The Binding of Isaac: Rebirth

Four hundred and thirty-nine thousand reviews. Ninety-seven per cent positive. There is not much argument left to have about whether The Binding of Isaac is good.

So this review is about the argument its own veterans are having instead, which turns out to be one of the most interesting design questions in the genre it created.

What it invented

Isaac's contribution was items that talk to each other.

You descend through procedurally arranged floors shooting tears at increasingly unpleasant things, and each floor offers pickups: a modifier that makes your tears homing, a multiplier, a familiar that mirrors your shots, a trinket that changes how a stat scales. Individually they're small. Together they compound, and by the sixth floor you are frequently firing something the game has no vocabulary for — a wall of homing lasers, or exploding flies, or a single enormous tear that clears a room.

Nobody had done that at this scale in 2011, and everything since has borrowed it. The pleasure of a run isn't executing a plan, it's discovering what the pile of things you picked up has turned you into.

Around that sits an absurd volume of content: five expansions, hundreds of items, dozens of playable characters with different rules, alternate floor paths, multiple endings and an unlock chain that runs for hundreds of hours. The median recent reviewer has 37 hours; the corpus is full of people past a thousand. At NZ$17.99, the value question doesn't need asking.

The argument: too many items?

Here's the review that stopped me, from a player with 1,337 hours: talk to me five hundred items ago and the game was great — runs where a new synergy might dramatically change how things played. At this point, they write, there are simply too many items for an RNG game.

That is precisely the pool-dilution problem, and finding it here is significant, because Isaac is the game that made item synergy the genre's central pleasure. As the pool grows, the odds of drawing two specific items that transform each other fall. You get more variety per run and less combination. The game becomes broader and flatter at the same time.

I've now seen the identical complaint on two much smaller games this week — Dungeon Clawler's players and Luck be a Landlord's both report combos landing less reliably after content updates. The pattern is consistent: adding items to a randomised builder without adding tools to hunt for items makes it more random rather than richer.

Isaac has partial answers — shops, devil deals, item rerolls, planetariums — and they're why it handles the problem better than its imitators. But a thousand-hour player noticing the dilution is worth listening to, because they're the only people who can see it.

The information problem

The second complaint is sharper and easier to fix, and a 2026 reviewer states it perfectly: the most popular mod for this game exists to tell you what the items do.

Isaac withholds information deliberately. Pick up an item and you get a name, a picture and a cryptic one-liner. Working out what anything actually does is meant to be part of the game — and for the first thirty hours it genuinely is, because discovery is fun.

After that it's an obstacle. There are hundreds of items; nobody memorises them; and the practical result is that experienced players run an external wiki or a mod that displays descriptions in-game. When the community's most-installed modification is a fix for your interface, the interface is the problem.

To the developers' credit, a long-time reviewer with 2,822 hours notes that a later update added official in-game item descriptions. That's the right call, and it arrived about a decade late.

The rest of the objections

The expansions are divisive. A player of ten years with roughly 2,000 hours says the most recent major DLC changed the game away from what they'd loved. Negative reviews spike sharply in 2021 and again in 2026, which lines up with major reworks rather than with anything static about the game.

It's hostile. The most-upvoted recent negative, from someone who found it genuinely fun, is a long list of design decisions they consider punishing rather than difficult. Several players arriving from Hades or Slay the Spire report never reaching the "one more run" moment at all.

And the genre has caught up. A reviewer with 315 hours makes the fairest version of this: Isaac was their roguelike of choice for years, and having since played games that are both as polished and more carefully designed, it no longer holds the same place. That's not a criticism of Isaac so much as the cost of being first.

One practical note: expansion support on Linux has drawn specific complaints, with the base game supported and DLC reportedly not. Check before buying if that's your platform.

8.5 out of 10. Rated 8.5 out of 10.

Isaac remains essential. It built the vocabulary the entire roguelike genre now speaks, it contains more game than almost anything at its price, and 97% of nearly half a million people are not mistaken. If you've never played it, start with the base game, add the item-description mod, and discover why every synergy-driven roguelite since owes it money.

It sits at 8.5 rather than higher because two of its problems are structural: an item pool that has outgrown the odds of combining, and a design philosophy on information that its own community routes around with a mod. Neither stops it being one of the most important games of its decade.