H1B Weighted Lottery Explained: New Wage-Based Selection Rules (2026)

The H-1B cap lottery is no longer random. Since February 27, 2026, USCIS selects cap-subject registrations through a weighted selection process that gives higher-paid positions more entries in the pool. The change is the biggest structural shift in the H-1B program since the electronic registration system launched in 2020, and it already governed one full cap season: FY2027, registered in March 2026.

This guide covers the full rule: how the weighting works, the odds DHS estimates for each wage level, what happened in the first weighted lottery, who wins and loses, and what could still change in court. If you want the short version for your own situation, check your wage level → and then run your estimated odds →.

The rule in one table

Item	Detail
Rule name	"Weighted Selection Process for Registrants and Petitioners Seeking To File Cap-Subject H-1B Petitions" (DHS final rule)
Proposed rule	September 24, 2025 (90 Fed. Reg. 45,986); roughly 17,000 public comments received
Final rule published	Federal Register, December 29, 2025 (90 Fed. Reg. 60,864)
Effective date	February 27, 2026
First applied	FY2027 cap registration period, March 4–19, 2026
Core mechanism	OEWS wage level I/II/III/IV = 1/2/3/4 entries in the selection pool
Annual cap	Unchanged: 65,000 regular + 20,000 U.S. advanced degree exemption
Finalized changes vs. proposal	None — DHS adopted the proposal without changes

Timeline: from proposal to the first weighted lottery

• September 24, 2025 — DHS publishes the proposed rule in the Federal Register. The comment period draws about 17,000 comments.
• December 23, 2025 — DHS announces the final rule. It makes no changes from the proposal.
• December 29, 2025 — Final rule published in the Federal Register (90 Fed. Reg. 60,864).
• February 27, 2026 — Rule takes effect.
• March 4–19, 2026 — FY2027 registration window (noon ET to noon ET). Registration fee: $215 per beneficiary. Employers must now declare the SOC code, work location(s), and OEWS wage level at registration.
• March 31, 2026 — USCIS announces it completed the first weighted selection and notified registrants through online accounts.
• April 1 – June 30, 2026 — 90-day petition filing window for selected registrations. Approved workers can start October 1, 2026.
• July 2026 — USCIS confirms the FY2027 cap is full and there will be no second lottery.

Two adjacent policy moves matter for context. A separate presidential proclamation (September 2025) imposed a $100,000 supplemental fee on certain H-1B petitions for beneficiaries outside the United States. And on March 26–27, 2026, DOL published a proposed rule that would raise the prevailing wage percentiles behind every wage level — that second wave is covered in our DOL wage percentile rule tracker.

How the weighted selection actually works

Under the old system (FY2021–FY2026), every eligible registration had an identical chance. If USCIS received more registrations than needed, it ran a random draw, and each unique beneficiary was counted once regardless of salary.

The final rule keeps the beneficiary-centric design — one person, one cap number, no matter how many employers register them — but layers a weight on top:

OEWS wage level	Typical profile	Entries in selection pool
Level I	Entry level	1
Level II	Qualified	2
Level III	Experienced	3
Level IV	Fully competent	4

The wage level is not chosen by the employer at will. It comes from comparing the offered wage against DOL prevailing wage thresholds for the job's SOC code and work location — the same four-level OEWS structure that has governed H-1B Labor Condition Applications for years. We break that system down in H1B Wage Levels Explained.

Three mechanics deserve attention:

Multiple employers, one lowest level. If several employers register the same beneficiary, USCIS counts the person once and assigns the lowest wage level among all registrations. An employer cannot inflate a candidate's odds by filing a parallel high-level registration.

Multiple worksites, same floor. If the job spans locations with different prevailing wage levels, the registration must use the lowest corresponding level across worksites that the offered wage would satisfy.

Registration-petition consistency. The level declared at registration is expected to match the LCA and petition that follow. A petition filed at a materially lower level than the registration could draw scrutiny — if your compensation changes between March and filing, talk to an immigration attorney before submitting.

How registration changed in practice

The March registration used to be thin: beneficiary details, passport information, the fee, done in minutes. Wages barely came up.

