Chessable is a $11.99 a month opening and course platform, owned by Chess.com, built around one core training loop: MoveTrainer, a spaced repetition drill that turns an authored repertoire into moves you can recall over the board. It has a genuinely strong library of courses from named grandmasters, put above 1,000 by third party counts, a 4.8 out of 5 rating on roughly 5,900 iOS reviews, and one significant limitation almost nobody writing about it names specifically: MoveTrainer runs the exact same fixed interval schedule for every move in every course, whether that move is obvious or counterintuitive, and a single miss sends it back to the bottom of that schedule. This Chessable review covers what the platform actually does, what it costs, where it is worth the money, and what a per move scheduling approach, ours, does differently, with our own generated numbers rather than a claim we cannot back up.
What Chessable actually is
Chessable is an online chess training platform built around interactive courses: an author, usually a titled player or coach, builds out a repertoire or set of study material as a "book," and the platform turns each line into a drillable variation instead of a page you passively read. Third party sources put the library above 1,000 courses, with named authors including Magnus Carlsen, Anish Giri, Judit Polgar, Sam Shankland and Peter Svidler. A figure of 600 plus also circulates, including in summaries dated this year, so the counts genuinely disagree rather than one simply being older than the other. We could not settle it: chessable.com returned a 403 on every fetch attempt during this research. Treat any specific course count for Chessable, including ours, as unverified at source.
Chessable vs Chess.com: ownership, not rivalry
Chessable is not a rival to Chess.com, it is one of Chess.com's own brands. Chess.com's acquisition of Play Magnus Group, Chessable's parent company, closed 16 December 2022, folding in Chessable alongside Chess24, AimChess, Magnus Academy, Everyman Chess, New In Chess, iChess.net and GingerGM. Play Magnus Group had itself bought Chessable back in February 2019. So "Chessable vs Chess.com" is not really a comparison to make: Chessable is a standalone app and brand that Chess.com owns and operates, not a competing product.
In April 2026, Chessable shipped a Repertoire feature that lets any user pull individual variations or whole chapters out of courses they already have access to, edit them, add their own notes and alternative moves, and train the result through MoveTrainer independent of the source course. Chess.com's own announcement carries an April 14 timestamp, though a Chessable blog post references it under a March 29 date, so we are not pinning the exact ship day tighter than that.
What MoveTrainer actually does
Per Chessable's own help centre, MoveTrainer runs a fixed, eight level graduated interval ladder. Answer a move correctly and it advances to the next level; miss it and it resets back toward level one. Each move inside a variation keeps its own independent timer, so you can sit at different levels on different moves within the same line. The levels themselves:
| Level | Interval |
|---|---|
| 1 | 4 hours |
| 2 | 1 day |
| 3 | 3 days |
| 4 | 1 week |
| 5 | 2 weeks |
| 6 | 1 month |
| 7 | 3 months |
| 8 | 6 months |
That schedule does not change based on the move. A forced recapture nobody in a course has ever missed, and a quiet, easy to overlook prophylactic move, both climb the same eight rungs at the same pace, and both fall back toward level one the same way on a miss. Chessable's documentation does not name an algorithm family behind this, no SM 2, no FSRS, no per card ease factors or fitted memory parameters. A separate help article titled "What is the Schedule setting" exists but was inaccessible during our research, so we cannot say what it configures. One rival product, ChessAtlas, states on its own marketing page that "SM 2 powers Chessable's MoveTrainer," and separately claims FSRS needs 20 to 30 percent fewer reviews than SM 2 for the same retention, citing "an open spaced repetition benchmark across 500+ million review logs." Neither claim cites a primary Chessable source, and the 20 to 30 percent figure is a general SM 2 versus FSRS number from that outside benchmark, not something measured against Chessable specifically. Treat the SM 2 label as an educated guess from a competitor, not a documented fact. The eight level ladder itself is the part we can point at a primary source for.
Chessable pricing in 2026
A basic Chessable account is free, and it includes MoveTrainer for whatever courses you own or have free access to. PRO membership is listed on Apple's App Store at $11.99 a month or $74.99 a year; two independent secondary sources also cite a $114.99 two year PRO plan. Individual, non PRO courses are typically sold separately, in roughly the $20 to $100 plus range depending on author and length. We are flagging this pricing as secondary rather than verified at source: chessable.com and chessable.com/pro returned a 403 on every direct fetch during this research, so these figures come from Apple's listing plus corroborating third party pages, not from Chessable's own pricing page read directly.
