FlyerIntel Research

Weather Share of Delay, Winter and Summer

Weather accounts for a larger share of attributed delay minutes in winter at 39 of 62 airports.

Published October 18, 2026 · Source period July 2023 – July 2026

Summary

39 of 62 large airports put a bigger share of their delay minutes down to weather in December to February than in June to August; 23 put a bigger share down in summer. The median difference is +1.3 points. Pooled across the airports, weather is 6.6% of attributed delay minutes in winter and 5.7% in summer.

Key findings

  • ANC has the largest winter lead: weather is 13.5% of its attributed delay minutes in December to February against 1.8% in June to August, 11.6 points more in winter.
  • MCO has the largest summer lead: 10.5% in summer against 2.2% in winter, 8.3 points more in summer.
  • The median airport’s winter share is above its summer share by 1.3 points.
  • Pooled over the 62 airports, weather minutes are 6.6% of attributed delay minutes in winter and 5.7% in summer.

Which season blames the weather more

Delay minutes in the federal data come with a stated cause for every flight that arrived 15 minutes or more late. This report takes the share of those minutes the data files under weather, at each of 62 large airports, and compares December to February with June to August. It is a share of the delay that is attributed, not of all flights, so an airport with little delay in a season can show a high share on few minutes.

Largest winter lead, percentage points

ANC · Anchorage11.6 pts
SLC · Salt Lake City8.7 pts
PDX · Portland7.6 pts
SEA · Seattle6.9 pts
SBA · Santa Barbara6.5 pts
DTW · Detroit6.0 pts
MSP · Minneapolis5.7 pts
BUF · Buffalo5.5 pts
DFW · Dallas–Fort Worth5.2 pts
ORD · Chicago5.1 pts

Largest summer lead, percentage points

MCO · Orlando8.3 pts
MIA · Miami5.9 pts
FLL · Fort Lauderdale4.9 pts
IAD · Washington4.6 pts
EWR · Newark4.5 pts
JAX · Jacksonville4.3 pts
TPA · Tampa3.7 pts
RSW · Fort Myers3.6 pts
RDU · Raleigh–Durham3.3 pts
BWI · Baltimore2.9 pts

