Your City Is Worse at Filling Potholes Than New York City. Yes, That New York City.
New York City responds to potholes in about 2 days, most cities don't
Inverted incentives. Prevention is 6-to-14x cheaper, but savings take decades. Budget cycles run annual. Officials are going to defer.
Remove/Reduce the ability to defer. Cities that implemented automated dispatch means a detected pothole triggers a repair order without a human deciding whether to act.
Institutions compound, mayors don’t. NYC’s system over the past 20 years got better across four administrations because each inherited what the last built. The machinery survives the election.
On March 15, 2026, a Saturday, more than 80 New York City Department of Transportation crews fanned out across all five boroughs starting at 6 a.m. and filled 7,200 potholes before the day was over. That is roughly a full week’s worth of repairs compressed into a single shift. A second five-borough blitz followed the next weekend. By mid-March, the city had repaired over 66,000 potholes since Mayor Zohran Mamdani took office on January 1, maintaining an average response time of about two days. And they were doing it in the worst pothole season in recent memory: 311 complaints were running roughly 33 percent above the same period in 2025, driven by what Commissioner Flynn described as an extended cold spell that caused potholes to form “all at the same time” once conditions finally thawed.
The numbers are impressive. But the numbers are not the point (in this case). The point is the bet Mamdani is placing; one most American mayors never would (FOR SOME REASON).
The conventional (and cliché) political playbook says mayors earn capital from new things. Bridge openings. Park dedications. Groundbreakings with hard hats and oversized scissors. Economists have a name for this tendency: the “ribbon-cutting bias,” describing the documented preference of political systems for new construction over maintenance. The logic is straightforward. Maintaining existing infrastructure is invisible, unglamorous, and generates no ceremony. Building something new gets your name on a plaque.
Mamdani is running the opposite experiment. His framing, “no problem too big, no task too small, no pothole too deep,” is not subtle, and it does not need to be. Eighty crews at dawn is the ceremony. The Saturday deployment schedule guarantees local news coverage, or at least crew-members’ TikTok videos. The real-time 311 dashboard is that oversized ribbon. The pothole count is a metric every voter understands. He is making maintenance visible, making it fast, and betting that mundane competence generates more durable political loyalty than any groundbreaking ever could.
And if that bet pays off in New York City, the implication for every other American city is uncomfortable. If Mamdani’s DOT can hold a two-day average response time under historically bad conditions, the question for every other mayor is pointed: what’s your excuse? Can you beat the city that can’t get a building permit processed in under six months?
I think the answer, for most cities, is that they could. They just haven’t.
Before we explain why, a few things to note. A 2015 audit by the New York City Comptroller found that DOT’s tracking system contained duplicate and triplicate entries that inflated reported repair counts. Whether those data quality problems have since been corrected is not established in available documentation. Every pothole figure in this article, from every administration, should be read with that finding in mind. Beyond the counting question, more than a quarter of pothole service requests were still listed as open, pending, or in progress at the time of a recent review, and Queens alone accounts for nearly half of the city’s 11,300-plus complaints. The two-day average masks a distribution with real tails. And on March 9, Jaikarran Seenarian died after his electric scooter struck a pothole on Liberty Avenue in Richmond Hill, Queens. Like always, be very wary of aggregate statistics, they do not capture the distribution of failures. A single unaddressed pothole has killed.
His roughly two-day average sits slightly above the Adams administration’s benchmark of 1.8 days, but maintaining that level of performance under a 33 percent demand increase is arguably the harder achievement from an ops standpoint, especially how nasty the winter storms were.
The caveats are real. The improvement is also real. And the question this article sets out to answer is not simply “why are there potholes?” but something harder: why do cities that know potholes are coming, know how to prevent them, and have the technology to fix them still fail, year after year, to get the problem under control? The answer is institutional. And NYC’s trajectory from unmeasured chaos to two-day response is the proof.
But first: how big is the problem we are actually talking about?
How Big Is This (Pot)hole?
Pothole damage costs American drivers more than the GDP of Iceland. A 2022 AAA survey put the figure at $26.5 billion in vehicle repairs in 2021 alone, roughly $600 per affected driver, with one in ten drivers sustaining damage severe enough to require a mechanic. Earlier AAA estimates from a multi-year survey had put the figure at around $3 billion annually. The nearly ninefold jump reflects both worsening road conditions and the rising cost of vehicle repairs, but even the lower estimate understates the problem, because it counts only the direct cost to drivers and ignores the costs borne by municipalities, transit systems, and taxpayers.