PRO's value has shifted twice recently. Effective 2 January 2025, Chessable moved its free "Short & Sweet" trailer courses and free sample lessons, previously described as 400 plus hours of walkthrough video across 21,000 plus trainable variations, behind PRO; anyone who had already bought a course or membership got three months of free PRO as compensation. Then in 2026, some previously PRO only engine features, multi line analysis up to five lines and the seven piece tablebase, opened to all members. What remains PRO exclusive is discounted and exclusive paid content, pricing discounts that stack up to 60 percent off during sales, and features like Offline Mode.
Is Chessable worth it
On the numbers that actually represent a large sample, yes, most people paying for it are satisfied. Chessable's iOS app carries a 4.8 out of 5 rating across roughly 5,900 ratings; its Android app sits around 4.5 out of 5 on roughly 14,000 ratings, per a third party aggregator. Chessable's Trustpilot score is 2.5 out of 5, but that comes from only 10 reviews, a sample far too small and complaint skewed to represent general sentiment against nearly 20,000 combined app store ratings running the opposite direction. We are naming it only to say it is not representative, not to use it against Chessable.
Where it earns that satisfaction is real: nobody else has assembled a library this size of interactive courses actually authored by the players whose names are on them, Carlsen, Giri, Polgar, Shankland and Svidler among them. If what you want is a specific grandmaster's own repertoire, in that grandmaster's own words, buy the course. That is a product we do not build and are not trying to.
Where it falls short is specific too, and sourced rather than a vibe. Offline mode has been a recurring, multi year complaint on Chessable's own community forum, with separate threads across different years, "Offline mode Bugs," "Problems with Offline Mode," "App broken in offline," rather than one isolated report. Support response times draw real complaints too: one reviewer describes waiting three days for what they call an unhelpful answer, another says they contacted support multiple times with no response at all. And ChessDir's own review, generally favorable, still advises that for players rated under roughly 1400, "tactics and rated games return more per hour" than opening drilling on Chessable, an observation an independent 30 day hands on review echoes from a different angle: MoveTrainer teaches "what to play brilliantly, but the why lives in annotations," meaning the drill itself builds recall, not understanding.
The scheduling difference: one ladder for everyone, or a retention per move
This is the part nobody comparing Chessable to anything else has actually shown with real numbers, so here is ours.
MoveTrainer's eight level ladder is identical for every move, whether that move is a forced recapture nobody misses or a quiet positional idea most players never find. Chessdrive's scheduler, by contrast, computes a separate desired retention for every move in your repertoire, and never asks you to pick it. The formula, straight from our own scheduling code, is priority = reachProb * (1 - humanProb), where humanProb is Maia's estimate of the probability that a player at your rating plays that exact move in that exact position (the same Maia models covered in full in our Maia chess reference), and reachProb is how often that position actually comes up as you play your repertoire out. Each move's retention target is then the percentile rank of that priority inside your own full card set, mapped linearly onto a range from 0.80 to 0.97.
In plain terms: a move that shows up often in your lines, and that a player at your rating is unlikely to find on their own, gets held to a higher retention permanently. It comes back to you more often than an obvious move sitting in the exact same repertoire, for as long as you keep training it. The scheduler itself runs on ts-fsrs 5.4.1, the exact FSRS library the app depends on. We generated the schedules below directly from that code, not from a hypothetical: a fresh card walked through consecutive on time "good" reviews, with fuzz turned off so the numbers are exactly reproducible. The shipping app runs enable_fuzz set to true, a small random jitter, so real intervals land close to these numbers rather than precisely on them. The app also exposes exactly two grades, again and good, no hard, no easy, and sub day learning steps are switched off, so every interval below is a whole number of days.
