Skip to main content
PERM Tracker

Data

H-1B lottery odds, year by year

How many H-1B cap registrations USCIS received and selected each year, from its own table, and how the odds change now that the draw is weighted by wage.

FY2026 cap

34.9%

of eligible registrations selected

120,141 of 343,981. USCIS counted about 339,000 unique beneficiaries, with 1.01 registrations each on average, so one person's chance was close to this rate. The FY2027 lottery was the first weighted by wage; USCIS hasn't added it to its table (last revised 2026-09-21).

Eligible and selected H-1B registrations, FY2021 to FY2026FY2021: 269,424 eligible registrations FY2021: 124,415 selected 46.2% FY2021 FY2022: 301,447 eligible registrations FY2022: 131,924 selected 43.8% FY2022 FY2023: 474,421 eligible registrations FY2023: 127,600 selected 26.9% FY2023 FY2024: 758,994 eligible registrations FY2024: 188,400 selected 24.8% FY2024 FY2025: 470,342 eligible registrations FY2025: 135,137 selected 28.7% FY2025 FY2026: 343,981 eligible registrations FY2026: 120,141 selected 34.9% FY2026

Outline: eligible registrations. Filled: selected. The share above each bar is selected over eligible. From FY2025 USCIS drew people rather than registrations, which is why the FY2024 pile of repeat registrations disappears.

From FY2027, the odds depend on the wage

A registration now enters the draw once at OEWS wage level I and up to four times at level IV. DHS estimated what that does to one beneficiary's chance, in the final rule, before the first weighted lottery ran:

  • Level I 15.3%
  • Level II 30.6%
  • Level III 45.9%
  • Level IV 61.2%

Dashed line: 29.59%, DHS's figure for every beneficiary under the random draw. DHS's estimate, 90 FR 60864 (Dec. 29, 2025), estimate at 60947-60948, assumes employers keep their current wages and pools the regular cap with the master's cap; DHS notes it may understate selections at higher levels. Find the OEWS wage level for a job.

USCIS's table

Cap year Registrations Eligible One per person Several per person Selected Selected share
FY2026 358,737 343,981 336,153 7,828 120,141 34.9%
FY2025 479,953 470,342 423,028 47,314 135,137 28.7%
FY2024 780,884 758,994 350,103 408,891 188,400 24.8%
FY2023 483,927 474,421 309,241 165,180 127,600 26.9%
FY2022 308,613 301,447 211,304 90,143 131,924 43.8%
FY2021 274,237 269,424 241,299 28,125 124,415 46.2%

From USCIS's H-1B electronic registration page, last revised 2026-09-21, read 2026-09-26. Eligible leaves out duplicates, registrations deleted before the period closed, invalid passports and failed payments. The selected count is USCIS's total for the year.

What this can't tell you
  • Your own odds. They depend on your wage level and on how many registrations came in at each level, which USCIS hasn't published for FY2027. The percentages above are DHS's estimate, not a count.
  • How the FY2027 lottery actually turned out. USCIS's table stops at FY2026.
  • How many of a year's selections came in a second round. The table gives the year's total and doesn't split it.
  • Anything about cap-exempt jobs. Employers such as universities and their affiliated nonprofits and research organizations file outside the lottery under INA 214(g)(5).
  • Whether a selected registration becomes an approved petition. Selection only lets the employer file.

Common questions

What were the H-1B lottery odds for FY2026?
USCIS selected 120,141 of 343,981 eligible registrations, about 35%. It counted roughly 339,000 unique beneficiaries, and almost all had a single registration (1.01 on average), so a beneficiary's chance was close to the registration rate.
How does the weighted lottery work?
From the FY2027 cap, a registration goes into the draw once for OEWS wage level I, twice for level II, three times for level III and four times for level IV. The level is the highest one the offered wage meets for the job's occupation and area. A beneficiary is still selected only once.
What are my odds at my wage level?
DHS estimated them in the final rule: about 15% at level I, 31% at level II, 46% at level III and 61% at level IV, against about 30% for everyone under the old random draw. That's DHS's estimate from past petitions, assuming employers keep their current wages, not a count of an actual lottery.
Why did the odds jump in FY2025?
The lottery changed from drawing registrations to drawing people. Before FY2025, one person with several registrations had several chances. From FY2025 each unique beneficiary is entered once, and registrations for the same person fell from 408,891 in FY2024 to 47,314 in FY2025.

The rules behind these numbers

Other calculators