In New York City alone, potholes and road defects cost the city nearly $138 million in legal settlements for pedestrian injuries and vehicle damage over a six-year period, on top of an additional $7.3 million the city allocated to DOT in a single winter to address an extraordinary volume of repairs. Multiply that pattern across thousands of American municipalities and the total dwarfs any individual city’s budget.
The 2025 ASCE Report Card for America’s Infrastructure gave roads a D grade. As the Report Card documents, 39 percent of major U.S. roads remain in poor or mediocre condition, an improvement from 43 percent in 2020 but still a failing grade by any standard that takes driving seriously. Poor road conditions cost the average driver an estimated $1,400 per year in vehicle damage and time lost to delays.
But the backlog is worse than the current snapshot suggests. The Pew Charitable Trusts, in a careful 2025 analysis of Bureau of Economic Analysis data, found that state and local governments have accumulated a deferred maintenance liability of nearly $105 billion on roads and bridges since 2004. That number represents the cumulative gap between what governments spent on preservation and the value of annual depreciation. Even with nominally increasing dollar figures, real, inflation-adjusted road investment actually declined over the first two decades of this century. ASCE estimates that an additional $2.9 trillion is needed across 11 infrastructure sectors to reach a state of good repair, with a $684 billion gap specific to roads over the coming decade.
And the damage is not distributed evenly. Research from Mavis Tire finds that the Northeast experiences 60 percent more pothole-related tire repairs per location than shops in the South and West. The reason is physics: freeze-thaw cycles are the primary mechanism by which potholes form. Water penetrates pavement cracks, freezes, expands, and then thaws, progressively destroying the pavement structure from within. The Northeast and upper Midwest are structurally disadvantaged by their climate, and no technology fully overcomes that.
Most people’s response (especially suburban voters with pavement princesses) to numbers like these is to say the country needs to spend more on roads. That is true, to a point. It’s more about the distribution of the money being spent, rather than spending a total amount.
The Money Problem Is Real. Also Not Real.
The funding argument is correct as far as it goes. The most structurally important fact about American road funding is one that almost never appears in local news coverage of potholes: the federal gasoline tax has been frozen at 18.4 cents per gallon since October 1993, and it is not indexed to inflation. According to the Peter G. Peterson Foundation, that 18.4 cents is now worth roughly 45 percent less in real terms than when it was set. Wikipedia’s fuel tax tracking datashows that inflation has risen more than 122 percent since 1993 with the rate unchanged, and construction costs for roads have risen faster still. The net result, as Transportation for America documents, is that annual gas tax revenue declined by roughly $10 billion per year in real terms from 2010 to 2025. Congress has been patching the Highway Trust Fund with general revenue transfers, $53 billion over a recent five-year period according to Tax Notes, rather than raising the underlying tax.
Why does no one raise it? Before the whole breakdown in the Persian Gulf, increasing a visible per-gallon tax is politically costly in a way that deferred road maintenance is not, at least not until the pothole blows out a tire and the driver calls 311. The result is a revenue system built on a 1993 assumption that has decayed in real terms by more than half, funding a network of roads that costs multiples more to maintain than it did three decades ago.
The Infrastructure Investment and Jobs Act, signed in November 2021, provides approximately $500 billion for roads and bridge programs over five years, a 34 percent increase in surface transportation funding. That is substantial. It is also, given the scale of deferred maintenance, not sufficient to close the gap. And the IIJA is set to expire in September 2026, with reauthorization most likely going to be a no.
All of that is true. But even the money we do have goes to the wrong places.
The Congressional Budget Office has documented this: in 2022, 96 percent of the $52 billion the federal government spent on highways went to capital investment, meaning new construction and major rehabilitation of existing structures, rather than to routine operations and maintenance. That allocation has been typical since the 1950s. The federal government builds. States and localities maintain. And CBO’s reauthorization analysis makes the structural mismatch explicit: federal highway funds are restricted to “federal-aid highways,” the Interstate system and other major roads, leaving most local streets and roads entirely outside the program. The neighborhood streets where potholes cause daily misery are, by design, ineligible for the bulk of federal road dollars. Local roads account for 99 percent of road length in the United States and carry the majority of pothole complaints, yet they are maintained by counties and municipalities from their own general fund revenues, with no systematic federal support.
Even if Congress doubled highway funding tomorrow, most of the increase would flow to state DOTs for state highways. The potholes on your street would remain your city’s problem, funded from your city’s budget.
But spending does not reliably predict outcomes. High-spending cities have persistently terrible pothole records. Lower-spending cities with better-designed systems sometimes outperform them. If money were the binding constraint, that could not happen. Something else is going on.
