Predicting the 2026 Mat-Su Valley Salmon Runs — And Why Real-Time Counts Still Decide the Day

The Mat-Su Valley is road-system salmon country. For thousands of Southcentral anglers, the Deshka, Little Susitna, Fish Creek, and Jim Creek are the closest thing to a backyard fishery — which is exactly why the timing matters so much. These aren’t million-fish sonar rivers. They’re smaller, weir-counted systems where a run can build and fade inside a week, and where the difference between a great Saturday and a slow one is often just which Saturday you picked.

This post does two things. First, it takes the daily passage records for the Mat-Su’s main counting projects and turns them into concrete 2026 expectations — when each species’ run is likely to start, peak, and tail off, and how strong the magnitude looks against historical analogs. Second, it’s honest about the limits. Mat-Su king and silver runs have been weak in the recent regime, several of these counting records have real gaps, and 2026 is an even year — which changes everything for pinks. Where the data supports a real call, we make it. Where it doesn’t, we say so. That second half is where Salmon Finder earns its place on your phone.

How We Read the Numbers

For each species at each river, we computed four timing anchors per year of data: the day on which 5% of the run had passed (season start), 50% (midpoint), 85% (late-season threshold), and the consecutive 10-day window with the largest total passage. We also weight lifecycle analogs — 2022 (4-year), 2021 (5-year), 2023 (3-year), and 2020 (6-year) — over simple linear trend, because salmon return on a cycle, not a calendar. Pinks return on a strict 2-year cycle, so for 2026 (an even year) we lean hardest on the most recent even years: 2024, 2022, 2020, 2018.

We separate timing confidence from magnitude confidence. On several Mat-Su rivers the timing has more year-to-year spread than the big sonar systems do, so timing confidence here is more often “medium” than “high.” We flag it honestly.

Day-to-date convention: this analysis treats the spreadsheet’s day_1 as May 1, day_31 as May 31, day_32 as June 1, through day_123 = August 31. If the source file uses day_0 = May 1, every date below shifts one day later.

Two call-outs about this dataset. First, coverage is uneven. Several Mat-Su records have missing years inside their span — Fish Creek Chinook, for example, has only 11 of a possible 25 years — which lowers magnitude confidence wherever it appears. Second, the file also contains Coghill River (a Prince William Sound system) and Gulkana River (a Copper River drainage system). Neither is in the Mat-Su Valley, so we’ve kept them out of the Mat-Su forecast and summarized them separately at the end as adjacent reference rivers.

Executive Summary: The 2026 Mat-Su Outlook at a Glance

The honest headline: Mat-Su Chinook and coho have been running well below their long-term averages, and 2026 is unlikely to break that pattern. The recent-regime numbers are stark — Deshka Chinook has fallen from a pre-2021 average above 22,000 fish to a recent-3-year average under 3,000. This is a plan-conservatively, watch-closely year for kings and silvers.

Best overall opportunity: Fish Creek sockeye. It’s the steadiest run in the valley — a stable 26-year record averaging about 52,000 fish, with analogs clustered near the mean. Forecast: average run, medium confidence on both timing and magnitude. If you want a Mat-Su run you can actually plan around, this is it.

The even-year wildcard: Pinks. 2026 is an even year, and Mat-Su pinks are heavily even-cycle (the Deshka even-year average is roughly 14× its odd-year average). That makes pinks the species most worth showing up for in 2026 — but recent even years have run below the long-term even-year average, so we temper the magnitude call.

Most predictable timing: Jim Creek coho (50%-passage standard deviation of just 2.9 days) and Deshka Chinook (3.83 days). These are the two windows you can mark with the most confidence.

Least predictable timing: Little Susitna sockeye (standard deviation 16 days) and Little Susitna Chinook (11.5 days). The runs happen — but when they peak swings widely year to year, which is precisely where live counts beat a calendar.

Where we won’t forecast: Fish Creek Chinook (a once-real king run now counting in the low single digits and sparsely recorded), Jim Creek pink (essentially absent), and a couple of “Other”-species records too thin to model. We flag these as insufficient rather than invent a number.

Chinook Salmon

The Mat-Su king story is a decline story, and we’re not going to dress it up.

Deshka River. The flagship Mat-Su king run, with a full 26-year record. Long-term mean: about 19,400 fish. Recent-3-year mean: 2,959 — roughly 13% of the pre-2021 average. The trend is clearly down. Timing, by contrast, is one of the tightest in the valley (50%-passage standard deviation of 3.83 days), with the midpoint clustering in mid-June and the 5% start in late May to mid-June. Lifecycle analogs: 2022 returned 5,436, 2021 returned 18,583, 2023 returned 3,741. Weighted forecast for 2026: weak relative to long history, with medium timing confidence and medium magnitude confidence. Expect a mid-June core to whatever run shows.

