Reviewed by [[AUTHOR_NAME]], [[AUTHOR_CREDENTIAL]]

TL;DR

A law school predictor is a tool that estimates how your LSAT score and GPA compare to a law school's historical admitted-student data. It runs a statistical model, usually logistic regression, over past LSAT/GPA combinations. It does not measure your essays, letters, or story. Treat the output as a benchmark against past cycles, not a real probability of admission.

What is a law school predictor?

A law school predictor is an online tool that compares your LSAT score and undergraduate GPA against a law school's historical admitted-student data to estimate where you land relative to past applicants. Most predictors draw on two data sources: official percentile bands and crowdsourced applicant reports. Law schools disclose 25th, 50th, and 75th percentile LSAT and GPA figures for each admitted class in required ABA Standard 509 reports. Predictors overlay your numbers onto that band, or onto a larger pool of self-reported admit-and-deny data from applicant forums, to generate a result.

The term "predictor" is doing a lot of marketing work here. Nothing about the tool actually predicts an individual outcome. It pattern-matches your numbers against numbers from people who applied in prior cycles, some of whom got in and some of whom didn't, and reports how your profile sits within that distribution. That's useful information. It is not a forecast of your specific decision letter.

Most predictor tools fall into one of two camps. The first pulls exclusively from official ABA 509 disclosures, which are audited and standardized across schools. The second pulls from self-reported data — applicants voluntarily submitting their own LSAT, GPA, and outcome to a shared database. Self-reported pools are larger and more granular, since they can include cycle year, URM status, and work experience tags, but they are also unverified and skew toward applicants active on forums, which is not a random sample of the applicant pool.

How predictors calculate your odds

Predictors combine two or three data layers to generate a result: published percentile bands, self-reported admit/deny history, and a statistical model — usually logistic regression — that weights your LSAT score more heavily than your GPA. The output is a modeled likelihood based on numbers alone. It is not a probability of admission calculated for you specifically, because soft factors aren't in the math.

Here's what most predictor models actually take as inputs:

• LSAT score. This is typically weighted more heavily than GPA in the underlying model, because LSAT has historically shown a stronger statistical relationship to admit outcomes across schools in aggregate self-reported datasets.
• Undergraduate GPA. Schools also disclose GPA percentile bands under ABA 509 rules, so this gets compared the same way as LSAT.
• Cycle year. Median LSAT and GPA figures shift from year to year as application volume rises and falls, so a predictor trained on last cycle's data can be stale by the time you apply.
• Self-reported tags, where available: work experience, first-generation status, state residency, or major. Not every predictor collects these, and none of them can verify that a user entered accurate information.

What the model cannot see, and never will, is everything that actually swings a real admissions decision at the margins: your personal statement, your letters of recommendation, addenda explaining a gap year or a bad semester, and how a specific admissions committee is weighting diversity of experience or geographic reach in that particular cycle. A predictor treats every LSAT-175/GPA-3.8 applicant as interchangeable. Admissions offices do not.

That's the core limitation to hold onto: a percentage from a predictor describes a historical pattern, not your file. Two applicants with identical numbers can get very different results from the same school in the same cycle, and neither predictor accuracy nor bad luck fully explains why — soft factors do.

Predictor vs calculator vs consultant

A predictor outputs a modeled percentage using historical LSAT/GPA combinations, a calculator benchmarks your numbers against a school's officially disclosed percentile bands without producing an odds figure, and a consultant reviews your entire application — numbers, essays, letters, and narrative — to build a strategy that no algorithm can replicate.

Each tool answers a different question, and picking the wrong one for the job is where applicants waste time.

• Tool: Predictor (e.g., crowdsourced tools like LSData) | Primary output: Estimated percentage or odds figure | Data basis: Self-reported admit/deny history from prior applicants | What it can't see: Essay quality, letters, addenda, soft factors | Typical cost: Usually free | Best use: A fast, rough gut-check on where you stand numerically
• Tool: Calculator (e.g., Lovare Institut's admissions calculator) | Primary output: Your percentile position against a school's published median | Data basis: ABA Standard 509 disclosed LSAT/GPA bands | What it can't see: Doesn't generate an odds figure at all | Typical cost: Free / transparent pricing, varies by platform | Best use: Benchmarking your numbers against verified, official data
• Tool: Consultant (e.g., Spivey Consulting Group) | Primary output: Personalized application strategy | Data basis: Your full file, read holistically | What it can't see: Can't compare you across thousands of profiles simultaneously | Typical cost: Hourly or package fees, [[VERIFY: current consultant hourly rate ranges]] | Best use: Positioning your narrative, school list, and addenda

Each of these has a genuine strength worth naming honestly. Self-reported predictor tools like the ones built on LSData's applicant pool have real scale — thousands of data points per cycle — which is more granular than any single school's disclosed 509 report. Independent consultants like Spivey Consulting bring pattern recognition from reading full files across many admissions cycles, which a numbers-only tool structurally cannot do. A calculator's advantage is narrower but firmer: it never overstates certainty, because it reports where your numbers sit against a school's own disclosed data instead of manufacturing a percentage.

How Lovare approaches admissions benchmarking: Lovare Institut's admissions calculator ties your LSAT and GPA to each school's published ABA 509 percentile bands rather than generating a probability figure, because a number derived from your file alone can't responsibly claim to predict a committee's decision. That benchmarking approach is informed by a dataset drawn from over 10,000 applicant outcomes and 4,000 scholarship negotiations Lovare has tracked across its advising practice, and it sits inside a platform that continues past the admissions letter — into 1L tools, bar prep, and career support through the Legal Mentor Network — rather than stopping once you're accepted. If you want a strategy conversation instead of a benchmark, that's a separate service, and Lovare's advisors are upfront about which one you're getting.

FAQ

Are law school predictors accurate?

Law school predictors are only as accurate as the data feeding them, and self-reported applicant data is not a verified, random sample. They're reasonably useful for a rough sense of where your numbers land historically. They are not reliable enough to treat as a real forecast of your individual decision.

What data do law school predictors use?

Most predictors use a mix of officially disclosed ABA Standard 509 percentile bands and self-reported LSAT, GPA, and outcome data from prior applicants on forums or databases. Some also collect optional tags like work experience or state residency, though these are unverified and not universally available.

Can a predictor guarantee admission?

No predictor, calculator, or consultant can guarantee admission to any law school. Admissions decisions weigh essays, letters, and committee priorities that no numbers-only tool captures. Any tool claiming a guaranteed outcome should be treated with real skepticism.

Should I use a predictor or a calculator?

Use a predictor for a fast, rough read on historical patterns, and use a calculator when you want your numbers compared directly against a school's own officially disclosed percentile bands. A calculator makes fewer claims and is generally the more defensible starting point.

Do law schools use predictors themselves?

Law schools do not use third-party predictor tools to make admissions decisions. Admissions committees rely on internal file review, published median targets for ranking purposes, and holistic reading of essays and letters, not an outside algorithm's percentage.

How often should I rerun a predictor?

Rerun a predictor whenever a school's admitted-class medians update, typically once per cycle after ABA Standard 509 disclosures are filed. Also rerun it after a retake, since a new LSAT score meaningfully shifts your position in the underlying model.