Something changed at the border in 2026, and only the car lanes broke
By Alan Scott · Who makes this · Tijuana, Baja California ·
It isn't in your head, and it isn't just San Ysidro. From May through August, all three crossings ran far above the same months last year — and Tecate doubled.
But the finding that surprised us is further down: Otay Mesa's freight lanes got faster over the same period. That rules out the easy explanation.
Each month against the same month last year
🔴 We compared each month to the same month of 2025, not to the month before it. It's the only valid comparison, because the border always slows down going into summer — we measured it, it's about 32 minutes of seasonality — so comparing July to March of the same year measures the season, not the change.
Median, car (General) lane, northbound:
| Month | San Ysidro 2025 → 2026 | Otay Mesa 2025 → 2026 | Tecate 2025 → 2026 |
|---|---|---|---|
| January | 67 → 59 · −12% | 51 → 45 · −12% | 36 → 53 · +45% |
| February | 55 → 80 · +46% | 50 → 62 · +24% | 39 → 57 · +45% |
| March (1–26) | 65 → 71 · +10% | 54 → 54 · 0% | 44 → 58 · +32% |
| April | ⚠️ no data | ⚠️ no data | ⚠️ no data |
| May | 80 → 96 · +20% | 65 → 90 · +39% | 49 → 106 · +114% |
| June | 78 → 121 · +56% | 51 → 88 · +74% | 45 → 116 · +160% |
| July | 89 → 120 · +34% | 67 → 95 · +41% | 49 → 133 · +169% |
| August (1–9) | 81 → 131 · +62% | 64 → 107 · +67% | 53 → 123 · +132% |
January and March were normal or better. From May on, all three are up — and the gap widens every month.
And Tecate is its own case: it was already running 45% above 2025 back in January, before whatever else happened. At Tecate there are two things going on, not one.
⚠️ And 20 days are missing from the public record
We need to say this before drawing any conclusion:
🔴 SANDAG's open data contains no records at all from March 27 to April 15, 2026. Twenty consecutive days, across all three crossings and every lane type at once.
All three ports and every lane going missing simultaneously means this is a source failure, not something that happened at the border. In 2025, not one day of that same window is missing.
Why we're telling you instead of papering over it: anyone can download that data and count the days. And because it forces us to admit something inconvenient: we cannot tell you the exact day the border changed. The change is clear from May onward, but the onset falls inside the gap or right around it, and there's nothing there to measure with.
If you ever read that the change started on a specific day in April, ask where April came from.
🔥 And here's what nobody has published
If a crossing were simply overwhelmed — more people, fewer booths open — every lane would get worse. We split each lane out and compared January–July against the same period in 2025:
| Lane | San Ysidro | Otay Mesa | Tecate |
|---|---|---|---|
| Car (General) | 74 → 93 · +27% | 56 → 75 · +34% | 44 → 80 · +84% |
| SENTRI | 13.3 → 14.3 · +8% | 13.4 → 17.4 · +30% | ❌ doesn't exist |
| Pedestrian | 15.4 → 16.6 · +8% | 16.7 → 20.2 · +21% | 7.0 → 8.6 · +23% |
| Pedestrian Ready | 3.0 → 3.1 · +3% | ❌ doesn't exist | ❌ doesn't exist |
| Commercial | ❌ doesn't exist | 21.8 → 19.4 · −11% | 13.8 → 15.0 · +9% |
| Commercial FAST | ❌ doesn't exist | 14.4 → 11.7 · −19% | ❌ doesn't exist |
Read those last two rows again. At Otay Mesa, the region's only crossing with freight lanes, trucks are moving faster than last year. FAST improved 19%.
Same crossing, same year: freight improved and cars got 34% worse.
That rules out Otay being "overwhelmed." If it were general volume or short staffing, freight would fall as hard as cars or harder. The opposite happened.
And the cleanest contrast is at San Ysidro: the Pedestrian Ready Lane went up 3% — which is to say it didn't move — while the car lane right next to it went up 27%.
What this data does NOT tell you
🔴 It does not tell you why. And we're not going to invent a reason.
It could be how booths are assigned across lanes, construction, a change in inspection procedure, or several things at once. We have no way to distinguish between those, and writing one down as if we knew would be exactly what this site exists not to do.
The only thing we can assert is what we ruled out: at Otay it isn't general saturation, because freight improved.
And that isn't an explanation either. It's one hypothesis fewer.
⚠️ August is 9 days — compared against the same 9 days of 2025, which is the fair comparison, but still 9 days.
And we don't know if it's permanent. Four months is four months. If September goes back to normal, we'll say so here in the same size type.
🔴 How we got here, and what we corrected on the way
The first version of this article said the border "jumped in April." That was wrong for two reasons, and both are ours:
- We compared 2026 months against each other instead of against the same month of 2025. That measures the summer season, not the change.
- The "April jump" came from 15 days of April, because the other 15 sit inside the source's gap. We hadn't counted them.
We redid it month-against-same-month and excluded the gap from every year. The change is still there, it's bigger than we said, and now it's measured correctly — but the start date falls apart, which is why this article no longer says "April."
It's on our public list of mistakes
What to do with this
- Prior-year averages are no longer useful for planning. If you're going by what the border was like in 2025, you're short — by 20% to 130%, depending on where you cross
- 🔴 Tecate is not the shortcut anymore. It's still the fastest of the three, but by 13 minutes in July instead of the 40 it used to be
- Otay is still better than San Ysidro, and in July by 25 minutes
- 🟢 If you have Ready Lane and can walk across at San Ysidro, nothing happened to you: 3.1 minutes, same as last year
- And this is the best argument for checking the current number instead of trusting a pattern: the patterns moved this year
You might also want:
- San Ysidro border wait
- Otay Mesa border wait
- Tecate border wait
- The border is running worse than last year
Sources
Data: Average Daily Border Waiting Time, SANDAG Open Data Portal, from U.S. Customs and Border Protection records. Northbound. Period: January 29, 2023 to August 9, 2026. Method: medians, never averages. Month-against-the-same-month comparison, so seasonality doesn't get counted as change. 🔴 The 20 days from March 27 to April 15 are excluded from EVERY year, not just 2026, so the comparison is like against like. March is trimmed to the 26th in both years for the same reason. Retrieved: August 12, 2026.
We are not affiliated with CBP or SANDAG.
CHARTS · build spec
Same two charts as the Spanish version — see 6_CONTENIDO/articulos/09_la_linea_cambio_en_2026.md. 🔴 Do not re-derive the values.
Label translations only, plus the two rules that stay: 🔴 April is drawn as a visible GAP in all three lines, never interpolated, labeled "the source published nothing." The gap is part of the argument. 🔴 August keeps reduced opacity with the footnote "August is 9 days against the same 9 days."