| Review # | Retention 0.80 | Retention 0.90 | Retention 0.97 |
|---|---|---|---|
| 1 | 8 days | 3 days | 3 days |
| 2 | 75 days | 14 days | 3 days |
| 3 | 521 days | 57 days | 6 days |
| 4 | 2,762 days | 196 days | 11 days |
The 0.80 column stops at four rows on purpose. Its fifth review lands at 11,761 days, and by the sixth the interval reaches ts-fsrs's own hard ceiling of 36,500 days, 100 years, a limit the library itself enforces rather than a real review date anyone will hit. Printing either of those in a schedule table would make them look like planned intervals; past the fourth review at this retention the scheduler has effectively decided the move is learned, which is the honest reading, not a forecast of your reviews in the 2050s. Reviews five through eight at the higher retentions, where the numbers stay in a realistic range, look like this:
| Review # | Retention 0.90 | Retention 0.97 |
|---|---|---|
| 5 | 586 days | 19 days |
| 6 | 1,559 days | 31 days |
| 7 | 3,760 days | 50 days |
| 8 | 8,346 days | 79 days |
Same card. Same grades, good every time. Same library. The only variable across all three columns is the retention Chessdrive derived for that specific move from your repertoire and your rating. That is the entire mechanism MoveTrainer's fixed ladder does not have: nothing in Chessable's documented scheduling adjusts per move based on how likely you specifically are to find it.
Read this the right way, not the wrong one. This is FSRS behavior under Chessdrive's own parameters, not a head to head retention study against Chessable, which we have not run. We are not claiming our scheduler retains moves better than MoveTrainer does. The honest claim is about what each system optimizes for: Chessable's ladder treats every move identically once it is in a course; Chessdrive's decides, move by move, how hard that specific move is for someone at your rating to find, and schedules accordingly.
What happens on a miss: reset, or partial forgetting
Chessable's documented behavior on an incorrect answer is a reset: the move drops back toward level one of the eight step ladder, the same four hour starting point regardless of how many levels it had already climbed.
Chessdrive's FSRS scheduler does not reset on a miss, it recalculates. Take a card at retention 0.80 that has reached its third on time "good" review, sitting on a 521 day interval with a fitted stability of 157.24. Grade that same card "again" and it does not go back to day one: it comes back in 18 days, with stability falling to 5.49, a real drop, but a fraction of the accumulated memory strength, not the whole thing. The card keeps a record of how much you had already learned it; it just gets more cautious about when to trust that again.
Chessable alternatives
Chessable is not short on comparison pages, despite what a quick search might suggest. ChessAtlas runs a dedicated vs Chessable page, CircleChess publishes a Chessable alternatives post, and ChessDir carries its own directory review. None of them, in our reading, actually sources a scheduling comparison the way this page does; they compare price and course counts, not documented algorithm behavior.
Named alternatives that come up repeatedly in user discussion: Chess Tempo, Chess Position Trainer, CT Art and Listudy for tactics and opening drilling; DecodeChess, which markets itself as explaining the why behind a move rather than just scoring it; Lichess, free and open source, for playing and basic study tools; and Anki, the generic spaced repetition app some players use to build their own opening flashcards by hand instead of relying on a proprietary trainer. ChessAtlas itself is worth naming directly: a paid rival product built specifically on FSRS, positioned against Chessable on pricing and algorithm.
Chessdrive does not compete on course library, we do not have one and are not building one. One distinction worth drawing carefully, since Chessable's own April 2026 Repertoire feature sounds like the same thing: theirs assembles a repertoire out of courses you already own, which still starts from someone else's authored lines. Ours generates one from a short taste questionnaire without a course underneath it (how that maps onto realistic opening choice by rating is covered in our chess openings by rating piece), then apply the per move retention scheduling described above to it.
Who should actually buy Chessable
Buy Chessable if you want a specific titled player's repertoire, explained in that player's own words, with video walkthroughs and prose annotation attached to each line. That is a real, well executed product, and reviewers who have used it for weeks agree the recall mechanism works; the 4.8 out of 5 App Store rating on nearly 6,000 reviews backs that up. If your priority is a huge, pre built course library across openings, tactics and endgames from named grandmasters, and you are fine with the same fixed schedule applying to every move you drill, Chessable is the more complete product today.
If what you want instead is a repertoire generated for you rather than authored by someone else, an opponent that plays like a specific rating rather than a course's fixed line, and a scheduler that decides per move how hard that move actually is for you, that is the gap Chessdrive is built to fill, and we are the interested party saying so.
