Here is the thing nobody selling you a chain wants to lead with: your drivetrain has already chosen your chain.An 11-speed bike takes an 11-speed chain. A 12-speed bike takes a 12-speed chain. Get that wrong and nothing else on this page matters — the chain will shift badly, wear fast, or simply not fit between the sprockets. Get it right and you have narrowed six options down to two or three before brand has entered the conversation at all.
That constraint is unfashionable to write about because it makes the roundup shorter. It is also true, and it is the single most useful thing on this page. So: count the sprockets on your cassette. That number is your filter. Everything below is what’s left after it.
The number Shimano publishes and nobody quotes
Shimano runs a public product information database, and in it every chain carries a figure called average weight (g) at 114 links. It is a genuinely comparable number — same link count, same measurement basis, published by the manufacturer, updated with a version and a date. We read it directly, and it produces a result that the rest of this category has apparently never gone looking for.
The Ultegra chain (CN-HG701-11) and the 105 chain (CN-HG601-11) both weigh 257 g at 114 links. Identical, to the gram, in Shimano’s own table. One of them costs meaningfully more than the other. The published difference is a SIL-TEC coating on the pin link plate, and Shimano prints no wear figure, no durability figure and no efficiency figure that would let you decide whether that coating is worth the money. That is not us being cynical about Ultegra; it is us reading the spec sheet and reporting what is and isn’t on it.
For contrast, the 12-speed XT chain (CN-M8100) is published at 252 g, and the top-tier CN-HG901-11 at 247 g with hollow pins and SIL-TEC on the plates and the rollers. So the entire published weight spread across Shimano’s road-relevant chain range is about ten grams. Ten grams. If you are choosing a chain to save weight, you have chosen the wrong component to think about.
What the manufacturers refuse to publish
Shimano publishes weights. KMCdoes not — neither the X11 nor the X12 page carries a weight or a link count, though both publish pitch and compatibility. SRAM publishes no weight for the PC-1110 that we could retrieve. So three of the six chains here cannot be compared on weight at all, and we are not going to fabricate the missing cells to make a tidier table. The What the makers publish table below shows the gaps as gaps.
Nobody — not Shimano, not KMC, not SRAM — publishes a wear rate, a service life, or a friction figure for any chain on this page. That is the number you actually want, and the entire industry declines to print it.
The wear data that exists, and who owns it
There is one substantial public dataset on chain wear: Zero Friction Cycling, which publishes chain efficiency and wear-life testing and states it has run over 300,000 km of controlled testing. Zero Friction Cycling also sells chains and lubricants. Its own site says the core products that reach its store “will have undergone the world’s most exhaustive controlled testing”, and its chain-testing page states the programme exists in part to assist “ZFC in what products to stock / avoid”. A testing operation that retails the winners has a commercial interest in its own results. That is not an accusation — a shop that only stocks what it rates is a defensible way to run a shop — but you are entitled to know it before you weigh the data.
Competitors in this category describe ZFC’s work as independenttesting. It is rigorous, it is public, and it is far more than we could do. It is not independent. We will keep citing it, we will keep linking its own pages so you can read the raw material yourself, and we will keep saying this every single time. There is also a public methodological critique of ZFC’s testing published by Hambini Engineering; we could not retrieve that page, so we neither characterise nor adjudicate it — we note only that the dispute exists.
For this page specifically: ZFC does not publish retrievable figures for these six exact models, so no pick above is driven by a ZFC result. The ranking rests on published specs, published compatibility, and live price.
The chain is the cheap part
A chain is a wear item that protects two expensive things. Let it stretch past its limit and it starts reshaping your cassette teeth, and then you are buying a cassette as well. When to replace a bike chain covers the 0.5% and 0.75% thresholds and what ignoring them costs. Whatever you buy here, the habit beats the badge: a mid-range chain measured monthly and replaced on time will save you more money than the best chain on this page ridden until it grinds.
The same logic applies to what you put on it — see the best bike chain lube, and how to clean a bike chain, because every wax product assumes a chain that is genuinely clean.
Speed count first, and it is not negotiable. Count the sprockets on your cassette. Eleven means an 11-speed chain; twelve means a 12-speed chain. Chains get narrower as the count goes up, and a chain that is wrong by one speed is wrong permanently. This one constraint eliminates more than half this page for you, and it does so before you have thought about brand at all.
Then check whose 12-speed you have.Shimano 12-speed and SRAM 12-speed road are not interchangeable: SRAM’s own support documentation lists road Flattop chains for Force and Rival AXS. If you have SRAM AXS, none of the 12-speed chains on this page is the chain you need, and we would rather tell you that than sell you one.
Link count is a length, not a quality.The listings publish 114, 118 and 126 links. You cut a chain to your bike — longer chainstays, bigger sprockets and big chainrings all need more. More links than you need costs you nothing but the offcut; fewer links than you need makes the chain unusable. If you run a large cassette or a long frame, favour 126.
Weight is the wrong reason to choose.Shimano’s published range across these chains spans about ten grams. Ten grams is a mouthful of water. If a shop is selling you up the range on weight, they are selling you the least significant published difference on the spec sheet.
A quick link is worth more than a coating.Three of these chains ship with a connecting link and three don’t mention one. A quick link is the difference between a roadside repair you can do and one you can’t. That is a real, usable feature with a published presence — unlike the durability claims nobody prints.
Buy the chain checker before the chain.The chain is cheap; the cassette and chainrings it destroys are not. Measuring wear on a schedule is worth more than any difference between the six chains here — when to replace a bike chain explains the thresholds and how to read them.