Little Susitna River. 13-year record (2013–2025), full coverage. Mean: about 2,450 fish; recent-3-year mean: 623. Also declining. Timing here is genuinely unpredictable — the 50%-passage standard deviation is 11.5 days, meaning the midpoint has swung from mid-June to early August across the record. Forecast: average relative to its own (modest) recent regime, but with low timing confidence. This is a river where you should not trust a calendar date; watch the counts.

Fish Creek. Once a counted king system, now returning low single digits (recent years: 1–2 fish) with a sparse, gap-filled record. We do not issue a meaningful 2026 Chinook forecast for Fish Creek — there isn’t a run left to forecast from this data.

Practical implication. Mat-Su kings in 2026 are a plan-around-the-restrictions, watch-the-counts proposition. Be aware that low runs frequently trigger inseason management actions. The historical timing windows (mid-June core on the Deshka) are your planning anchor; the live counts are your go/no-go signal.

Sockeye Salmon

Sockeye is where the Mat-Su offers its most plannable 2026 opportunity.

Fish Creek. The valley’s sockeye workhorse — a complete 26-year record, stable trend, mean of about 51,800 fish. Recent-5-year mean (41,200) sits modestly below the long-term mean but well within normal range. Timing is reasonably tight (50%-passage standard deviation of 3.86 days), with the midpoint in late July and the 5% start in mid-July. Analogs (2022: 58,351, 2023: 44,711, 2021: 22,271) bracket the mean. Forecast: average run for 2026 with medium timing and magnitude confidence. This is the Mat-Su’s most reliable sockeye call.

Little Susitna. 13-year record, stable, but small — mean of about 1,190 fish. The trouble is timing: a 50%-passage standard deviation of 16 days makes this the least predictable run window in the valley. Forecast: average magnitude, low timing confidence. The run is real; pinning the week from history alone is not possible.

Jim Creek. 8-year record (with gaps), declining, mean about 3,050 sockeye, recent-3-year mean about 2,020. Timing is fairly tight (standard deviation 3.39 days), midpoint in early August. Forecast: weak magnitude with medium timing confidence and low magnitude confidence (limited and gappy record).

Deshka. Sockeye are a minor, sparsely recorded presence on the Deshka (long-term mean around 60 fish in the pre-2021 record, with one anomalous recent spike). We flag this as a low-confidence minor run — not a planning target.

Practical implication. For Mat-Su sockeye, Fish Creek is the call you can build a trip around. Everywhere else, the historical record gives you a rough window and the live counts give you the actual timing.

Coho Salmon

Silvers are the late-summer Mat-Su staple, and like kings, they’ve softened in the recent regime.

Little Susitna. The valley’s premier road-accessible silver fishery, with a full 26-year record. Mean: about 8,800 fish; recent-3-year mean: 3,006 — a clear decline. Timing midpoint runs mid-to-late August (50%-passage standard deviation 9.6 days, moderate). Analogs (2021: 10,923, 2020: 10,765, 2022: 3,162, 2023: 3,726) are split between strong older years and weak recent ones. Forecast: below average for 2026 with medium magnitude confidence. The August window holds; the volume has been thinner.

Deshka. 26-year record, mean about 17,900, recent-3-year mean 2,109 — one of the steeper coho declines in the dataset. Midpoint in mid-August (standard deviation 8 days). Forecast: weak, medium magnitude confidence.

Jim Creek. 8-year record, mean about 2,560, declining. The bright spot: timing is the most predictable in the entire dataset (50%-passage standard deviation of 2.9 days), with the midpoint locked around August 21. Forecast: weak magnitude but high timing confidence. If you fish Jim Creek silvers, the late-August window is close to bankable — the only question is how many show.

Fish Creek. 22-year record (some gaps), declining, mean about 3,570, recent-5-year mean about 1,019. Forecast: weak, low confidence on both axes.

Practical implication. Mat-Su silvers are a “right week, modest numbers” story for 2026. The timing windows — mid-to-late August across all four rivers — are the planning anchor. Because the magnitude has been thin and variable, real-time counts are the difference between hitting a push and fishing a lull.

Pink Salmon

Here’s the species that makes 2026 interesting: pinks run on a strict 2-year cycle, even years dominate in the Mat-Su, and 2026 is even.

Deshka. 26-year record. Even-year mean: about 249,000 fish. Odd-year mean: about 18,000 — a 14× even/odd ratio. That’s the loudest cyclic signal in the valley. The catch: recent even years have run below the long-term even-year average (the recent even analogs average closer to 81,000), so while 2026 should be a meaningful pink year by Mat-Su standards, we forecast it as moderate for an even year, not a blowout — call it weak-to-average against the long-term even baseline, with medium timing confidence. Timing is tight (standard deviation 3.47 days): expect a late-July core, with the 5% start around July 14–28 and the peak 10-day window roughly July 15–31.