The way American cities maintain roads is shaped by a set of interlocking incentive problems that make individually rational people choose collectively expensive, irrational outcomes, repeatedly, predictably, and at every level of government. This makes road spending one of the few places where we can cut budgets, and quality can improve. Reducing the amount of roads built, redirecting the existing budgets towards deferred maintenance, or even making roads more manageable by shrinking them.
Why Cities Keep Choosing the Most Expensive Option on the Menu
Consider the math. The Federal Highway Administration has documented that every dollar spent on preventive pavement preservation can save up to six dollars in future reconstruction costs. Michigan DOT research found that proactive maintenance is 14 times cheaper than reactive maintenance, and estimated that the state could save $700 million over eight years through systematic preventive intervention. A widely cited 1996 FHWA report puts preventive treatments at four to five times more cost-effective than allowing pavement to deteriorate to the point of reconstruction. Research summarized by the IMF and FREE Network finds that one dollar of deferred maintenance generates over four dollars in future repair costs, with some industry estimates running as high as seven to one over five years.
The physics compound the economics. Pavement deterioration is non-linear. It begins with cracks. Water (in addition to making new cracks) infiltrates the old cracks, freezes, expands, and destabilizes the road’s base layer. Traffic then removes the compromised asphalt in chunks. Each stage accelerates the next. A crack that costs pennies to seal becomes a pothole that costs dollars to patch and eventually a structural failure that costs thousands to reconstruct.
This isn’t new; most civil engineers will tell you this. That means every transportation agency in America is fully aware. And yet cities systematically choose cold-patch repairs, which last approximately one year, over permanent repairs that require excavating and replacing the base material. The same pothole gets filled three or four times before it is finally reconstructed at five to seven times the total cost of a single proper repair. The math is unambiguous. The behavior is economically irrational. So why does it persist?
Because the people who pay for prevention are not the people who benefit from it. Preventive treatments yield savings over a 10-to-20-year horizon. Municipal budget cycles run annual. The official who authorizes a $50,000 crack-sealing program today will not be in office, and may not even be in the same job, when the savings materialize. The official who lets roads deteriorate and then funds a high-visibility pothole-filling blitz gets to take credit immediately, on camera, with a crew and a truck behind them. The incentive structure is inverted. The economically rational choice and the politically rational choice point in opposite directions, and politics wins every time.
The ribbon-cutting bias runs deeper than preference for novelty. The IMF and FREE Network research identifies four compounding reasons that governments systematically underinvest in maintenance: political economy, meaning officials prefer projects that generate visible credit; fiscal, meaning maintenance budgets are cut first in lean years; institutional, meaning capital and operating budgets are managed separately (with FHWA’s own guidance actually drawing a sharp bureaucratic distinction between “preventive” and “corrective” maintenance that reinforces the silo); and informational, meaning asset condition data is often simply unavailable. The Economic Policy Institute adds a further wrinkle: because most states operate under balanced budget requirements, infrastructure maintenance is discretionary spending that gets squeezed during recessions, exactly when potholes proliferate most as neglected roads face winter freeze-thaw cycles and revenue-starved cities defer even routine crack-sealing. Maintenance spending is procyclical. Conditions are countercyclical. The system is designed to fail at the worst possible moment.
Even if these incentive problems were fixed, the underlying math of American development patterns would still make the task harder every year. Since World War II, American municipalities have systematically extended road infrastructure to accommodate new development, accepting near-term tax revenue in exchange for long-term maintenance obligations that the new revenue stream rarely covers. Strong Towns, the nonprofit led by Charles Marohn, has built an influential body of work arguing that this pattern constitutes a fiscal Ponzi scheme: cities use revenues from new growth to service the maintenance obligations of previous growth, requiring ever-more growth to remain solvent. The “Ponzi” framing is contested — Market Urbanism and Better Cities Project note that the empirical evidence is uneven, and City Journal has found the analysis broadly credible while questioning the label. But the core insight, that municipalities routinely assume maintenance liabilities well in excess of the tax revenue new development generates, is well-grounded. The practical consequence: total lane miles grow; maintenance budgets don’t. Consider Tulare, California, which told the state legislature in 2017 that, forced to prioritize arterial and collector roads due to funding shortfalls, its residential street network had declined from a Pavement Condition Index of 61 in 2010 to 53 in 2016, and the cost to restore it was growing exponentially with each year of deferral.