Where Chessdrive stands, honestly
Chessdrive is our own app, an AI chess coach built around Stockfish for post game review (covered in full in our Stockfish explainer) and the real maia3-5m model on device as the opponent, self rated from 600 to 2600. It is live on iOS and Android, priced at $5.99 a month or $39.99 a year on the US storefront, with prices varying by region, and a free trial to start on either plan.
We do not have Chessable's course library, its named author video content, or its offline mode, buggy as that mode's own history has been. We have not run a study measuring whether our retention scheduling produces better long term recall than MoveTrainer's ladder does; nobody has published one, and we are not going to imply one exists by omission.
Last verified: 29 August 2026.
Frequently asked
- What is Chessable?
- Chessable is an online chess training platform, owned by Chess.com since Chess.com's acquisition of Play Magnus Group closed 16 December 2022, built around interactive courses authored by titled players and coaches. Its library is put above 1,000 courses by third party counts, though those counts disagree and Chessable's own site could not be read to settle it, including named authors like Magnus Carlsen, Anish Giri, Judit Polgar, Sam Shankland and Peter Svidler, each turned into drillable variations through its MoveTrainer feature rather than pages you just read.
- What does Chessable's MoveTrainer actually do?
- Per Chessable's own help centre, MoveTrainer runs a fixed eight level interval ladder: 4 hours, 1 day, 3 days, 1 week, 2 weeks, 1 month, 3 months, 6 months, identical for every move in every course. A correct answer advances a move up a level; an incorrect one resets it back toward level one. Each move keeps its own independent timer, but the schedule itself never adapts to how hard that specific move actually is.
- How much does Chessable cost?
- PRO membership lists on Apple's App Store at $11.99 a month or $74.99 a year, with two independent secondary sources also citing a $114.99 two year plan. Individual, non PRO courses typically run $20 to $100 plus separately. We are flagging this as secondary rather than site verified pricing: chessable.com returned a 403 on every direct fetch during our research, so these figures come from Apple's listing plus corroborating third party pages, not Chessable's own pricing page read directly.
- Is Chessable worth it?
- For most people paying for it, yes. It holds a 4.8 out of 5 rating across roughly 5,900 iOS ratings, and around 4.5 out of 5 on roughly 14,000 Android ratings. Its 2.5 out of 5 Trustpilot score comes from only 10 reviews, too small a sample to treat as representative against nearly 20,000 combined app store ratings pointing the other way. Where it is documented to fall short: chronic offline mode bugs reported across multiple years on its own community forum, slow support response in some user reports, and, per an independent review, less value per hour than tactics or rated games for players under roughly 1400.
- What are the best Chessable alternatives?
- Depends what you are replacing. For tactics and opening drilling: Chess Tempo, Chess Position Trainer, CT Art, Listudy. For move explanations rather than just a score: DecodeChess. Free and open: Lichess for play and study tools, Anki for players who want to build their own flashcards by hand. ChessAtlas is a direct paid rival built specifically on FSRS scheduling. Chessdrive, our own app, replaces the authored course model entirely with a repertoire generated from a short taste questionnaire, reviewed by Stockfish and played against on device by Maia3.
- How is Chessdrive's scheduling different from Chessable's MoveTrainer?
- MoveTrainer runs one fixed eight step ladder for every move, with a full reset to the bottom on a miss. Chessdrive computes a separate desired retention for every move, from 0.80 to 0.97, based on how often that move comes up in your repertoire and how likely a player at your rating is to find it, using ts-fsrs 5.4.1. A miss recalculates the card's memory stability rather than resetting it: a card sitting on a 521 day interval that gets missed returns in 18 days, not at the bottom. This is FSRS behavior under our own parameters, not a measured claim that our approach retains moves better than Chessable's, which we have not tested head to head.
- Who should buy Chessable instead of using Chessdrive?
- Anyone who wants a specific grandmaster's own repertoire, explained in that grandmaster's own words with video walkthroughs, should buy the Chessable course. Its library of titled author courses, counted above 1,000 by third parties, is a real product we do not build and are not trying to. Chessdrive is the better fit if you want a repertoire generated for you rather than sold to you, an opponent that plays like a real rating band instead of a fixed line, and per move scheduling that adapts to how hard each move actually is for you.
Curious what your own repertoire's retention curve looks like?
Chessdrive builds a repertoire for both colors from a short taste questionnaire, then reviews each move against a retention it computes from your rating and how often that position actually comes up. Free to start on iOS and Android.