An FY2027 registration asks for the job itself:

• The SOC code for the offered position, chosen from the actual duties.
• The area(s) of intended employment — the worksites that determine which OEWS wage table applies.
• The offered wage, which USCIS maps against DOL's thresholds to assign the wage level — and with it, the number of pool entries.

The employer attests that the declared level reflects a bona fide offer at that wage, and the final rule treats that March declaration as the reference point for everything that follows: the certified LCA and the petition are expected to line up with it.

One practical consequence: the lottery conversation moved from March to offer stage. By the time a registration is submitted, everything that sets the weight — duties, location, salary — was negotiated months earlier. A candidate who knows their level in January can still do something about it. In March, the number is already fixed.

What DHS estimates your odds to be

DHS published estimated selection probabilities in the rule's analysis, based on a simulated pool drawn from recent registration data:

Wage level	Old random lottery	Weighted lottery (DHS estimate)	Change
Level I	~29.6%	15.3%	−48%
Level II	~29.6%	30.6%	+3%
Level III	~29.6%	45.9%	+55%
Level IV	~29.6%	61.2%	+107%

A few things to understand about these numbers:

They are estimated odds, not guarantees. DHS assumed a pool of 320,711 unique beneficiaries — distributed 89,911 at Level I, 177,216 at Level II, 37,928 at Level III, and 15,657 at Level IV — competing for roughly 94,900 selections. That assumption produced the 29.6% random baseline (94,900 ÷ 320,711) and a weighted pool of about 620,755 entries, where one Level I entry is worth a 15.3% chance and each higher level multiplies it. If the real pool in a given year skews differently, actual selection rates shift with it.

The weighting is roughly linear by design. Level IV's 61.2% is almost exactly four times Level I's 15.3%. That is a modeling simplification — in a real draw without replacement, four entries are worth slightly less than four single entries — but DHS's own simulation landed close to these marks, and they are the figures practitioners use for planning.

Level II is the quiet winner. Most cap-subject petitions historically sit at Levels I and II (DHS cites a distribution of roughly 28% / 55% / 12% / 5% across Levels I–IV). Under the new math, the modal H-1B worker — Level II — sees estimated odds slightly better than the old lottery, not worse.

For a personalized estimate that layers in the master's cap, use our H1B lottery odds simulator, and see the companion piece What Are My H1B Lottery Odds in 2026?.

The master's cap under the weighted rule

The statutory cap is untouched: 65,000 regular numbers plus 20,000 for beneficiaries with a U.S. master's degree or higher. The two-stage order also survives: the regular-cap selection runs first, and advanced-degree registrations not chosen there are pooled again for the 20,000 exemption.

What changed is that both stages are now weighted. A Level III registration draws with three entries in the regular round and, if it misses, three entries again in the master's round. The practical result: a U.S. master's graduate with a Level III or IV offer now stacks two structural advantages — double exposure to the draw and multiplied entries in each.

What actually happened in FY2027

The first weighted lottery is done, and early USCIS figures reported by immigration counsel show the mechanism working as designed:

• Registrations collapsed. Properly submitted registrations fell from 343,981 (FY2026) to 211,600 — a 38.5% drop. With entries weighted and the $100,000 fee hanging over abroad hires, speculative registrations largely disappeared.
• Selection skewed senior and advanced-degree. 71.5% of selected beneficiaries held U.S. advanced degrees, up from 57% the prior year.
• Level I got squeezed. Only 17.7% of selected registrations were in the lowest wage category.
• No second lottery. Petitions filed by the June 30 deadline filled the cap; USCIS confirmed no additional selection rounds.

Firm-level data is starting to fill the gap. Manifest Law, reviewing its own FY2027 caseload, reported Level I selections at 24.5% — well above DHS's 15.3% estimate — a useful reminder that actual rates depend on the real pool and that DHS's figures are planning estimates, not outcomes.

USCIS has not yet released complete FY2027 selection statistics by wage level. When it does, we will recalibrate the odds simulator against actuals.

Who is hit hardest

Entry-level hires on Level I wages. A new graduate whose offer clears only the 17th-percentile threshold saw estimated odds fall from ~29.6% to 15.3%. With three years of STEM OPT, that is still multiple bites at the apple — but each bite is smaller, and the old "register every year and wait" strategy is weaker than it used to be.