Fragmentation compounds everything else. On any given urban road, responsibility for the pavement may be shared among city, county, state, and private utility companies. The Federal Highway Administration’s manual on pavement utility cuts documents how utility excavations, in Washington, D.C., alone more than 5,000 per year at peak, structurally weaken adjacent pavement and dramatically shorten road lifespan. Utility companies are typically required to patch their cuts temporarily, but permanent restoration is often delayed years due to funding gaps. When the resulting pothole appears, it is frequently unclear which agency is responsible, and each agency has an interest in pointing to the other. Cities like Cleveland explicitly disclaim responsibility for utility-cut repairs, referring drivers back to the contractor. The driver, meanwhile, does not care whose jurisdiction the hole is in. They want the hole filled.
Former New York City DOT deputy commissioner Lou Riccio, speaking to Crain’s New York Business, captured the structural problem with characteristic directness: “New York City streets, in particular, get cut so often it’s like continuous open-heart surgery.” Every utility repair, every underground pipe, every cable installation reopens the street and disrupts freshly laid asphalt. Riccio estimated that the city needs at least 1,000 lane miles of resurfacing per year just to maintain current quality, and more than 1,200 to actually improve conditions.
The result is a system where no one bears the cost of what they defer. Federal policymakers who set gas tax rates do not drive on the local roads that deteriorate when revenues fail. State legislators who cut transportation budgets during recessions are not accountable for the potholes that appear three winters later. City officials who defer crack-sealing this fiscal year have moved on by the time the road requires full reconstruction. Utility companies that cut pavements are not on the hook for the premature road failure that follows. The FHWA and the Pew Charitable Trusts are right that better data and asset management plans would help. But the deeper problem is not informational. It is incentive-structural. Until that changes, the potholes will persist.
So what does it look like when a city breaks this pattern? The answer has less to do with technology than you might expect, and more to do with closing the loop between knowing about a problem and doing something about it.
The Cities That Broke the Pattern
In Chicago (to most people’s surprise), the Department of Transportation filled approximately 143,000 potholes in the first two months of 2024. February alone: over 93,000 potholes, a 63 percent increase over February 2023 and a 100 percent increase over February 2022. But the volume is not what makes Chicago interesting. What makes Chicago interesting is the systems design behind the volume.
The city’s 311 portal feeds directly into the open data platform at data.cityofchicago.org, allowing both internal management tracking and public transparency. Service requests are generally completed within seven days. The design choice that matters most: when a crew arrives to repair a 311-reported pothole, they do not fill that one hole and leave. They fill every pothole on the block. This sounds obvious. It is not. Most municipal repair systems dispatch crews to specific complaints. Chicago’s system dispatches them to streets, which means each complaint triggers a sweep that catches unreported defects. It is a force-multiplier that prevents the whack-a-mole dynamic of returning repeatedly to the same location. And because progress is publicly visible on a real-time dashboard, elected officials have a political incentive to maintain performance, not merely to claim it. Transparency creates accountability.
Pittsburgh illustrates a different facet of the same principle: that data quality itself can be a form of governance. Through the Bloomberg Philanthropies What Works Cities initiative, the city’s Department of Innovation and Performance built “Dashburgh,” a public data dashboard presenting 311 data broken down by neighborhood, providing granular accountability the city had never had before. But the more consequential innovation was internal. Pittsburgh’s Department of Public Works built analytical tools that automatically link multiple requests for the same pothole into a single ticket, flag requests that lack precise location details, and, most importantly, identify street segments that have been patched three or more times in two months. That last flag is the key. A street segment patched three times in eight weeks is not a street that needs patching. It is a street whose underlying pavement needs rehabilitation. The data system recognizes the pattern that human decision-making, under the pressure of daily complaint volume, reliably misses. This is the shift from reactive to systemic maintenance in microcosm: using information to detect when a short-term fix is producing long-run waste.
In Memphis, the breakthrough came from an unlikely place: the city’s transit buses. Working with Google Cloud partner SpringML, Memphis equipped its buses, which were already running cameras for traffic monitoring, with an AI system that analyzes road footage to detect and classify potholes. The model was trained on existing images, refined with data from higher-quality cameras, and then integrated directly with the city’s 311 work-order system so that confirmed potholes automatically generate a repair ticket. No human triage step. No 311 call required. The results: a 75 percent increase in potholes detected compared to prior manual methods, an estimated cost saving of approximately $20,000 per year, and 63,000 potholes repaired in one year. The lesson from Memphis is simple: use what you already have. The city did not buy a specialized detection fleet or hire a team of data scientists. It leveraged buses already running existing routes, cameras already in place, and cloud tools developed for other purposes. The marginal cost of detection approached zero.
Tarrant County, Texas pushed the detection-to-dispatch connection even further. The county integrated reports from Waze, the navigation application with over 140 million users that already allows drivers to flag road hazards, directly into its work-order system. A Waze pothole report auto-generates a maintenance dispatch with no human triage step. No 311 operator, no supervisor review, no scheduling queue. The bottleneck that plagues most municipal repair systems, the chain of handoffs between someone knowing about a pothole and someone doing something about it, is eliminated entirely.
