Northline Forensics: Counting vs. Measuring - What 593 Treasure Valley Bridges Show About Reading Condition Data

Short answer: in Canyon County, 2.06% of bridges carry a Poor condition classification when you count structures - but 5.34% when you measure deck area. The Poor-rated structures there average 2.6 times the deck area of the county's typical bridge. Statewide the relationship runs the opposite direction: Idaho's Poor-rated bridges average 0.70 times typical size. Same dataset, same year, two different answers depending on whether you count deficiencies or measure them. That distinction is the entire difference between a punch list and a capital plan, and it applies to buildings exactly as much as it applies to bridges.

Northline Forensics is a recurring segment. Each episode starts from a primary source document or a public dataset and derives something from it that has not been published elsewhere. This one is a dataset dive.

The source

The Federal Highway Administration publishes the National Bridge Inventory, and among its annual tables is Bridge Condition by County. The 2025 table reports, for every county in the United States, how many bridges are classified Good, Fair, and Poor, and the total deck area in each classification. It includes federal bridges. It is free, it is public, and almost nobody reads it at the county level.

The classification itself is worth understanding before looking at any numbers. FHWA determines bridge condition from the lowest of four NBI condition ratings: Item 58 (Deck), Item 59 (Superstructure), Item 60 (Substructure), and Item 62 (Culvert). If that lowest rating is 7 or above, the bridge is classified Good. If it is 4 or below, the classification is Poor. Ratings of 5 or 6 are classified Fair.

Two things follow from that definition. First, the classification is driven by the worst component, not by an average - a structure with an excellent deck and a substructure rated 4 is classified Poor. Second, it is a condition classification. Load capacity and load posting are tracked separately in the NBI and published in their own tables. Nothing in this article is an assessment of any individual structure, and nothing in it should be read as a safety judgment. This is an analysis of how condition data behaves when you slice it two different ways.

The raw numbers

Straight from the 2025 table, with no manipulation:

Ada County (FIPS 001): 302 bridges total - 134 Good, 165 Fair, 3 Poor. Deck area: 233,713 square metres total, of which 1,813 square metres is Poor.

Canyon County (FIPS 027): 291 bridges total - 86 Good, 199 Fair, 6 Poor. Deck area: 115,024 square metres total, of which 6,141 square metres is Poor.

Idaho statewide: 4,646 bridges total - 1,468 Good, 2,928 Fair, 250 Poor. Deck area: 1,846,239 square metres total, of which 69,356 square metres is Poor.

Those figures reconcile internally in both dimensions - the three condition classes sum to the totals for both bridge count and deck area in all three rows - which is a useful check that nothing has been transcribed wrong.

The derivation

Four calculations, shown so anyone can repeat them.

Poor share by count = Poor bridges divided by total bridges. Ada: 3 / 302 = 0.99%. Canyon: 6 / 291 = 2.06%. Idaho: 250 / 4,646 = 5.38%.

Poor share by deck area = Poor deck area divided by total deck area. Ada: 1,813 / 233,713 = 0.78%. Canyon: 6,141 / 115,024 = 5.34%. Idaho: 69,356 / 1,846,239 = 3.76%.

Average deck area of a Poor-rated bridge = Poor deck area divided by Poor bridge count. Ada: 1,813 / 3 = 604 square metres. Canyon: 6,141 / 6 = 1,024 square metres. Idaho: 69,356 / 250 = 277 square metres.

Size ratio = average Poor-rated deck area divided by average deck area of all bridges in that geography. Ada: 604 / 774 = 0.78 times. Canyon: 1,024 / 395 = 2.59 times. Idaho: 277 / 397 = 0.70 times.

What the numbers say

Statewide, and in Ada County, Poor-rated structures are smaller than average. That is the intuitive pattern: small rural crossings are numerous, they age, and they accumulate in the Poor bucket without representing much total deck area. Idaho's Poor share is 5.38% by count but only 3.76% by area, because the average Poor-rated structure in Idaho is about 70% the size of the typical one.

Canyon County inverts it. Its Poor share more than doubles when you switch from counting to measuring - 2.06% by count, 5.34% by area - because its six Poor-rated structures average about 1,024 square metres of deck each, against a county-wide average of 395. A count-based summary of Canyon County understates its Poor-rated deck area by a factor of roughly 2.6.

Two more figures fall out of the same table. Ada and Canyon together hold 593 bridges - 12.8% of Idaho's total - but only 9 of the state's 250 Poor-rated structures, or 3.6% of them. And 364 of those 593 bridges, 61.4%, sit in Fair.

Fair is the interesting bucket

Poor gets the attention because it sounds alarming. Fair is where the money is. A Fair classification means the governing component rated 5 or 6 - past new, not yet at the threshold where the classification changes. In a two-county inventory where 61.4% of structures sit there, the dominant condition question is not what is broken. It is what is halfway through its life, and in what order it will arrive.

That is exactly the shape of a commercial building portfolio. Very little of a well-run building is broken at any given moment. Most of it is somewhere in the middle of its service life, and the useful question is sequencing.

Why this matters for a building

The transferable lesson is not about bridges. It is that a condition inventory which counts deficiencies and a condition inventory which measures them produce different priority orders from identical observations.

A report that says a property has fourteen roofing deficiencies and three pavement deficiencies implies roofing is the bigger problem. Weight those same findings by affected area, by replacement cost, or by consequence of failure, and the order can reverse - one pavement deficiency across an entire truck court can outweigh fourteen localized roof observations. This is the same arithmetic that separates Canyon County's count-based picture from its area-based one.

It is also why the worst-component rule in the NBI is worth borrowing. A bridge with an excellent deck and a substructure rated 4 is classified Poor, because averaging would hide the thing that governs. Building systems behave the same way. A roof assembly is not in average condition; it is in the condition of its worst detail, and that is usually a flashing, a penetration, or a drain.

Three practical questions to put to any condition report, on any asset: Is the priority order based on counting findings or on weighting them? Is the classification driven by the governing component or by an average? And how much of this property is in the middle of its life rather than at either end - because that middle is what the capital plan is actually for.

Limits of this analysis

Canyon County's Poor-rated population is six structures. That is a small enough number that a single large structure moves the ratio substantially, and the finding should be read as a description of this dataset in this year rather than as a trend. Deck area is a proxy for scale, not for cost or complexity. County-level NBI tables aggregate structures across all owners and all functional classifications. And the whole analysis rests on the 2025 publication; earlier and later years will differ.

Stating those limits is part of the point. A derived number without its caveats is not analysis.

Sources

Primary dataset: Federal Highway Administration, National Bridge Inventory - Bridge Condition by County, 2025 (includes federal bridges). https://www.fhwa.dot.gov/bridge/nbi/no10/county25a.cfm

Classification definition: Federal Highway Administration, Bridge Condition definitions (NBI Items 58, 59, 60, 62). https://www.fhwa.dot.gov/bridge/britab.cfm

All percentages, averages, and size ratios in this article were computed by Northline Inspection Co. from the published counts and deck areas. No figure here is estimated, modelled, or carried over from another source.

Northline Inspection Co. performs commercial property inspections, Property Condition Assessments, and building maintenance inspections and cost planning throughout Boise, the Treasure Valley, and Southern Idaho. info@northlineinspection.com | (208) 254-1940