Why Wage Backpay results differ in New Hampshire
6 min read
Published April 15, 2026 • By DocketMath Team
The top 5 reasons results differ
When you run a Wage Backpay calculation in New Hampshire (US-NH) with DocketMath, you may notice that two seemingly similar scenarios produce different outputs. Those differences are usually caused by (1) time-limit filtering and (2) how wage amounts are converted and aggregated across the pay periods you input.
Baseline to keep in mind first: In New Hampshire, the general statute of limitations (SOL) for civil actions is 3 years under RSA 508:4. Jurisdiction data for this project did not identify a claim-type-specific sub-rule, so treat 3 years as the general/default period for SOL lookback in this calculator. (If a specific claim type applies differently, the result could change—but with the information provided, the default general rule is what DocketMath will model.)
Here are the top 5 reasons results differ in US-NH:
**Different “lookback start” date (SOL cutoff)
- DocketMath uses the backpay start/end dates you enter to determine which backpay months fall within the 3-year window.
- Practical impact: moving the earliest asserted shortfall date forward can exclude earlier months, reducing total backpay; moving it earlier can include more months, increasing total backpay.
Misaligned pay periods vs. calendar months
- Two runs can effectively “bucket” wages differently (e.g., weekly/biweekly pay applied across months vs. monthly aggregation).
- Even if the same pay rate is used, the way wages are distributed over time can change what DocketMath totals for each month.
**Rate inputs aren’t equivalent (hourly vs. salary conversion)
- Wage backpay outputs can shift depending on whether you input:
- an hourly rate plus expected hours, or
- a salary amount plus an assumed hours basis.
- Example: If the same salary figure is treated with different expected hours (e.g., 35 vs. 40 hours/week), the effective hourly/monthly wage changes—so the backpay difference can be large.
Mitigation / earnings offsets applied with inconsistent conventions
- If your scenario includes interim earnings or offsets, users sometimes structure inputs differently (e.g., entering offsets as a net reduction vs. entering additional earnings to be subtracted later).
- DocketMath will follow the structure you provide, so differences often come from inconsistent offset formatting between two runs.
**Start/end dates vary by a day (boundary effects)
- A one-day shift can move an entire pay period into or out of the 3-year SOL lookback.
- Practical impact: when the earliest included wage period is near the cutoff, small date changes can produce outsized total differences.
Common pitfall: If the earliest alleged underpayment date is off even slightly, SOL filtering can exclude/include multiple pay periods, leading to a noticeably different total.
For context on the time limit driving many of these differences: RSA 508:4 provides a general 3-year SOL, so DocketMath may “prune” backpay months outside that 3-year window when the calculation period extends farther back.
How to isolate the variable
To understand why two DocketMath results differ, isolate the inputs like a mini-audit. The goal is to determine whether the change came from (a) SOL cutoff filtering, (b) wage calculation/aggregation, or (c) offsets.
- Freeze the jurisdiction and tool settings so both runs use the same rule set.
- Compare one input at a time (dates, rates, amounts) and re-run after each change.
- Review the breakdown to see which segment or assumption drives the difference.
Step 1: Freeze the date range first
- Keep wages (rates), expected hours, pay cadence, and offsets the same.
- Re-run by changing only the backpay start/end dates.
What you’ll learn: If totals change primarily with the dates, SOL lookback under the general 3-year rule (RSA 508:4) is likely the driver.
Step 2: Hold dates constant; compare period math assumptions
- Keep start/end dates the same.
- Change only one of the following:
- pay cadence / aggregation approach (weekly vs. biweekly vs. monthly style inputs)
- expected hours used for converting rate types
What you’ll learn: If totals change when you change cadence or expected hours, the difference is likely due to conversion/aggregation, not SOL filtering.
Step 3: Validate offsets consistently
- Keep dates and wage assumptions fixed.
- Adjust only offset/earnings inputs—making sure the convention matches between runs.
What you’ll learn: If differences cluster across specific months where offsets overlap, inconsistent offset structure is likely the cause.
Step 4: Use a one-change-at-a-time checklist
Before you interpret results, confirm:
To compare quickly, start from the same base inputs and run small variations rather than changing multiple fields at once.
To run your own comparison, use DocketMath here: /tools/wage-backpay.
Next steps
Create a single source-of-truth timeline
- Write down the earliest date you can support for the wage shortfall.
- Write down your calculation end date.
- Because New Hampshire models a general 3-year SOL under RSA 508:4, the earliest date can strongly affect which months are included.
Standardize wage inputs
- Pick one input style (hourly-based or salary-based) and use it consistently across runs.
- Keep expected hours and pay cadence aligned.
Confirm offset entry method
- If interim earnings exist, ensure both runs represent them the same way in DocketMath.
- Inconsistent conventions are one of the most common causes of “mysterious” result differences.
Run a controlled boundary test
- Make two versions:
- Version A: backpay start at the earliest asserted date
- Version B: backpay start moved forward by ~14–30 days
- If totals change sharply, you’re likely near the 3-year SOL cutoff under RSA 508:4.
Keep a short calculation log
- Save the input set that produced the result you consider most accurate.
- When you correct dates or assumptions later, you’ll be able to trace why totals changed.
Note: DocketMath can help you model outcomes based on your inputs and New Hampshire’s general 3-year SOL (RSA 508:4). This is guidance for understanding calculation mechanics, not legal advice.