And in Kansas City, Missouri, a pilot program tried to go one step further: not just closing the loop between detection and repair, but getting ahead of the problem entirely. The program combines traffic camera data, weather data, traffic volume, and pavement condition assessments to predict where potholes will form before they appear, with the goal of applying preventive sealant while the road is still intact. Chief Innovation Officer Bob Bennett reported early successes, with fewer potholes reported the following spring, though the sample was small and the causal mechanism difficult to isolate from weather variation. The program comes closest to addressing the root cause: spending money today where data suggest it is needed, rather than where the public happened to complain yesterday.
Bennett also happens to have the best one-liner in the pothole policy literature. Telling Governing magazine: “If we fail to fill the potholes or pick up the trash, we’re going to hear about it. Potholes are one of those things people kvetch about.” He is right that potholes generate immediate, visceral political pressure. The problem is that the political energy they generate goes to reactive speed, not systemic prevention. People kvetch about the pothole in front of their house. They do not kvetch about the crack-sealing program that was cut from the budget eighteen months ago.
What connects Chicago, Pittsburgh, Memphis, Tarrant County, and Kansas City? It is not any particular technology. Chicago uses spray injection trucks. Memphis uses machine learning. Tarrant County uses Waze. Pittsburgh uses internal analytics. Kansas City uses predictive modeling. The technologies are different. The underlying design principle is the same: each city closed a feedback loop connecting detection to dispatch to performance measurement to public visibility. Each removed the discretion and the delay that normally sit between knowing there is a problem and acting on it.
The absence is equally telling. Pennsylvania DOT’s AI road survey program, which partnered with RoadBotics to survey over 2,500 miles using computer vision, found that standardized defect scoring improved regional coordination and reduced inspection costs. But translating condition data into repair schedules required sustained institutional commitment that not every participating municipality provided. Better cameras produced better-documented backlogs unless someone rewired the institution between the camera and the crew.
Technology transfers. Institutions do not.
When did New York start to do something right?
Considering how people feel about the city, this is a bit of a shock. Yet, no city proves the compounding power of institutional investment more completely than New York (in this case), and no city’s story makes clearer that the improvement belongs to the institution, not to any single occupant of the mayor’s office. What Mamdani’s crews did on that Saturday in March 2026 was possible because of decisions made in 2007, 2010, 2014, and 2022 by people who could not have known they were building a system together.
Start at the floor. A 2002 audit by the New York City Comptroller found that DOT lacked any useful operational standard for pothole repair. The agency’s informal target was merely to close 65 percent of repair orders within 30 days. The auditors noted that the 30-day threshold was set arbitrarily, with no operational rationale. DOT could not explain why 30 days rather than 15, the number that actually mattered, since NYC Administrative Code Section 7-201(c)(2)shields the city from civil liability only if potholes are repaired within 15 days of complaint. The tracking system was unreliable. Repair counts were inflated by duplicates and triplicates. There were no written policies or procedures governing the work. This was the institutional floor: roughly 100,000 potholes filled per year, by a count that could not be trusted, on a timeline the agency could not reliably measure.
Michael Bloomberg’s contribution to pothole management was primarily infrastructural rather than operational, and it is the kind of contribution that gets no credit at the time and turns out to be indispensable later. Three investments from his tenure shape everything that followed. First, the city had long operated one municipal asphalt plant at Hamilton Avenue in Brooklyn. Bloomberg opened a second plant in Queens in 2010, giving DOT the materials supply, and the independence from outside vendors, that makes large-scale blitz operations financially and logistically viable. NYC DOT is one of the rare urban agencies in the country that produces its own asphalt, and that structural asset remains central to every subsequent administration’s pothole story, including Mamdani’s. Second, in 2007 and 2008, Bloomberg launched the Street Conditions Observation Unit, or SCOUT, a team of inspectors deployed on three-wheeled scooters to traverse every one of the city’s 6,000 miles of streets at least once per month, reporting conditions via GPS-enabled devices linked directly to 311. As Gothamist reported at the time, Bloomberg’s framing was explicitly about shifting from reactive to proactive: “It’s government’s responsibility to find the problems and fix them, not to sit there and say ‘Duh we didn’t know.’” Third, Bloomberg’s DOT under Commissioner Janette Sadik-Khan launched “The Daily Pothole”, a public-facing blog tracking repair efforts in real time with maps and crew counts. Trivial by today’s standards. Significant at the time. It created a transparency norm around pothole performance that each subsequent administration has had to honor, escalate, or outrun.
