Why Your Bread Scores Close Up in the Oven

A plant manager called me a few years back about a rustic loaf that had stopped opening. Same formula, same oven, same operators. The scores went in clean and came out of the oven nearly invisible, just a faint pale line where a proper ear should have been. He had already reformulated twice.

The formula was never the problem. It almost never is.

Scores that close during bake are one of the most common calls I get, and after forty years of building and rebuilding scoring machines I can tell you the cause is usually mechanical, usually cheap to fix, and usually something the bakery has walked past a hundred times. Here is how I work through it.

What a score is actually doing

Before troubleshooting, it helps to be precise about the job the cut is doing.

When dough enters the oven, the gas inside it expands fast. That expansion has to escape somewhere. A score is a deliberate weak point in the surface tension of the loaf, and its entire purpose is to give that expansion a preferred direction. Cut it properly and the loaf opens along the line you chose. Skip it, or cut it badly, and the dough finds its own weak point, usually a side seam or the base.

So when a score closes up, what you are really seeing is a cut that failed to become the weakest point on the loaf. Either it was never weak enough to begin with, or the surface pulled itself back together before the oven could act on it.

That gives you three places to look.

One: the cut was too shallow

This is the first thing to check because it is the easiest to change and the most common cause.

A shallow cut scratches the surface without genuinely breaking the tension underneath. It looks like a score on the bench. It is not one. Under oven heat the dough swells, the surface stretches, and that shallow line simply smooths itself out.

The fix sounds obvious, but the trap is that most bakeries over-correct. They go deeper, the loaf spreads or collapses, they pull back too far, and now they are chasing the setting up and down every shift.

Change depth in small increments and read the baked result each time, not the raw dough. The raw score tells you almost nothing about what will happen in the oven. What matters is whether the baked loaf shows a genuine ear with a torn, slightly ragged edge. That ragged edge is what a real cut looks like after it has done its job.

Two: the dough was too slack to hold the opening

If depth is right and the score still closes, look at proof.

Over-proofed dough has lost the structure it needs to hold a cut open. The gluten network is already stretched near its limit, the dough is slack, and when you cut it the edges relax and flow back toward each other almost immediately. You will sometimes see this on the bench if you watch a scored piece for thirty seconds. The line softens and narrows before it ever reaches the oven.

The same thing happens for a different reason on high-hydration products. Wetter dough flows more. That is the whole point of it in terms of crumb, but it works against you at the score line.

Two useful checks:

• Score a test piece slightly earlier in the proof cycle and bake it alongside a normal one. If the earlier piece opens and the normal one does not, your proof has drifted.
• Watch whether the problem is worse at the end of a run than at the start. Proof time creeping as the line backs up is one of the most common invisible variables in a bakery.

This is also why I ask every customer whether they are scoring after proof. It is not a throwaway question. The answer changes what the machine has to do.

Three: the blade dragged instead of slicing

Here is the one that gets missed, and it is the one I would bet on if the first two check out.

A sharp blade slices. It parts the dough and leaves two clean edges that stay where you put them. A dull blade, or a blade loaded with built-up dough, does something different. It drags. It pushes the surface ahead of itself, stretches it, and then releases it. The dough springs back and the two edges of the cut move toward each other. You get a cut that looks acceptable on the bench and closes in the oven.

The tells are worth learning:

• The score edge looks smooth and rounded rather than crisp
• Dough is visibly building on the blade faster than it used to
• The cut is inconsistent along its own length, opening in places and closing in others
• You can hear the difference. A dragging blade is quieter and duller than a slicing one

On wet or enriched doughs, build-up is the usual culprit rather than genuine dullness. Dough sticks to the blade, the effective edge gets thicker and blunter with every pass, and the cut quality degrades through the shift even though nothing about your process changed. This is exactly the problem a Teflon coated scoring blade is meant to solve. The coating slows the build-up so the blade keeps a working edge for longer between wipe-downs.

If you are running uncoated blades on a sticky dough, that alone is worth testing. Run one coated and one non-coated blade side by side for a week and let the baked loaves settle it.

When it is the machine, not the blade

There is a fourth cause, and it shows up in a specific pattern worth recognising.

If the score closes on some positions across the pan but holds on others, stop looking at dough. Consistent problems point at dough and depth. Inconsistent problems across the width of the pan point at the machine.

What you are seeing is depth wandering, and the usual causes are mechanical wear rather than the blade itself:

• A worn scoring blade hub letting the blade run slightly out of true
• Play in the shaft hanger or its bearing, which transmits directly into cut depth
• Blades at different wear stages across the same shaft

That last one catches people out. If you replace blades individually as they fail rather than as a set, you end up with a shaft carrying blades at four different stages of life, cutting four different depths. On a multi-up pan that reads as random inconsistency when it is actually completely predictable.

Bearings are the cheapest part on the machine and one of the most common root causes of wandering depth. If your scoring head has developed any noise it did not used to have, check there first.

A working order of operations

When a bakery calls me about scores closing, this is the sequence I walk them through:

1. Pull baked samples, not raw dough. Photograph them.
2. Check whether the problem is consistent across every position or varies across the pan. Consistent points at dough and depth; variable points at the machine.
3. Inspect the blade edge and look for build-up. Change it and re-run if there is any doubt. It is the cheapest test you can run.
4. Increase depth in small increments, reading baked results each time.
5. Test a piece scored earlier in the proof cycle.
6. If it is variable across the pan, check hub, hanger and bearing for play, and check whether your blades are all at the same wear stage.

Most cases resolve in the first three steps. The ones that do not are usually telling you something about the machine that is worth taking seriously, because a head that has developed play will keep drifting.

When to stop adjusting and call someone

If you have worked through that list and the scores still will not hold, the issue is generally that the machine is not right for what you are asking it to do. That happens most often when a product has changed since the machine was specified. A bakery moves to a wetter dough, or adds a product that sits above the pan lip, or increases line speed, and the scoring head that was correct five years ago is now being asked for something it was never built to do.

That is a fixable problem and it does not always mean a new machine. We retrofit and rebuild existing equipment regularly, and there are plenty of cases where the honest answer is that a blade change and a hub will get you there.

If you want a second opinion, send photographs of your scores after bake along with your pan size, and I will tell you what I see. The ten questions on our scoring page cover everything I would need to give you a real answer rather than a guess.