Can You Still Get a Grade 9? GCSE Maths 2026 Paper 1-3 Review & Predictions

So I get asked about the maths grade boundaries all the time, it's the most frequent question, so in this video I'm going to give you my predictions for what the grade boundaries will be like in August and exactly how I got there. Now before I start I should point out that these are predictions and there is no human being on the planet that can tell you what the grade boundaries will definitely be, but I can go through how the exam board decides on grade boundaries and show you kind of maybe so we can make a more informed prediction of what they could be. The first question I want to answer though is every year there's conversations about whether or not the grade boundaries are fair and the answer to that is yes by design. So the way that they decide maths grade boundaries is two main factors and these main factors basically ensure that the grade boundaries are fair. So the first one is that a grade nine student should get a grade nine in any exam sitting, so this is what they mean by grade boundaries go down when the paper's harder and they go up when it's easier, because if the exam's harder then of course a grade nine student wouldn't necessarily get a grade nine unless the grade boundaries are lower. Now this is true actually for all of the other grades as well, so if you passed on any sitting you should still pass on any other sitting. The other thing which is where we're going to encounter the biggest problems, and I'll get into that a bit later, is the same proportion of students getting each grade which is predominantly predicted on SATS data. So everyone that's SATS, they essentially get to predict a grade for the end of year 11 and that is used in order to help us decide if the grades are going to be higher or lower. It also shows us if the general group of students is stronger or weaker as a country, which it does change year on year, but I'll get into why this is a bit difficult in a moment. So let's start with difficulty because that's the easy one I can get through. So I've gone through the 2026 papers again, I've done them all myself as well, and I looked back through 2025 because I think that's the easiest one to compare it to, and 2024, so the two most recent sets of exams. So with paper one, actually sorry, before that, just as a bit of evidence as to why I might be slightly more correct, is I've been tutoring SATS teaching maths since, well for 10 years now, so that's 2016. I remember specifically going through the 2022 exams as they came out, so literally when paper one did it, the day after I did the paper one as well, and so on and so forth. Now I'm pretty sure I did this all the way back in 2016, but I can't remember so I didn't put on the slides. Now the reason why this is important is every year when the new papers come out, us maths teachers look at the new styles of questions that the example gave, and then we teach it to our next year 11s or year 10s. So this is probably why you found 2025 and 2024 relatively easy, because your teachers taught you those questions in their slides when they're teaching the topic. However, when a paper first comes out, obviously the questions are very different, and so you will find naturally 2026 to be slightly harder regardless. Now because I've done these as they've come out, I can remember how difficult they were as they came out, so that might help a little bit. So paper one, I think it was genuinely a really really standard paper. The only two weird questions were, for example, the histogram question, and of course the proving the circle theorem question. However, with the circle theorem proof, some people just got lucky. They happened to look at it the night before, they happened to look at it that morning, and they will get full marks, so that's going to skew the grade boundary slightly, but with the histogram question, that was very weird and different. If you can't remember, it was one where it said something like, you know, there are five more people between 10 and 12 than there were between four and eight, and then find the probability that, insert whatever. So overall, a fairly standard paper, and it really wasn't that difficult. Paper two, again, it was fairly standard overall, but there was a lot more weird worded questions, and the numerical rearranging was quite strange. So for example, this question at the end about the pressure, like something was resting on a table, and then you had to get a cert, which is quite difficult, quite weird. But again, in terms of difficulty, if you really knew what you were doing, it wasn't that bad. But the way they worded questions, made it odd, made it strange. So that's going to increase the difficulty of paper two compared to previous years. Now paper three, very odd and very tricky. And instead of just giving you one difficult five or six marker, like they have in 2024, for example, they just gave you many, many, many questions that were all quite weird, even though there were only three or four marks. Overall, I think that makes the paper slightly harder on average, because it means that the total number of marks of difficult questions was greater. Now when I talk about these weird questions as well, you should know that they are almost always problem solving questions, which is the type of questions that I like to go through on the channel, because they are the things that people struggle with. So depending on how much practice you have with problem solving, you would have either found paper one, two and three, quite