Staffing and consulting companies. The IT staffing model runs on high volumes of Level I/II registrations placed at third-party worksites. DHS said the quiet part out loud in the rule: the point is to shift selections toward higher-wage petitions. The FY2027 registration collapse — down 138,000 in one year — is largely this segment pulling back.

Small employers and nonprofits that are cap-subject. They often cannot match big-tech wage bands, and their offers more often land at Level I or II. (Cap-exempt employers — universities, affiliated nonprofits, nonprofit and government research organizations — remain outside the lottery entirely, which makes them newly attractive to candidates.)

Who benefits: employers whose offers clear Level III or IV — mostly large tech companies, finance, and senior roles everywhere — plus master's-degree holders, and foreign graduates negotiating from strength. A candidate with competing offers now has a concrete, quantifiable reason to take the higher-paying one.

Five misconceptions making the rounds

"A bigger salary always means more entries." Only relative to local thresholds. The level is set by where your wage falls in the OEWS distribution for your SOC code and metro — a salary that is Level IV in Cleveland can be Level II in San Jose.

"The employer picks the level." The level follows from comparing the offered wage to DOL thresholds. Declaring a level the job and wage do not support is not strategy; it is exposure.

"Registering through more employers helps." The beneficiary-centric rule counts you once no matter how many employers register you — and the lowest level among all registrations applies. A stray Level I registration can drag down a Level III one.

"The master's cap is separate from the weighting." Both stages are weighted now. Advanced-degree registrations draw with their full entries in the regular round and again in the 20,000 round.

"Level IV means selection is locked." 61.2% is DHS's estimate for a modeled pool, not a promise. Actual odds move with each year's registration volume and level mix, and four entries in a draw without replacement are worth slightly less than 4× a single entry.

Ranking vs. weighting: why this version went live

This was not DHS's first attempt to tie selection to wages. In January 2021, at the end of the first Trump administration, DHS finalized a rule that would have ranked registrations: all Level IV registrations selected first, then Level III, and so on until the cap filled (86 FR 1676). In most years, that design would have shut Level I out entirely.

That rule never operated. It was delayed, challenged, and formally withdrawn before a single registration was ranked.

The 2026 rule is the same policy instinct in a legally sturdier package. Weighting instead of ranking means:

• No level is excluded. Level I keeps a real chance — DHS's own estimate is 15.3% — which undercuts the argument that the agency wrote whole categories of workers out of the program.
• The beneficiary-centric system survives. One person, one cap slot, with the integrity framework USCIS built in 2024 intact.
• The mechanism is arithmetic, not discretion. USCIS applies a fixed multiplier; it does not pick winners by occupation or industry.

Whether that is enough to survive court review is the open question — the next section covers it.

Could the rule be struck down?

The rule could face legal challenges. Immigration law firms, including Fragomen, flagged litigation risk from the day it was announced, and industry groups have argued DHS exceeded its statutory authority by ranking beneficiaries — something the statute does not obviously authorize, and a rationale that helped sink the Trump administration's 2021 wage-based selection rule (which DHS withdrew before implementation).

As of this writing, no court has enjoined the rule, and it governed the full FY2027 season. But the ground can shift in three ways, and each changes what you should do:

• A court vacates the rule mid-cycle. USCIS would likely revert to random selection for the next season; past selections already made would probably stand.
• DOL finalizes its percentile increase. The wage levels themselves get more expensive to reach — see our tracker on the DOL proposal.
• Nothing happens. The weighted system beds in, and wage-level strategy becomes a permanent part of H-1B planning, the way SOC-code hygiene already is.

What to do now

1. Find your level. Your odds turn on one number: which OEWS level your offered wage hits for your SOC code and metro. Check your wage level → — it takes under a minute and shows the exact gap to the next level.
2. Quantify the gap. If you are $4,000 short of Level III, that figure is negotiating leverage: it is the difference between ~30.6% and ~45.9% estimated odds.
3. Run your odds. Layer in the master's cap and see your estimated odds against the old system in the H1B lottery odds simulator.
4. Plan FY2028 early. Registration opens in March 2027. Level strategy — SOC selection, worksite planning, offer structure — is now decided in the offer letter, months before the lottery. Browse real employer filings in the H1B Levels salary database to see the wage levels employers actually certified for your role.

