, and cancels contracts.

Hardware

Toast’s hardware and professional services gross loss widened to $85M in Q3 from $57M a year ago, and the $225M lost through September already tops the $220M lost in all of 2025:

It doesn’t help that hardware revenue was flat at $45M while hardware costs rose 29% to $130M. Management has tied that pressure to rising memory chip prices, which are largely outside Toast’s control.

What Toast can control is when and how it buys, and that’s exactly what it’s been doing. It’s been buying memory chips early, which took inventory from $114M (as of December 31, 2025) to $244M (as of September 30), with another $191M of hardware purchases committed over the next 12 months.

Gomez said on the Q3 call that “nothing has really changed” since the Q2 call, when she said Toast had already cut the expected memory cost for both 2026 and 2027 by using earlier hardware generations, cheaper memory, and spot-market buying.

Toast also started buying Toast Capital loans from its bank partner and holding them itself in 2026, buying $145M through September after none in the first nine months of 2025.

Gomez said that lets Toast keep the full return instead of just a fee for arranging the loans. But Toast now also takes the loss when a restaurant doesn’t repay, just as the FTC challenges its practice of making the full loan due when a customer cancels.

So far, Gomez says defaults are “within our expectations,” and most new loans will still go through Toast’s existing program rather than its own balance sheet.

Valuation

At $34.67/share, Toast’s enterprise value is $18.3B ($20.1B market cap on 581M shares, less $1.89B of cash and marketable securities, with no debt), or 21.6x 2026 adjusted EBITDA guidance.

In my August 10 update, I said I’d add if a pullback took Toast down to the high teens on forward EV/EBITDA. On the 2026 guide, that’s ~$29/share, which the stock briefly touched at the start of October.

Reading Senno’s “very modest margin progression” as a margin near 2026’s ~36%, 20% recurring gross profit growth would mean ~$1.0B of 2027 adjusted EBITDA.

Fuel Tech (Nasdaq: FTEK), the power plant pollution-control company, reported its first quarter under CEO Ramesh Nuggihalli on Tuesday, and used the release to make acquisitions part of the plan.

I bought Fuel Tech at $1.39/share in my August 27 deep dive, as a bet on its pipeline converting into orders and on Nuggihalli, a former COO of competitor CECO Environmental who took over that month.

The stock closed Friday at $1.35, 2.9% below my entry, with $27.8M of cash and investments and no debt:

The cash is what made Fuel Tech worth buying in August, since even if the pipeline never converted into orders, I’d still own $0.89/share of cash and a FUEL CHEM business that turns a profit.

The Quarter

Here’s what Fuel Tech’s Q3 report showed:

On margins, Nuggihalli said on the Q3 call that recent APC wins came back after customers compared competing bids, which to him shows “there is ability to have premium pricing,” and that “overall, the margin is going to be stable or growing.”

So the base business is doing its part, at least on revenue: APC brought in $8.1M through Q3 and needs just $1.9-2.9M in Q4 to hit the $10-11M I estimated for 2026 in August.

Backlog

APC backlog reached $16.6M as of September 30, up from $14.3M as of June 30 and $7.0M as of December 31, and the $3.3M of APC contracts announced with the results take it to $19.9M:

In my deep dive, I said a first data center order or signed capacity reservation could land by the end of Q3.

But none of 2026’s announced awards is a data center order, and neither the release, the 10-Q, nor the Q3 call mentions the reservation agreement management said in August it was negotiating with a potential data center customer.

Nuggihalli did call the data center opportunity “significant” on the call, and said the 2027 budget factors in “some of the mega project bidding” that’s still underway.

I don’t think management misled anyone, since it only called a Q3 award “a possibility.”

But data center awards Fuel Tech hinted at have now failed to land in two straight quarters, so don’t be surprised if the data center option doesn’t come by Q4.

The New Plan

Nuggihalli spent most of the Q3 call on strategy rather than the quarter. He’s running a planning process with outside advisors he’s worked with before, plus a smaller team already identifying acquisition targets, and plans to take the strategy and an “M&A playbook” to the board in early 2027.

He listed three gaps acquisitions would fill, in this order:

He didn’t say how much of the $27.8M the board would spend, but at an H.C. Wainwright conference in September, he said he doesn’t need “$50 million-$100 million deals,” just “a $2 million EBITDA company” that fills a product or channel gap.

Nuggihalli’s also reviewing R&D spending on DGI, Fuel Tech’s technology for dissolving oxygen into water. Adoption in aquaculture (fish farming) “has been slower than anticipated,” so Fuel Tech is evaluating larger parts of the water market.

In August, a first commercial hatchery system was the DGI milestone I said to watch.

But DGI barely matters to the thesis: it’s still in pilots, Nuggihalli called Fuel Tech’s investment in it “modest,” the only 2026 revenue management has mentioned is $100K from a wastewater rental, and I gave it no value in August.

Valuation

At $1.35/share, Fuel Tech trades 13.4% above its $1.19/share book value, and the market values the operating business at $14.3M ($42.1M market cap less $27.8M of cash and investments), ~0.5x TTM sales.

Newbury Street II (Nasdaq: NTWO), the SPAC taking robot-safety maker FORT Robotics public, lost 82% of its trust to redemptions at its October 28 extension vote, before the merger filing (S-4) is even public.

In my September 16 deep dive, I found the valuation hard to justify at $10.93/share, and set my entry for FROB (the ticker after the merger) below ~$7/share, but only with low redemptions.

I also called this vote’s redemption rate the first read on how much of the trust would reach FORT.

Here’s what the recent November filings (8-K and 10-Q) show:

On today’s commitments, the trust and the PIPE add up to ~$74M before fees, against the $200.8M of trust and PIPE money the August 2026 investor deck assumed with no redemptions.