And yet, for all of this institutional investment, the operational numbers barely moved during the Bloomberg years. A data science analysis by researchers at Lehman College examining 311 records found a mean repair time of approximately 4.7 days with a standard deviation of seven days, meaning the distribution had a very long right tail. The Adams administration’s own retrospective benchmarks Bloomberg at approximately 4.4 days average. By fiscal year 2015, a second Comptroller audit found the average at 5.6 days on local streets, with nearly 10 percent of complaints unresolved past the 15-day legal liability threshold. The infrastructure was better. The operations had not yet caught up.
Bill de Blasio came into office in a brutal pothole winter. By his own DOT’s count, 113,131 potholes had been patched by early 2014 before his comprehensive plan was even announced. But the move that mattered was not reactive. It was strategic. De Blasio shifted the city’s fundamental logic from patching potholes after they appeared to preventing them from forming in the first place. The centerpiece was a $1.6 billion commitment over ten years to resurface New York City’s streets. The theory of change was explicit: a freshly resurfaced street is less likely to develop potholes, meaning fewer complaints, fewer reactive repairs, and lower long-run costs. This is exactly the prevention-over-reaction economics that FHWA had documented for decades, applied at city scale for the first time.
It worked. Between 2014 and 2018, the de Blasio administration repaved 5,000 lane miles and drove pothole counts down by 44 percent. By the end of his tenure, DOT had resurfaced 8,356 lane miles total and filled more than 2 million potholes. He also invested in field operations technology: crews received tablets for one-touch, real-time tracking of work, replacing paper-based reporting with digital logistics that could route crews more efficiently. DOT introduced a warm-weather asphalt mix requiring less heat, reducing energy use and extending the paving season closer to year-round. Average response time fell from Bloomberg’s 4.4 days to approximately 3.4 days (though official 2021 releasesreport 3.83 days, suggesting the improvement concentrated in later years). But the 44 percent reduction in pothole formation arguably mattered more than the response-time improvement. Fewer holes means less demand on the repair system even at constant operational capacity.
Eric Adams inherited the prevention framework and operationalized it. By January 2025, his administration claimed an average response time of approximately 1.8 days, nearly a full day faster than de Blasio and more than twice as fast as Bloomberg. But the more revealing number was demand: 88,466 pothole complaints via 311 over the first three years of his administration, compared to 147,640 at the same point under de Blasio, a 40 percent reduction. The administration resurfaced an average of nearly 1,200 lane miles per year, right at the expert-estimated threshold Riccio identified, and consistently deployed 25 to 75 crews during pothole season. When Adams framed his 500,000th pothole milestone as filled “in half the time it used to take” because “proactive paving of 1,200 miles of lanes” cut pothole formation, that was the correct analytical framing. Improvement came from two compounding forces: faster dispatch and lower demand. Both were improving simultaneously.
Mamdani inherited a better system under harder conditions. Approximately two-day average response under 33 percent elevated demand. Over 66,000 potholes through mid-March. The blitz format itself, 80-plus crews in coordinated Saturday deployments, is not new; DOT’s own 2021 press releases from the de Blasio era describe the identical model. What Mamdani adds is not operational novelty. It is the political insight that maintenance, made visible and fast, can do the work that ribbon-cuttings do. Executive attention flows resources and accountability to operations that might otherwise drift. Whether that attention converts into structural improvement, specifically the 1,150 lane-miles resurfacing commitment for 2026 that represents the preventive investment rather than the reactive repair, is the unanswered question. And whether the response time can reach same-day is the unanswered dare.
Most of what drives these numbers is (again) institutional, not mayoral. Asphalt plants, crew systems, 311 integration, and resurfacing programs were built over decades and persist across administrations regardless of who occupies City Hall. The most important long-run metric is not response time but pothole formation rate, which is a function of resurfacing investment, utility excavation coordination, and pavement quality. Riccio’s description of “continuous open-heart surgery” captures a structural vulnerability that no blitz weekend can permanently fix. And all of this data is self-reported by agencies with strong incentives to report improvement. The one time an independent auditor looked carefully, in 2015, the numbers were inflated. We do not know if that problem has been corrected. The evidentiary gap we would most like to close is independent verification of DOT pothole counts, along with borough-level and complaint-source breakdowns that would let us assess the distribution of the response-time average, not just its mean.
What Works, What Is Coming, and the Gap That Is Not Technical
Every success story so far shares a principle: institutional design matters more than technology choice. But the technology is changing, some of it is promising, and understanding what is deployable now versus what remains speculative matters. The persistent question is whether any of it changes the institutional calculus or merely produces more polished documentation of the same dysfunction.