Weather share of attributed delay minutes by airport

Weather share of attributed delay minutes by airport
#AirportCityWinter shareSummer shareDifference, pointsCompleted flights, winterCompleted flights, summer
1ANCAnchorage13.5%1.8%+11.610,40022,069
2SLCSalt Lake City13.0%4.4%+8.782,68398,167
3PDXPortland9.0%1.4%+7.639,59955,585
4SEASeattle9.8%2.9%+6.9104,562157,088
5SBASanta Barbara10.3%3.9%+6.55,2056,974
6DTWDetroit12.2%6.1%+6.084,450108,179
7MSPMinneapolis12.4%6.7%+5.778,576108,287
8BUFBuffalo8.1%2.6%+5.513,05918,403
9DFWDallas–Fort Worth10.7%5.5%+5.2209,188270,907
10ORDChicago12.7%7.7%+5.1198,513267,462
11SANSan Diego7.3%2.6%+4.764,63584,448
12SFOSan Francisco6.2%1.9%+4.395,232122,136
13SMFSacramento5.8%1.5%+4.337,25949,428
14BOSBoston5.8%2.6%+3.292,196122,818
15INDIndianapolis7.6%4.6%+3.031,69840,131
16SJCSan Jose4.2%1.4%+2.832,36342,161
17LAXLos Angeles4.5%1.7%+2.8133,450168,390
18BNANashville6.8%4.2%+2.667,46589,353
19CLECleveland6.1%3.7%+2.426,55335,367
20ONTOntario3.7%1.4%+2.317,10422,777
21BURBurbank4.6%2.4%+2.220,80226,110
22MKEMilwaukee7.1%5.0%+2.118,50424,648
23DENDenver6.0%4.0%+2.0213,948274,948
24TUSTucson10.0%8.0%+2.015,16813,696
25ATLAtlanta8.4%6.4%+2.0223,665280,229
26OMAOmaha9.3%7.6%+1.816,84421,058
27OAKOakland2.6%0.9%+1.726,53732,919
28HNLHonolulu3.0%1.3%+1.744,34351,418
29DALDallas4.1%2.7%+1.551,06661,518
30SATSan Antonio6.1%4.7%+1.428,07334,972
31CMHColumbus5.3%3.9%+1.428,89037,085
32PITPittsburgh6.4%5.1%+1.328,06637,637
33PHXPhoenix3.6%2.3%+1.3143,296150,473
34JFKNew York5.0%3.9%+1.177,33894,676
35MYRMyrtle Beach7.6%6.5%+1.16,81116,234
36SNASanta Ana4.7%3.7%+0.931,33837,677
37ABQAlbuquerque4.7%3.8%+0.816,28721,399
38STLSt. Louis4.3%3.8%+0.642,85655,958
39MEMMemphis7.9%7.4%+0.414,71520,940
40SJUSan Juan1.4%1.4%0.027,19931,523
41MCIKansas City5.6%5.7%-0.132,30943,838
42LASLas Vegas3.2%3.5%-0.3129,174151,508
43HOUHouston6.0%6.5%-0.538,44647,801
44AUSAustin3.6%4.2%-0.658,87376,349
45PHLPhiladelphia6.1%7.0%-0.960,33489,207
46MDWChicago3.8%4.9%-1.147,85672,848
47IAHHouston6.8%8.0%-1.280,48198,081
48CHSCharleston2.9%4.5%-1.615,59721,842
49MSYNew Orleans3.7%5.7%-2.036,45538,610
50LGANew York4.1%6.6%-2.5100,511118,347
51DCAWashington5.5%8.0%-2.597,233113,097
52CLTCharlotte7.1%9.9%-2.8136,983165,833
53BWIBaltimore3.6%6.5%-2.964,24184,247
54RDURaleigh–Durham2.7%5.9%-3.338,42349,781
55RSWFort Myers2.6%6.2%-3.631,64020,605
56TPATampa2.1%5.8%-3.758,59460,909
57JAXJacksonville3.3%7.7%-4.320,24924,275
58EWRNewark3.8%8.4%-4.587,792100,428
59IADWashington7.6%12.2%-4.636,03048,130
60FLLFort Lauderdale1.9%6.7%-4.972,49967,069
61MIAMiami2.8%8.7%-5.987,33585,144
62MCOOrlando2.2%10.5%-8.3121,457128,128

Sorted from the largest winter lead to the largest summer lead. Weather share is weather delay minutes over all attributed delay minutes, the sum of the five cause categories the data reports, for flights that left the airport in the months named, added up before dividing. Difference is the winter share less the summer share.

What the share leaves out

The data’s weather category covers extreme weather only. Ordinary weather that slows the air traffic system is filed under that system, so every share here is lower than the weather’s true part, and by how much differs by airport and season. A share can fall because another cause grew, so a lower weather share in one season does not by itself mean fewer weather minutes. The table gives shares only; it does not give minutes. Airports are included when they have flights in all three months of both seasons and at least 5,000 completed flights in each season; 2 did not and are left out.

Pages for what this report covers

Methodology

Calculated by FlyerIntel from 21,708,321 individual flight records in the US Department of Transportation Bureau of Transportation Statistics Reporting Carrier On-Time Performance dataset, covering July 2023 through July 2026. Following BTS methodology, a flight counts as delayed when it arrives 15 or more minutes after its scheduled arrival time. Cancelled and diverted flights have no arrival time and are excluded from on-time rates rather than counted as late; cancellation rates are reported separately. Full definitions are published at /methodology/. Each calendar month is pooled across every year in the period, and a season adds its three months. The share divides weather delay minutes by the sum of the five cause columns on the same rows; months where any cause column is empty are left out. Flights are counted once, at the airport they leave. The pooled share adds minutes across the listed airports before dividing, so large airports weigh more.

Limitations

The source dataset covers scheduled domestic flights operated by carriers above the DOT reporting threshold; smaller carriers and international flights are not included. Flights sold under a mainline brand but flown by a regional carrier are attributed to the operating carrier, which is how the DOT records them. Past performance describes what has happened and is not a prediction about any individual flight.

Sources

  • FlyerIntel analysis — FlyerIntel · source
  • Reporting Carrier On-Time Performance — US Department of Transportation, Bureau of Transportation Statistics · source

Full FlyerIntel methodology · Report a problem with this analysis