The Great Old Sodbury Fudge Disaster

I fancied some fudge. It seemed simple enough. Sugar, butter, and condensed milk in a pan, a bit of stirring, and you’re off. How hard could it be? Harder than it looks, as it turns out. What I ended up with was less “melt-in-the-mouth indulgence” and more “runny disappointment in a tin.” A sort of sugary soup. Tasty, yes, but about as sliceable as custard.

Letting runny fudge sit won’t fix it. Water doesn’t just politely evaporate from a pan of cooling fudge. It lingers, keeping the sugar concentration too low for setting. If it didn’t hit the right temperature the first time, it never will without a return trip to the hob. You can’t fudge fudge.

Undeterred, I consulted the internet. The smug consensus? I’d failed to hit the magic number: 115 degrees Celsius. Fudge-making, it turns out, is less about intuition and more about precision – sugar concentration, water evaporation, soft ball stage. It read like GCSE chemistry, but with more chance of heart disease. The lesson was clear. I needed a sugar thermometer.

Amazon obliged. Next day, armed with my shiny new instrument of confectionery righteousness, I scraped my failed fudge back into the pan. I’d like to say my troubles ended there, but no. Foolishly, I’d lined the tin with greaseproof paper, blissfully unaware that “greaseproof” doesn’t mean non-stick. When hot sugar meets greaseproof paper, it bonds like concrete. Peeling it off the cooled fudge was like trying to undress a Velcroed toddler. Baking parchment, by contrast, is coated with silicone, meaning the fudge lifts out cleanly. Why greaseproof paper still exists for home use is anyone’s guess. It’s about as fit for purpose as a chocolate teapot.

This time, I watched the thermometer like a hawk, stirring as the numbers climbed. Ninety. Ninety-five. A hundred. The mixture bubbled and thickened, but nothing revolutionary seemed to be happening. It wasn’t until I hit 115 that I realised the point. It’s not magic. It’s maths.

At 115°C, the sugar solution hits around 85% sugar and 15% water. That’s the Goldilocks zone – just enough moisture left to keep the fudge creamy, but not so much that it stays liquid. Anything less and you’re still in syrup territory. Anything more and you’re heading into brittle toffee land. The temperature itself isn’t the goal. It’s just a proxy for sugar concentration, because once the water boils off, the syrup heats faster. No structural alchemy, no caramelisation, no Maillard reaction. Just evaporation until the sugar takes charge.

Silicone baking trays seemed like a clever alternative, but they’re more trouble than they’re worth for fudge. The flexible ones, at least. When you try to cut the set slab, they bend and buckle, leaving you with ragged, uneven chunks instead of neat squares. They’re fine for muffins, but fudge needs structure. I’d used one of those silicone moulds designed for six small cakes, and that worked well enough. Each cavity held its shape, and the fudge popped out cleanly. Still, nothing beats a rigid tin lined with proper parchment if you want clean, satisfying slices.

Even with the right temperature, the job wasn’t done. Once I hit 115 and took the pan off the heat, I had to let it cool – not until it “felt about right,” but down to 43°C, the sweet spot for beating. Stirring too soon makes the sugar crystals form too quickly, giving you gritty fudge. Leave it too late and the mixture thickens like concrete before you get it into the tin. I waited, thermometer in hand, until the mercury dipped to 43, then beat the living daylights out of it. That’s when it thickened, lost its glossy sheen, and started looking like sand rather than a sticky mistake.

My real mistake was reheating Day 1’s attempt. There was precious little left, and heating it resulted in a slight charring of the bottom. Not disastrous, but it was rather grainy and left an aftertaste somewhat redolent of burnt tyres. I'd invented a new flavour. I should have started fresh.

Which I did, a few days later. This time, I applied the heat sparingly. It’s not the intensity that brings it to 115 – it’s the water fraction reducing. Slow heating over 20 minutes or so does the job without burning the sugar.

When it comes to beating, don’t, whatever you do, use a whisk unless it’s very open, or you'll end up with a ball of immovable fudge in the middle of the whisk. I made the mistake of beating too soon again and the result was admittedly a bit grainy. As it was cooling, I thought that it would too solid to beat at 43 degrees and did it around 60 degrees.

Pan choice matters, too. If you’re using a large pan, the thermometer registers the heat at the bottom. For a single can of condensed milk, a deep, narrow pan works best, so the thermometer’s business end sits halfway up the side. Alternatively, if you’re set on a wide pan, double or triple the batch.

When pressing the fudge into a mould, use a silicone spatula. The fudge will stick to anything else like glue.

So, if you’re thinking of making fudge and reckon you can wing it, don’t bother. Get a thermometer. Boil the mixture slowly and don’t be tempted to boil the arse out of it. Avoid wide pans unless you’re doubling up. Use parchment, cool to 43 before beating (not with a whisk), and steer clear of anything labelled “non-stick” unless it’s been tested under battle conditions and made of silicone. Fudge doesn’t care how confident you feel. It only cares about the numbers.

Here's the final result – a tad dark, as I'd used muscovado sugar.

It certainly held its shape and was much smoother than the previous batch. Not perfect, but I'm getting there.

Next time I make any I will invest in a marble slab and knead the fudge, rather than beating it. Kneading at 43°C often yields superior results – smoother, less grainy, and more consistent. It’s easier to judge when the texture is just right, compared to the more aggressive approach of beating with a spoon. If you’re after indulgent, creamy fudge rather than crumbly tablet-style squares, kneading wins hands down, apparently. There is (or was) a shop in Bath where the stuff was kneaded in the shop window. Bath Fudge - with added Bath water for that authentic, slightly Roman flavour. 

Just as an aside, regional fudge is only designated as such because it uses local cream and butter. Most mass-produced fudge uses butter from anywhere, and the condensed milk almost certainly isn’t local. Read the label. Even then, the amount of cream and butter is tiny compared to the sugar and condensed milk. It should really be called West Indian fudge, regardless of where it’s made.