Little Susitna. 13-year record. Even-year mean about 11,600, odd-year about 5,000 (a 2.3× ratio — a real but milder even cycle). Timing is tight (standard deviation 4.6 days), midpoint late July. Forecast: average even-year run with medium confidence on both axes. The peak window lands roughly July 14–Aug 2.

Fish Creek. 17-year record with gaps; even/odd ratio about 3×, but recent even years have been thin (the 2022 even year recorded essentially zero). Forecast: weak, low confidence.

Jim Creek. Pink data is sparse (3 years) and essentially zero. No meaningful forecast.

Practical implication. If you fish Mat-Su pinks, 2026 is the on-cycle year — particularly on the Deshka and Little Susitna, where the late-July window is the target. Build around late July, and let Salmon Finder confirm the wave is actually breaking, because even-year magnitude has been sliding and the peak compresses fast.

A Note on “Other”

The dataset includes an “Other” species category at several Mat-Su sites (likely chum and miscellaneous). The most substantial is Little Susitna “Other” — a roughly 26,600-fish average run with tight timing (standard deviation 3.77 days) peaking in late July. It’s a secondary opportunity rather than a target species, and we mention it mainly so the numbers in the app make sense.

How the Four Mat-Su Rivers Compare

Most plannable overall: Fish Creek — the steady sockeye run anchors it, with the most reliable magnitude call in the valley.

Best 2026 even-year pink target: Deshka and Little Susitna, late July.

Most predictable timing: Jim Creek coho (late August, standard deviation 2.9 days) and Deshka Chinook (mid-June, 3.83 days).

Least predictable timing — watch live: Little Susitna sockeye (standard deviation 16 days) and Little Susitna Chinook (11.5 days). These two need real-time tracking more than any others on the list.

Earlier season: Deshka Chinook (late-May to mid-June front edge) and Little Susitna kings/sockeye when they run early.

Later season: coho across all four rivers (mid-to-late August) and the late-July pink window.

2026 Mat-Su Prediction Table

Why Predictions Aren’t Enough — and Where Salmon Finder Closes the Gap

Everything above is the playbook. It gives you the seasonal shape, the windows, and the analog-weighted run sizes where the data supports them — and an honest shrug where it doesn’t. None of it replaces what’s happening in the water.

The Mat-Su is, in fact, the strongest case for real-time tracking of all three regions we’ve looked at. The runs are smaller, so a single push is a bigger share of the week. The timing is looser — Little Susitna sockeye can peak three weeks apart from one year to the next. The magnitude has been declining and variable across kings and silvers, which means inseason management actions can open or restrict a fishery on short notice. And the even-year pink signal, while real, has been sliding, so 2026’s pink magnitude is genuinely a watch-it-build question.

That’s the gap Salmon Finder closes. The app pulls the same weir and counting-tower numbers these forecasts are built on — as they’re posted, not at season’s end. It plots every Mat-Su counting project on one map so you can see the Deshka kings, the Fish Creek sockeye, and the Little Susitna silvers at a glance, and it pushes notifications when a daily count crosses a threshold, when a run starts to build, or when a number spikes.

Predictions tell you when to get ready. Real-time counts tell you when to go. The blog gives you the seasonal playbook. The app tells you what’s happening today. Don’t just guess the run — watch it build.

Adjacent Reference Rivers (Not Mat-Su)

For completeness, two rivers in the source file sit outside the Mat-Su Valley. We summarize them here but keep them out of the Mat-Su forecast.

Coghill River (Prince William Sound). A strong, increasing sockeye system — 26-year record, recent-3-year mean about 78,800 (well above the long-term mean of 47,600), midpoint around July 7. Forecast: strong sockeye run for 2026, medium confidence. Its pinks are odd-cycle dominant, so 2026 (even) projects weak.

Gulkana River (Copper River drainage). A stable interior system. Chinook: 24-year record, average forecast around 4,000 fish, midpoint around July 9. Sockeye: below-average forecast around 14,400, with very loose timing (standard deviation 13.8 days). Both are flagged as separate-drainage references.

Conclusion

If you fish the Mat-Su in 2026, here’s the short version. Treat Fish Creek sockeye as your most plannable run. Treat the Deshka and Little Susitna late-July pink window as the on-cycle opportunity. Treat kings and silvers as plan-conservatively, watch-closely fisheries where the timing windows hold but the numbers have been thin. And treat Little Susitna’s loose timing as a standing reminder that the calendar only gets you so far.

Then download Salmon Finder — free on the Apple App Store and Google Play — open the map, and turn on notifications for the rivers and species you actually fish.

Read the prediction. Open the app. Follow the map. Turn on the alerts. Fish the windows.