Where to surf in July and early August: what 75,932 modelled spot-days suggest
The archive contains 75,932 model-score rows across 1,097 breaks. This is an exploratory snapshot of one late-June-to-July window and one late-June-to-mid-August window—not a climatology, universal destination ranking, or forecast validation.
The honest verdict
Was two years enough?
Yes—for describing this sample
Counts such as 11.5% classified Good-or-better and 37 Epic spot-days are exact summaries of the archived rows in the two stated windows.
Suggestive—for broad seasonal patterns
The southern-hemisphere advantage appeared in both matched 28-day comparisons, making the direction more credible than a one-window anomaly.
No—for long-run destination odds
Two asymmetric, partial windows cannot establish climatological probability, stable fine-grained rankings, or what will happen in a future July or August.
What the sample contained
75,932
archived score rows
11.5%
Good or better
2.5%
Firing or Epic
37
Epic spot-days
All score categories
The broad signal repeated
Across the combined sample, southern-hemisphere breaks received Good-or-better classifications in 17.3% of archived rows versus 7% in the north. More importantly, the direction was the same in each matched 28-day window:
| Archive window | Archived rows | Southern Good+ | Northern Good+ | All Good+ |
|---|---|---|---|---|
| 29 Jun 2024–26 Jul 2024 | 30,660 | 17.2% | 5.3% | 10.5% |
| 30 Jun 2025–27 Jul 2025 | 30,685 | 18.8% | 7.4% | 12.4% |
This is a descriptive comparison, not a causal estimate. Breaks are clustered by coastline, model-score rows are not independent, and the analysis does not adjust for spot mix between hemispheres.
Highest Good-or-better frequency
Pooled rankings include only the 1,015 breaks with at least 60 of the 75 possible archived dates. They rank this sample—not the world's best waves.
| Break | Country | Archived dates | Good+ | Firing+ |
|---|---|---|---|---|
| 1Cimaja | Indonesia | 67 | 91% | 37.3% |
| 2Nosara | Costa Rica | 66 | 89.4% | 22.7% |
| 3Jimmy's Right | Indonesia | 75 | 89.3% | 53.3% |
| 4Carrizalillo | Mexico | 69 | 88.4% | 43.5% |
| 5Honey Smacks | Indonesia | 74 | 86.5% | 43.2% |
| 6Camana | Peru | 72 | 86.1% | 25% |
| 7El Transito | Nicaragua | 74 | 85.1% | 39.2% |
| 8Asuchillo | Nicaragua | 66 | 83.3% | 56.1% |
| 9Poneloya | Nicaragua | 73 | 82.2% | 37% |
| 10El Brazo | Chile | 75 | 81.3% | 12% |
| 11Playa Chinchorro | Chile | 75 | 81.3% | 16% |
| 12Lacerations | Indonesia | 69 | 81.2% | 50.7% |
| 13Balangan | Indonesia | 73 | 80.8% | 49.3% |
| 14Bingin | Indonesia | 67 | 79.1% | 40.3% |
| 15Las Gatas (Zihuatanejo) | Mexico | 69 | 78.3% | 18.8% |
Country-level results
Shown for countries with at least five ranking-eligible breaks. The percentage is weighted by archived score rows, so countries with more available dates contribute more rows.
Costa Rica
15 breaks · 1,046 archive rows
36.5%
Indonesia
79 breaks · 5,549 archive rows
34.2%
Nicaragua
26 breaks · 1,829 archive rows
31.8%
Chile
49 breaks · 3,452 archive rows
30.7%
New Zealand
17 breaks · 1,197 archive rows
25.9%
Angola
9 breaks · 628 archive rows
22.5%
El Salvador
7 breaks · 497 archive rows
21.1%
Peru
54 breaks · 3,799 archive rows
20.9%
Mexico
59 breaks · 4,148 archive rows
18.7%
Panama
11 breaks · 771 archive rows
17.3%
Australia
95 breaks · 6,666 archive rows
13.2%
Fiji
9 breaks · 642 archive rows
10.7%
Limits you should carry with every citation
- Partial, asymmetric windows. The 2024 window covers 28 days and no August; 2025 covers 47 days through 15 August. Neither is a long-run July–August record.
- Modelled inputs, not eyewitness reports. Historical API values cannot observe crowds, access, sandbars, water quality, or every local effect.
- Model-on-model result. Strike Score evaluates modelled inputs against stored spot parameters. It is not independent validation against surfer observations.
- Incomplete, unbalanced panel. Only 108 breaks have all 75 possible dates. The archive is missing 6,343 of 82,275 theoretical rows, and coverage falls sharply late in the 2025 window.
- Historical pipeline limitations. These stored rows have no tide adjustment, used the older direction averaging and period-selection logic, and did not preserve a pinned upstream model identifier.
- Spatial dependence. Nearby breaks often share the same swell and weather system, so 75,932 rows are not 75,932 independent experiments.
- No future guarantee. A high sample frequency does not mean a break will be good on any particular future date. Use a fresh forecast before booking.
Method and reproducibility
- 1
Source rows
Historical marine and weather values accessed through Open-Meteo. Rows carry an internal source='era5' label, but the upstream model parameter was not pinned or stored, so this report does not claim pure ERA5 provenance.
- 2
Historical daily aggregation
Hourly inputs were aggregated over each break’s 06:00–18:00 local window. This older path used arithmetic direction averaging and daily-max period selection; both are disclosed limitations.
- 3
Spot-specific scoring
Each stored day was evaluated against the break’s swell direction, size, period and wind parameters. The archived study rows contain no tide-series adjustment.
- 4
Frequency, not mean score
Results use thresholds (Good ≥65; Firing ≥80; Epic ≥95), avoiding distortion from the score’s floor at zero.
- 5
Ranking threshold
A break needed at least 60 of 75 possible archived dates to enter spot and country rankings.
- 6
Public aggregates
The downloadable CSVs include every archived break, its available-date count, eligibility flag, category counts, rates, and canonical URL.
Download and cite the data
The aggregate files are free to inspect. When republishing results, link to this report, name Strike Mission, preserve the archive date windows, and include the limitations relevant to your claim. Use is governed by the API terms.
Spot-level aggregates
1,097 rows · CSV
Country-level aggregates
All represented countries · CSV
Daily coverage table
75 dates · CSV
Methodology + checksums
Versioned manifest · JSON
Suggested citation: Cacciotti, Joel. “Where to Surf in July and Early August: What 75,932 Modelled Spot-Days Suggest.” Strike Mission, 27 July 2026, dataset v1.0.0.
Questions, corrections, or press use?
See the source notes or contact the data desk.