One caution on data: employer histories on this site reflect certified LCAs filed with DOL. A long filing history tells you a company knows the process; it does not mean any company will sponsor you for a specific role.

Frequently asked questions

Is the H-1B lottery still random in 2026? No. Effective February 27, 2026, DHS replaced random selection with wage-weighted selection. Registrations enter the pool 1–4 times based on the OEWS wage level of the offered job.

How many entries does each wage level get? Level I = 1, Level II = 2, Level III = 3, Level IV = 4. The level comes from DOL prevailing wage thresholds for the SOC code and work location, not from the employer's preference.

What are the estimated odds under the weighted lottery? DHS estimates 15.3% (Level I), 30.6% (II), 45.9% (III), and 61.2% (IV), versus ~29.6% for everyone under the old random draw. These are estimated odds based on a simulated pool; actual rates depend on each year's registrations.

Does the master's cap still exist? Yes — 65,000 regular plus 20,000 U.S. advanced degree numbers, unchanged. Advanced-degree registrations that miss the regular selection get a second weighted draw in the 20,000 pool.

What if two employers register me at different levels? You are counted once, at the lowest wage level among all registrations submitted for you.

Can the rule be overturned? It could face legal challenges, and firms including Fragomen have flagged that risk. The rule is in effect today and governed FY2027, but monitor litigation before relying on it for FY2028 planning.

Sources

• DHS final rule, Federal Register, December 29, 2025 (90 FR 60864): https://www.govinfo.gov/content/pkg/FR-2025-12-29/pdf/2025-23853.pdf
• DHS proposed rule with odds analysis, September 24, 2025 (90 FR 45986): https://www.govinfo.gov/content/pkg/FR-2025-09-24/pdf/2025-18473.pdf
• ABIL Immigration Insider, January 4, 2026 (17,000 comments; court challenges expected): https://www.abil.com/2026/01/abil-immigration-insider-january-4-2026/
• National Law Review: USCIS Completes FY2027 H-1B Lottery (March 31, 2026): https://natlawreview.com/article/uscis-completes-fiscal-year-2027-h-1b-lottery
• WR Immigration: FY2027 cap reached, no second lottery; registration and selection figures: https://wolfsdorf.com/h-1b-cap-fy2027-lottery-results-next-steps/
• Envoy Global FAQ on the weighted selection rule (master's cap mechanics, lowest-level rule): https://www.envoyglobal.com/insight/understanding-the-weighted-h1b-cap-selection-rule-faq/
• Brown Immigration Law: FY2027 lottery odds table and multi-worksite rule: https://brownimmigrationlaw.com/resources-library/h-1b-lottery-2025-fy26/
• Manifest Law: FY2027 firm-level selection rates vs. DHS estimates: https://manifestlaw.com/blog/immigration/news/analyzing-our-h1b-selection-results-fy2027/
• Buchanan Ingersoll & Rooney: FY2027 registration window and $215 fee: https://www.bipc.com/fy-2027-h-1b-cap-registration-opens-march-4,-2026-key-changes-and-action-items
• USCIS H-1B Electronic Registration Process: https://www.uscis.gov/working-in-the-united-states/temporary-workers/h-1b-specialty-occupations/h-1b-electronic-registration-process

This site is an independent data tool and is not affiliated with any government agency. Nothing on this site constitutes legal advice or immigration advice. Lottery odds shown are statistical estimates based on DHS-published figures and assumptions; actual outcomes depend on the annual registration pool and USCIS processing. Consult a licensed immigration attorney for advice about your specific situation.

Data source: U.S. Department of Labor, Office of Foreign Labor Certification (OFLC) disclosure files, updated quarterly. LCA filing data reflects applications certified by DOL; an LCA is a labor condition application and does not by itself constitute an approved H-1B petition or an employment offer. Wage level thresholds are derived from BLS Occupational Employment and Wage Statistics (OEWS). Last data update: Q2 FY2026.