Numbercade is seven arcade-style games that teach the math of grades 4 to 7: fractions, ratios, negative numbers, equations, and the rest of the stuff that suddenly stopped looking like math. Built around what randomized trials say actually helps kids learn, and playtested by two real kids with strong opinions.
Play free No account. No download. It runs in the browser your kid already argues with.SEVEN CABINETS • SEVEN SKILLS • SEVEN BOSSES
While we're in soft launch, every world of every game is open. No email required. If your kid doesn't ask to keep playing, you've lost nothing but fifteen minutes.
Here's a secret from someone who reviewed education research for a living: "research-based" can mean two very different things. It can mean the product itself was tested in real studies. Or it can mean the product was designed from practices that were tested. Most companies blur those two on purpose. We won't. Numbercade is the second kind: every mechanic in these games exists because randomized trials and the Department of Education's practice guides recommended the practice behind it, and we can point to which research. Tap "The Research" inside any game and you'll find the actual citations: practice guides, peer-reviewed studies, page numbers and all. Not "gamified learning science synergy." Actual receipts.
And lately there's a third thing "research-based" can mean: a badge. Some "ESSA certified" badges in this industry are sold by commercial certification services as part of a vendor's marketing toolkit. That is not the same as a rating from an independent registry like Evidence for ESSA, the one run by Johns Hopkins researchers, and at least one popular math app currently displays a purchased certification badge while that independent registry lists it as having no qualifying studies at all. So when you see a badge, ask who issued it and who paid for it. We'll always tell you what we have: not a badge, a bibliography.
Education law even has names for this. Federal ESSA rules sort evidence into four tiers, from "the product was proven in large randomized trials" (Tier 1) down to "designed from research, with a plan to study itself" (Tier 4, called Demonstrates a Rationale). We're Tier 4, we publish the logic model that tier requires, and we'll tell you exactly that instead of hiding it behind a vague "research-based" badge.
Built by a PhD scientist who used to review education research professionally, and who got tired of apps that say "research-based" the way cereal boxes say "part of a complete breakfast."
Plenty of math apps bolt a cartoon onto a worksheet and call it a game. Ours went through two kid playtesters with veto power. When a mini-game was boring, it got deleted (rest in peace). It was replaced by a slingshot target range the testers now fight over. Pirates, kitchens, block worlds, crystal mines: kids picked these themes because kids play these games.
Sessions are designed short because spaced practice, spread over days, beats cramming, and that's one of the most consistently replicated findings in learning research. The games also quiz kids on things they got wrong days ago (there's a Review Missed button, and it's not decorative), because retrieving a memory strengthens it more than rereading ever will. Short, spaced, and a little sneaky.
The humor in these games isn't decoration. Classroom experiments, including randomized studies by researcher Avner Ziv, found that relevant humor woven into instruction improved learning, while random clowning did not. So our jokes live inside the math: the scam artist whose "deals" teach percent, the villain whose wrong answers teach error-spotting. If your kid repeats a bit at dinner, that bit was teaching ratio.
Most apps check the answer. Ours teaches the moves. In Equation Pirates, kids pick each step and watch both sides of the equation transform, explain why a solver did what they did, and play detective on wrong solutions seeded with the exact mistakes kids really make. That's straight out of the Department of Education's algebra practice guide: solved problems, structure first, compare strategies. It's the difference between "wrong, try again" and "here's why that move sank your ship."
We collect one email address: yours. Kid profiles are nicknames and emoji, nothing else. No ads, ever. No third-party trackers, no "free" business model where your kid is the product, no selling anything to anyone. Progress saves on your device. We'd tell you what we do with kids' data, but we don't have any.
One grown-up email makes the account. Kids get profiles with nicknames and an emoji avatar. That's all the personal information there is.
Seven games, seven skill areas: equations, fractions and percents, order of operations, slopes, geometry, roots and exponents, probability. Start with what's wobbling this month, or start at game one and run the gauntlet.
Every skill area tracks lessons learned, bosses beaten, and missed questions cleared.
Most kids are designed to solidify all seven areas in one season at about 15 minutes a day, 4 days a week. That's our design target, not a guarantee; kids are gloriously non-standardized.
We think about this the way you think about soccer: sign up for the season, show up, get better, and nobody charges your card forever after your kid loses interest.
$14/month
$49/4 months
$99/12 months
If it's not working for your family in the first 14 days, tell us and we'll refund it. No exit interview.
Want a heads-up when passes go on sale? Email sign-up isn't wired up yet; for now, write to info@numbercade.com with the subject "season pass" and we'll tell you the day it opens.
This form goes live when passes do. Until then it's honest about being decorative.
No, and it's not trying to be. School (or your homeschool curriculum) covers the waterfront. Numbercade is targeted reinforcement for the seven skill areas where kids most often fall off the cliff: equations, fractions/ratios/percents, order of operations, linear equations and slope, geometry, roots and exponents, and probability. Think of it as the climbing gym, not the mountain.
It's fifteen minutes of screen time with a job. Sessions are short by design, there's no infinite scroll, no autoplay, and no mechanic engineered to keep kids on past the lesson. The game literally tells kids to grab scratch paper. When the session's done, it's done.