But PIPE investors who already own NTWO shares can pay part of their commitment by keeping those shares in the trust instead of redeeming them.

As of September 30, holders of ~1.41M shares (~$15M) had said they may do that. If they do, that ~$15M sits in the trust and also counts toward the $41.25M PIPE, so adding the trust and the PIPE (~$74M) counts it twice:

So the realistic range is $41-59M before fees, with the $41.25M PIPE as the floor, plus up to ~$17.8M if the other ~1.64M public shares stay in the trust through the merger vote.

The worst case in my deep dive, 95% of public shares redeemed, left FORT $39.73M before fees, and today’s floor only clears it because Tiger Global added $10M to the PIPE.

PIPE investors also get up to 1.29M founder shares from the sponsor and up to ~0.5M shares from FORT for free, so the $41.25M buys up to ~5.9M shares at ~$7/share, against ~$10.85 for the holders who just redeemed.

The 10-Q also shows Newbury Street II hired Donerail, a firm affiliated with former director William Wyatt, as a deal advisor on September 29, the day Wyatt resigned from the board. Donerail gets $350K in cash plus an interest in 599,166 of the sponsor’s founder shares, so the fee barely touches FORT’s cash.

The stock closed Friday at $11.00, 1.4% above the ~$10.85 that redeeming holders received.

Root (Nasdaq: ROOT), the car insurer that prices drivers on how they actually drive, closed Friday at $43.37/share, down 24.1% in two days from $57.15 before its Q3 report, even though it earned $14.4M against a $5.4M loss a year ago.

I first wrote up Root on May 14 at ~$60/share as a hold, with a plan to size up on a pullback to $44-50, which the stock is now below.

In my August 10 update, I said Root was riding out the soft market (when large carriers cut prices and spend heavily to win customers) by protecting margins instead of buying growth, and Q3 showed exactly that trade-off.

Shrinking on Purpose

Premiums in force (the annualized premium on every active policy) fell 5.4% y/y, nearly double Q2’s 2.8% drop.

Growth in policies offset the lower premium per policy through Q1, but policies were flat from June at 484,037 (+3.8% y/y) while premium per policy fell 8.9% to $1,441:

Basically, this means that Root’s customers are paying less on average, and Root’s customer count has barely moved since June, so the total it collects from customers each year is shrinking (down 5.4% from a year earlier, to $1.40B).

CEO Alex Timm said larger carriers raised marketing and cut prices during the quarter, so Root gave up direct-channel growth where returns fell below its targets. CFO Megan Binkley expects year-end policies in force to be “broadly in line with the third quarter.”

But new business through partners and independent agents still rose 19% y/y and made up 52% of the total, up from 44% a year earlier.

The quarter’s big partnership win was Carvana, the online used-car seller, which renewed its embedded insurance deal (coverage sold with the car) on August 31. Timm said the program had sold “well over 250,000 policies” by then.

Carvana’s new warrants for up to ~1.5M shares (~10% of the share count) vest on policy milestones through August 2028, so it only earns them by selling more Root policies.

In October, Root also launched a pilot with Hyundai, one of its existing partners. At 14 Phoenix-area dealerships, buyers who finance or lease a new Hyundai through Hyundai Capital America get six months of Root insurance free if they enroll and qualify. The pilot puts Root in front of car buyers at the dealership, the way Carvana does online.

Loss Ratio

The gross accident period loss ratio (losses from crashes in the quarter over premium earned) rose to 61.1% from 56.5% a year ago, which the 10-Q attributes to higher repair and medical costs per claim, more frequent claims, and a shift in channel mix.

But 56.5% is Q3 2025 after a year of revisions. Its first estimate was 59.5%, and every quarter since Q1 2025 has been revised down so far, by 0.5 to 3.4 points:

Regardless, 61.1% is still within Root’s 60-65% target range, and if Q3 develops the way 2025’s quarters did, its loss ratio ends lower. Timm also called claim frequency and severity “fairly benign,” with nothing “outside of the bounds of statistical noise” from quarter to quarter.

Binkley expects Q4 toward the top of the range, since November and December bring more animal collisions and bad weather.

The net combined ratio (losses plus expenses over premium; under 100% is an underwriting profit) improved 4.9 points to 97.2%. But Carvana warrant expense fell $15.3M after a one-time catch-up in Q3 2025, worth ~4.3 points, so the underlying improvement was under a point.

Growth Levers

Root began rolling out its new pricing model in Q4, which it expects to be its biggest jump in predictive power yet.

Timm said it should lower average premiums by “maybe a couple of points” while “meaningfully increasing conversion,” and the rollout runs through H1 2027.

Root also entered New Jersey in July and then Wyoming, bringing it to 38 states, and targets being near national by the end of 2027, with the larger new states coming last. Timm said New Jersey was Root’s most successful state launch yet in new business per capita.

Both levers make sense, but neither shows up in the numbers this year. The new pricing model only grows premiums if the extra conversion and retention outweigh the lower prices, and the states Root hasn’t entered yet cover just ~18% of the U.S. population.

So I’d expect policies to stay roughly flat until the rollout finishes in mid-2027.

Valuation

Root bought back 290,694 shares in Q3 at an average of $53.15/share ($15.5M), leaving $38.8M on the authorization, ~6% of today’s market cap.

At $43.37/share, the market cap is $659M on 15.2M shares, or 2.0x the $327.0M book value as of September 30.

Net of the $261.6M held outside the insurance subsidiaries, you’re paying $397M for the insurance operations, or 3.9x nine-month net income annualized (Q4 is the seasonally weak quarter).

Disclaimer: This write-up is for informational and educational purposes only and is not investment advice. I may hold positions in the securities discussed. Do your own research before buying or selling any security.