The most commercially mature technology is computer-vision road scanning. The basic model is straightforward: a smartphone or dash-mounted camera captures imagery at roughly 10-foot intervals as normal fleet vehicles run their routes. Images are GPS-tagged, uploaded to a cloud platform, and processed by neural networks that classify defects across a spectrum from hairline cracks to full-depth potholes, producing a color-coded condition map that feeds directly into maintenance management software. Vialytics, a German-American platform, now supports over 600 municipalities and recognizes 15 damage classes; case studies from municipalities including North Tonawanda and Metuchen document crews shifting from reactive patching to planned maintenance after a single survey drive. RoadBotics, now a Michelin company, partnered with the Pennsylvania Department of Transportation to survey more than 2,500 miles using AI-powered analysis, allowing 15-plus municipalities to adopt standardized defect scoring and compare conditions on a common scale. The program was expanded following strong results.
On the repair side, two methods significantly outperform the traditional throw-and-roll approach that remains standard in most cities. Infrared pothole repair uses infrared panels to heat existing asphalt to a workable state without burning it, allowing new hot-mix material to blend seamlessly with the old surface. The result is a seamless patch with no cold joints, eliminating the primary failure mode of conventional patching: water infiltration through seam edges. Repairs take under 20 minutes, and the recycling of existing asphalt reduces both material costs and environmental impact. Ohio DOT research comparing three patching methods found that infrared patches had significantly longer expected lifespans than throw-and-roll or spray injection, and proved more cost-effective for winter patching specifically. The limitation is not technical but organizational: the equipment requires upfront investment, the method demands trained operators, and it cannot address full-depth failures. It is most powerful as a preventive strategy, applied to developing cracks before they become potholes. Spray injection patching, which blasts debris with compressed air, injects asphalt emulsion, and covers with aggregate, is the workhorse for high-volume rapid response.
Better detection is valuable. Better repair methods are valuable. But both face the same institutional bottleneck. Does better detection actually produce faster repair, or does it produce better-documented backlogs? Memphis shows the connection working when detection is wired end-to-end into work-order dispatch. But Pennsylvania’s experience is the cautionary case: condition data improved coordination, but without sustained institutional commitment to translating data into schedules, the information sat unused in many participating municipalities. The technology is only as good as the institution it feeds into.
Further out, there is autonomous road repair. The ARRES PREVENT robot, developed by Robotiz3d in collaboration with the University of Liverpool and Hertfordshire County Council, passed its first live public road trial in Potters Bar, England, on March 6, 2024. The robot uses AI-powered imaging to detect cracks and road defects, then autonomously fills them with sealing material before they develop into potholes. Robotiz3d claims the technology could deliver a 90 percent cost reduction, a 70 percent increase in speed, and three times fewer CO2 emissions. These claims require heavy skepticism. They come from the developer, not from independent assessment. The trial was a single short demonstration, not a controlled evaluation. UK government funding amounted to just over 30,000 pounds. That said, the underlying concept, autonomous proactive crack sealing before pothole formation, directly addresses the root institutional failure we have been describing: the inability of reactive systems to intervene at the optimal moment in the pavement lifecycle. If the technology matures and costs fall, it could change the maintenance calculus. For planning purposes, it is not there yet.
A different approach comes from Professor Erik Schlangen’s laboratory at Delft University of Technology in the Netherlands: self-healing asphalt. The mix combines standard paving material with tiny steel wool fibers that make the asphalt conductive. When an induction machine passes over the road, the fibers heat up, melting the bitumen binder and allowing microscopic cracks to close and re-bond before they can propagate. The technology has been field-tested on a 400-meter stretch of the Dutch A58 motorway since 2010, and on 12 other Dutch roads by contractor Heijmans, including high-stress locations such as roundabouts and industrial estates. Laboratory research suggests road lifespan can be doubled if the treatment is applied approximately every four years. The material costs about 25 percent more upfront but is estimated to save the Netherlands up to 90 million euros per year if adopted nationwide. Delft’s team is also researching epoxy asphalt, which mixes epoxy resin with bitumen to create a surface less sensitive to temperature extremes, particularly relevant as climate change increases both peak summer temperatures and freeze-thaw frequency.
The limitations of self-healing asphalt matter, though. The process is not autonomous: the induction machine must be driven over the road on a schedule, which presupposes a systematic maintenance system capable of tracking when each segment was last treated. And the technology is most powerful on long-duration highway sections with infrequent disruption; urban roads that are excavated every few years for utility work gain little benefit from materials designed for 40-to-80-year lifespans. It has been in field trials for fifteen years and is still not standard practice even in the Netherlands. We would not bank on it for near-term planning.