The sweet spot is grades 4 to 7 (roughly ages 9 to 12). A strong third grader can start with the earlier worlds; an eighth grader shoring up algebra basics won't feel babied. The games never mention grade levels, so nobody feels behind.
Any modern browser: laptop, Chromebook, tablet, or phone. Nothing to install, nothing to update, no app store account. If it can load a webpage, it can run the arcade.
Fourteen days, full refund, just ask. And because passes don't auto-renew and we don't store your card, you will never discover a surprise charge from us in month nine.
We designed for compliance by collecting almost nothing: one parent email for the account, kid profiles that are just a nickname and an emoji. No ads, no trackers, no data sales, and game progress lives on your device. Parents create and control all profiles.
Fair question, because that phrase can mean two different things. Meaning one: the product itself was tested in studies. Meaning two: the product was designed from practices that were tested. Most companies blur the two; we think you deserve to know which one you're buying. Numbercade is meaning two: the games were designed around findings from randomized trials and the Department of Education's practice guides, and each game includes a "The Research" page listing which mechanic implements which recommendation, with citations. In the federal ESSA evidence-tier system, that makes us Tier 4, Demonstrates a Rationale: a research-anchored design plus a public plan to study our own effects. It does not mean this product itself has been through a trial, and we won't pretend otherwise. Designed around the evidence: yes. "Proven to raise your kid's grades": nobody can honestly say that about a product that hasn't been independently tested, so we don't.
Honest answer from a researcher: it depends on what you're optimizing for, and some of the big products have more product-level evidence than we do. DreamBox, for instance, has two randomized studies behind its K-5 rating on Evidence for ESSA, the independent Johns Hopkins registry; the average effect size recorded there is +0.10, which is real but modest, and the evidence covers grades K-5. If that fits your kid, it's a defensible choice, and we'd rather tell you that than pretend otherwise. What bothers us about the category is the distance between evidence and advertising. Prodigy's page on that same Johns Hopkins registry currently reads "No studies met inclusion requirements," and in 2021 the advocacy group Fairplay (then called Campaign for a Commercial-Free Childhood) and 21 other organizations filed an FTC complaint alleging that Prodigy's free home version aimed a stream of paid-membership advertising at children; the complaint reported counting 16 membership ads and four math problems in one 19-minute session. Those are allegations, Prodigy disputed the characterization, and no public FTC action followed. But the underlying design facts matter to us as parents: we built Numbercade with no ads, no in-game upsells, no membership tiers visible to other kids, and a pass that expires honestly. So here's the straight comparison: if you want the product with the most product-level trials behind it today, that isn't us, and we say so. If you want a product that never advertises to your kid and never claims one inch past its evidence, that's the one we built.
Yes, on purpose. Loot Lab puts kids on the other side of the counter: they run the loot-box shop, so they have to compute the real odds behind the shiny crates. Kids who can calculate that a "rare drop" is a 1-in-40 event, and what that costs over ten tries, are kids who are harder to fool later. It's probability class disguised as tycoon game, and media literacy disguised as both.
Not yet. It's on the list. Teachers who want to pilot it anyway: email us, we like you.
A PhD scientist and parent who used to review education research for a living, plus her two kids, who playtested every screen and got at least one boring mini-game executed. The full story is in the founder's note below.
I spent years as a researcher at Mathematica, the organization that conducts study reviews for the What Works Clearinghouse, the Department of Education's standard for "does this actually work?" My job was reading education research with a skeptical eye. (To be clear: that's my resume, not an endorsement. Neither Mathematica nor the What Works Clearinghouse has reviewed or endorsed this product.)
Then my own kids hit the math cliff.
You know the cliff even if you've never heard the name. Somewhere around fourth or fifth grade, math stops being counting and starts being ideas: fractions that are somehow bigger when the number on the bottom is smaller, numbers that live below zero, letters that stand for numbers nobody will tell you. Kids who were "fine at math" suddenly aren't. And the confidence goes first, before the grades do.
So I went shopping for help, the way you probably did before you landed here. I found two aisles. Aisle one: workbooks in disguise. Apps that are really just worksheets with a progress bar, that mark answers wrong without ever explaining why, exactly like the answer-key-only prep books that frustrated me the first time around. Aisle two: slot machines in disguise. Games with spinning rewards and streaks and gems, where the math is a toll you pay to get back to the fun part. Neither aisle teaches the thing my kids were actually missing: WHY the math works.
As a parent, I was underwhelmed by the options. As a scientist, I knew there was decades of randomized-trial evidence about what actually helps kids at this exact stage, sitting in practice guides almost nobody builds from. So I built it myself.
My two kids were the first players, and they were brutal. My nine-year-old read a dash in a math sentence as a minus sign and got the problem wrong through no fault of his own; there are no stray dashes anywhere in these games now. A mini-game I was proud of got a flat "this is boring," so it's gone, replaced by a slingshot target range they now fight over. Every lesson, every joke, every boss went through the two toughest reviewers I've ever had.
The result is seven games that teach why, not just what. Short sessions, real explanations, humor that carries its weight, and a design target we take seriously: about fifteen minutes a day, four days a week, for one season.
Your time is limited. Your kid's confidence is precious. Getting over the math cliff shouldn't take over your life.
[Founder name], PhD
Mom of the two harshest playtesters in the business
Mathematica and the What Works Clearinghouse do not review or endorse this product. References are biographical.