easy, or quite difficult. Now from speaking to my own students, that I tutor or at school, which is quite a few students this year, paper one, they found easy paper two, they found easy, but there was weird numbers and algebra. So they were slightly less sure paper three through a lot of them, like a lot of them shockingly as well. These are the grade nine students that said that they got 75 plus in paper one and two, paper three, they're not sure if they've even got 70, which is a pretty big deal. Because these students, again, they're the students that know how to do basically any question that can come up. Right now proportion here's where things get very, very difficult. Here are SAT scores. So 2019 was the last year that we did that we had SATs data, where the relative score was 104 for maths. Now for 2020 and 2021, it went higher, but there were no SATs. And so it's teacher assessed. And so pretty much everyone agrees that that was inflated. So it's not actually higher, it's probably the same, if not lower. Now in recent years, again, the average scores are one point higher, which for those of you that aren't aware of how SATs works, every one above 100 means you're 1% better than the average. So not really much of a difference. But again, this might be inflated. And this is a bit annoying, because the exam boards themselves with GCSEs want to return to pre pandemic pandemic pandemic standards. So yeah, we're basically looking at a group of students that are predicted higher. But then the exam boards are trying to become stricter in their marking as well. And so overall, I'm not confident in saying whether or not this will cause it to be higher or lower. I am still going to give you my predictions, by the way, but it's mainly going to be based on difficulty rather than SATs data, because we just don't really have enough data for So for context, here are all of the grade boundaries that have come up since 2017. Again, you can have a quick scan through you can see there's quite a bit of variation, you know, like for example, a nine has been 187 on the low end, and 217 on the high end. And you can also see that between years, there's been quite a big jump. So for example, 2024, June 197 for a nine all the way up to 217. That's 20 marks. That's pretty huge. That's almost 10%. So yeah, I will say that the average at the bottom is a little bit useless just because each year has so much variation in the in the grades. But you can use as a kind of rough guideline. So I'm going to be mainly using the 2025 June sitting now wide. Why am I ignoring the Novembers? Well, the only people that really sit in November series, there's two types of people. One is people that are re-sitting that didn't do as well in their previous exam. These students tend to be weaker, which means the grade boundaries do lower because of that, because they just do worse. The second type is people sitting it externally. So if you want in the future, I don't know why you would, but if you wanted to, you could sit GCSE maths again in November, right? So it may be in November 2030, you could just sit GCSE maths, whatever. And again, those students, because they haven't had two years of a teacher teaching them tend to do a little bit worse. And so that's why in November, the grade boundaries tend to be lower. It's not always true, of course, because it depends on paper difficulty as well. But that's why I'm going to ignore that because everyone that sat them in June, I'm pretty sure that 99% of you are just in school. So we're going to compare them to this. Here's what I think. Now, again, might be a bit much, but it's going to be 210 for roughly for a Grade 9. Now, again, I'm just going to show you what it was in 2025. It was 217. So I'm predicting the grade boundaries will go down by a little bit. So seven marks is roughly two and a bit marks per paper. Again, I think that might this again, this might be a bit generous, if I'm being completely honest, maybe it's a bit higher. But again, this is just what I came up with. Then a Grade 8 will be about 30 marks lower, a Grade 7 30 marks lower. Again, I know that these numbers look very nice, which is a bit suspicious, but then they start to change because the differences between grades actually differs slightly from year to year. And then Grade 5, 84 and a Grade 4, 50. And of course, anything below that is a three or a U. Again, maybe I'm being generous, maybe. I've been correct in terms of saying if it's higher or lower, but I've never actually done this where I've given them exact numerical values. I do think the grade boundaries will go down by a bit. But whether they go down by this much is probably a big question. And again, as I said at the beginning, no one knows what the grade boundaries will be. Who knows, maybe they'll even go up. But I think personally that they will go down. And again, just to reiterate, will they be fair? The answer has to be yes, because the same number of students get these grades each year. So it's roughly the top three and a half percent for a nine. It's roughly the top 10 percent again, eight and so on and so forth. The only one that doesn't apply for is the students that fail at the bottom. That can obviously change, which is why you're hearing the news about more people failing GCSE maths year on year. But let me know what you think. Maybe this is too generous. Maybe it's not not generous enough. Maybe you think the grade boundaries go up. Please do let me know what you think and we can have a proper discussion. But I will see you in the next one.