Same Asphalt, Different Institutions
If American institutions are the problem, a natural question arises: do other countries actually do this better, or is road maintenance just inherently hard everywhere? Two countries offer useful answers.
Over 90 percent of Japanese road pavements are asphalt, the same material used in the United States. Japanese roads are notably smoother and better maintained, but not because of some exotic technology. Repair work is simply performed far more frequently, ensuring the road surface remains intact before deterioration can compound. Asphalt has a service life of roughly 10 years, but maintenance is applied on a rolling basis so that roads rarely reach the deteriorated state at which potholes form. Japan’s Ministry of Land, Infrastructure, Transport and Tourism mandates systematic periodic inspection, bridges and tunnels once every five years by close visual inspection, and applies the same preventive maintenance philosophy to pavement. Nighttime paving is common, allowing crews to work without traffic disruption and producing better compaction. Procurement standards enforce lifecycle cost thinking rather than lowest upfront bid. These practices are not secret. They are well-understood by American transportation engineers. What prevents their adoption is not knowledge but institutional structure: the fragmented governance of American local roads, the annual budget cycle, and the absence of performance-based procurement at the local level. MLIT documents that 30 percent of Japanese towns and 60 percent of villages have no civil engineering technicians in their workforce, and Japan still outperforms the United States on road maintenance, because the institutional system compensates for local capacity gaps in ways the American system does not.
The Netherlands tells a complementary story. Rijkswaterstaat, the national water and road authority, applies rigorous lifecycle cost analysis to pavement decisions and maintains strong data infrastructure for road condition monitoring. Dutch porous asphalt is designed to absorb both sound and water, reducing noise pollution and improving drainage, but its open structure makes it more vulnerable to raveling, which is why the Delft self-healing research is particularly relevant to Dutch conditions. And the Dutch case demonstrates the lesson most directly: even self-healing asphalt requires institutional infrastructure. The induction treatment every four years presupposes a systematic maintenance scheduling system capable of tracking when each road segment was last treated. Innovation without maintenance institutions behind it does not stick.
Neither country is doing anything technically exotic. They are just more institutionally disciplined: systematic preventive maintenance, funded predictably, implemented through performance-based procurement, and measured transparently. The United States has most of the technical tools needed to match their outcomes. What it lacks is the institutional structure to deploy those tools at scale and sustain them across election cycles.
The Feedback Loop Is the Fix
Which brings us back to Mamdani, and to what his administration actually proves, which is less than the press coverage suggests and more than his critics allow.
The Saturday blitzes are the reactive half of the equation; the half that matters less. Eighty crews filling 7,200 potholes in a day is operationally impressive and politically savvy, but reactive speed is not what separates good road systems from bad ones. Prevention is. The real test of the Mamdani administration is not whether it can fill potholes fast on camera. It is whether the 1,150 lane-miles resurfacing commitment for 2026 actually gets funded and executed; whether the preventive investment survives budget season, whether it compounds across the next administration the way de Blasio’s and Adams’s did. We do not know the answer yet. A man died from a Queens pothole during the blitz. A quarter of service requests were still open. The data has historically been inflated. The hypothesis that visible maintenance can generate political capital is interesting. We will see whenever Mamdani can prove it or not.
But if the famously dysfunctional government of New York City can build a two-day-average system over twenty years, the question for every other American city is a little uncomfortable and specific: what’s stopping you?
The honest answer, for most cities, is not that they lack the knowledge or the technology. That every incentive in the system (annual budgets, political time horizons, fragmented governance, frozen federal revenue) pushes against doing the thing that works. There are proven institutional models. The political economy is designed to prevent its adoption. Pretending otherwise is how you get another decade of cold-patching the same hole four times.
The cities that broke the pattern did not do it by wishing the incentives or the political economy away. They did it by building systems that made the right choice easier than the wrong one: automated dispatch that removes the discretion to delay, public dashboards that make inaction visible, data flags that catch repeat failures before they become structural collapse. They did not fix the politics. They routed around them.
That is the only honest prescription this article can offer. Not that your mayor should try harder (well, that mayor should; you still need political will at the end of the day). Not that the ribbon-cutting bias is a failure of imagination. It is a structural feature of American municipal governance, and it will outlast every mayor who tries to overcome it through sheer will. What does not have to outlast them is the institutional machinery (the asphalt plants, the 311 integration, the resurfacing commitments, the transparency norms) that makes the next mayor’s job easier whether they care about potholes or not.
Hilarious graphic