NEW YORK ACCOUNTABILITY COALITIONDoing What the Campaign Can't
How we defined the race early, drew the contrast, and kept working people with Tom.
NEW YORK ACCOUNTABILITY COALITIONDoing What the Campaign Can't
How we defined the race early, drew the contrast, and kept working people with Tom.
The fight was in the city.
of the statewide Democratic primary vote came from New York City.
Outside the city it wasn’t a contest. DiNapoli took roughly 72% across suburbs and upstate.
Paid spending was highest in NYC across every campaign and IE — that’s where the race was won and lost.
Every other establishment candidate got wiped out. Tom didn’t.
Up and down the ballot, insurgents took down the establishment. Tom was the exception — because we set the terms of the race early.
In a night that punished the establishment across the board, Tom finished on top — and finished it by 9:01 PM.
Tom started this race a vulnerable incumbent.
Tom needed help — he was a vulnerable incumbent facing not one but two primary challengers who could potentially double-team attacking Tom early in the race.
Four possible stories.
We rule out 1 and 2, then weigh 3 and 4.
Progressive precincts engaged the race.
If progressive surge voters had turned out and skipped Comptroller, dropoff would rise in the most progressive precincts.
It does the opposite: dropoff falls from ~17% to ~7%. The most progressive voters completed the race at the highest rates and still chose DiNapoli.
It explains NY-07 — not NY-10 or NY-13.
Combined, the two progressives out-polled DiNapoli in the most progressive precincts (65% to 34%).
But a split only matters where the winner falls short of a majority. DiNapoli cleared 50%+ in NY-10 and NY-13. Only NY-07 turned on the split.
It wasn’t for lack of money.
The challengers had the resources — Goyle matched Tom ($3.1M) and outspent him on comms; Warshaw spent $2.4M. Drew and Raj had the money; their message failed to land.
Stacked on top of Tom’s $2.0M in campaign comms, our IE closed the gap and put him on top.
Drew and Raj reached voters, but couldn’t close.
Drew and Raj weren’t invisible — their combined share climbed from 25% to 65% as precincts got more progressive.
But they couldn’t close. Even in the most progressive precincts, a third of voters — theirs by ideology — stayed with DiNapoli.
Where the wave was strongest, Tom ran furthest ahead.
A large share of these voters cast ballots for progressives and Tom — the same voters who powered Valdez, Lander, and Avila Chevalier stayed with Tom for Comptroller.
Three strategies that shaped the program.
This was the plan we outlined in memos from the start — strategic knowledge, not hindsight.
Launch in April to define the race — forcing opponents into a difficult strategic dilemma and preserving Tom's ability to keep his powder dry.
Own the attack on Warshaw & Goyle so Tom's own campaign could stay positive. Our strategic assumption was Tom wouldn't go negative — so we did.
Mamdani surge voters — unengaged before the last year — were likely to turn out strong for whoever was furthest left. We had to plan targeting and messaging so that surge didn't drag Tom's number down in the contested congressional districts.
A precisely targeted universe to maximize our impact.
This was not a broadcast campaign. This was a targeted social, digital, and streaming TV campaign aimed at the 16% of registered Democrats who would decide this election.
- •Dems who voted in both the 2018 and 2022 statewide primaries.
- •Show up under any turnout scenario — our floor.
- •Modeled progressive Dems with strong general-election history.
- •Persuasion targets — the voters Warshaw and Goyle needed.
- •NYC Dems who voted in the 2025 mayoral primary with no prior primary history.
- •Targeted at key competitive congressional primaries where turnout was expected to surge around high-profile Mamdani endorsements.
These three audience segments combined for a total of 966,911 voters — just 16% of registered Democrats in the state. Total votes cast for Comptroller: 879,689.
Go early, boost Tom’s favorables.
- →2/3 of primary voters didn’t know Tom.
- →Most campaigns want to sit on their money until closer to the election.
- →We spent early to introduce Tom — by the time he went on air, his platform and name ID were already built.
“He is taking retirement funds very seriously and protecting people’s money.”
- ✓ Record of protecting pensions — our top-performing message.
- ✓ Fiscal-watchdog framing lands with progressives and independents alike.
Source: NYAC AIMI research, Spring 2026.
Flight 1: We Trust Tom.
16× avg. cross-platform screen exposures
Social engagement 2× benchmark
Workers with a stake in the pension made the case for Tom.
Hearing directly from the nurses, firefighters, and public workers whose retirements Tom protects moved the message from politics to trust — and elevated his credibility with primary voters.
Warshaw was forced out of the gate.
Going early forced Warshaw to open his checkbook from the jump. He rushed this ad to air — it was his big TV spend, and it didn’t land with voters. It set the tone for the rest of his campaign.
Drew Warshaw’s launch spot
YouTube · April 30
NYAC set the tempo of the race — opponents reacted to us.
NYAC set the pace — the rest of the field followed.
Source: linear TV spend for the three hard-side campaigns as reported to the New York State Board of Elections; NYAC IE spend per internal streaming TV records. Hard-side totals reflect linear TV only and do not include digital or other paid media.
Take the contrast lane.
Wealthy political insider benefiting from privilege and elite political connections.
GOP-aligned history in Kansas. Voted with Republicans 80% of the time. Opportunistic.
Hitting either opponent alone simply pushed voters to the other — but drawing the contrast against both together moved them to Tom.
Tom was the only candidate being attacked.
We leveled the field by surfacing our opponents’ records, so Tom could stay positive.
“I learned that there are two pseudo-progressives running.”
Flight 2: Phony Progressives.
13× avg. screen exposures
Social engagement 3× benchmark
13× avg. screen exposures
Blunt the DSA turnout surge — land the contrast where it mattered.
We anticipated higher turnout in DSA-aligned congressional districts and concentrated spend there so the Raj & Drew contrast reached the voters most at risk of drifting.
Source: NYAC program spend by district, all flights combined.
This could have been a different story.
We raised and spent expecting late attacks that never fully came. A few variables cut for us — and a few didn’t materialize against us.
Opponents had negatives on Tom that our polling showed as effective — especially Palantir and ICE. They never put paid behind them. We can’t explain why.
- •Palantir / ICE contract attacks tested strong — never aired at scale.
- •Drew went dark in the final month. Raj stayed up — but focused only on positive messaging.
- •This refers to paid only. Both were vocally negative on organic social — but without paid resources behind it, that organic reach was never going to travel far.
- ✓Tom’s debate performance. He delivered.
- ✓Letitia James’s endorsement. Gave progressives permission to vote for Tom. NYAC made the “phony progressives” claim — she validated it.
What we accomplished.
The team.
Thank you.
