Surely this is a slightly goady question since you said yourself you didn't know anything and only chose to RF because of info you found online. Not everyone will find that info, and it is the norm to sit FF so the assumption is usually that once babies are out of the first seat, it's a switch to FF. From our own childhood experiences we probably also only remember FF seats for toddlers, unless your parent was a very early ERF adopter or you grew up in Sweden.
Two things have changed this in my observation. The rule on newer seats that babies must remain RF by law until 15 months, which usually overlaps with at least their first few months in the second stage seat, and secondly that so many seats are available on the market allowing RF up to 105cm (approx age 4). But these are recent shifts, only in the last 10 years or so.
I think everyone should make their own decision and being a travel sickness sufferer myself, I totally understand turning for that and would agree the benefit outweighs the risk if DC is older than 15 months, and more so the older they are.
Ultimately there absolutely will be scenarios where FF makes sense for a family over RF regardless of the safety benefits. I am not as zealous as some ERF advocates because statistically, using a seat vs not using a seat (including major mistakes so that it doesn't work) makes much more difference than RF vs FF.
However, out of everything else, RF vs FF does make the biggest difference, so it is worth knowing about. There is even an argument that RF can counteract some of the more minor mistakes with car seats e.g. straps being slightly but not hugely too loose. If your seat meets any legal standard, is properly used according to the instructions, and is RF, it will offer better protection than any FF seat, with the possible exception of Cybex Anoris.
Still, some myth busting 
forward-facing seats are usually significantly cheaper than extended rear-facing ones
Not really true any more - the most popular seats sold are spin seats which can go in both directions. They cost about £150-200. Purely FF seats are actually much rarer to see now.
True that the super cheapest possible seats for £60ish are forward facing only, but there are a couple of super cheap RF seats as well for under £100 - Halfords Impala and the Joie/Graco belt fitted ones.
You don't need a £300+ Swedish seat to rear face. Although they can be worth it if you want more leg room, have a tall child, or need more front to back space.
no leg room/tall child
The seats sold in Sweden, where RF is a cultural norm and which are legally approved throughout Europe including the UK, are designed for long legged Swedish children. They can be moved away from the seat back to make more leg room. In the cheaper seats, children still don't usually get to a point their legs are actually squashed until they are more like 2. And it is safe for them to cross their legs in the seat, rest them up on the back rest or over the sides of the seat, if they are comfortable like this.
most accidents are from being rear ended, therefore rear [forward?] facing is safer. Makes me think if I was in a head on accident, would rear facing be pointless
It's not true most accidents are rear shunts. Statistically at least one car has to have a frontal collision, even in a rear or side impact. But also rear shunts are the least dangerous type of collision, and in a RF seat the child is further away from the point of impact, which can be beneficial especially if the crash is so severe there is intrusion into the passenger compartment. Apparently the only way a RF seat would come off worse is if you reversed at speed into someone. Most reversing accidents happen while parking, at low speed. RF seats also come off better in side impacts, which is thought to be because drivers will instinctively brake when anticipating an impact, which pushes the child backwards into their seat (or forwards, out of a FF seat) and so their head is better contained by any head protection. Side impact testing as done in a crash test lab is usually done as though the car were stationary.
Anti rebound protection which stops the seat ramping back up in the rebound phase of a frontal crash, such as anti rebound bars or the tethers on a larger RF seat can also help against rear impacts.
If it isn’t safe then surely they should stop selling forward facing for younger children.
The age has been raised - it is now 15 months minimum rather than the older rule of 9kg minimum with guidance of 9 months. In some US states and I think Canada, the rule is 2 years. Personally I would prefer the law to be 2 years, but I do also agree that parents should be able to make the decision about exactly when. However, I don't think parents have access to good information currently as the advice tends to be very polarised, which is not at all helpful. One year olds and younger are particularly vulnerable to spinal cord injuries without head impact, which means that the actual construction of the seat has caused the injury, this is likely to do with their relatively large heads. As children get older, the risk of this kind of injury in FF seats reduces. Under 15 months, the risk is unnacceptably high, but personally I would not be keen to rush to FF as soon as they hit 15 months (barring severe difficulties with RF) as it is still higher than it would be for an older child. OTOH I think the older the child is the more other factors may play into the decision, and as said previously, using any kind of seat still makes a bigger difference than using a RF compared with FF seat.
what does "much safer" mean?
Not 5x. That was a study which was later withdrawn because the maths was wrong. It's difficult to give exact numbers. The best and most helpful stat I saw was that up to age 4, a child in a RF seat has approximately the same risk of injury or death as an adult wearing a seatbelt. A child in a FF seat has approximately the same risk of injury or death as an adult not wearing a seatbelt. Children who are unrestrained or inappropriately restrained are at much higher risk of injury or death than adults, because their bodies are smaller and less developed. This is averaged over 0-4 years. So children at the lower end of that age spectrum will be at much higher risk regardless of restraint type and so RF is much more critical for them. A 4 year old in a properly used FF seat is probably closer to an adult in a seatbelt than an adult with no seatbelt. But averaged out across the age group, it's a startling figure.
The other way we can compare it is looking at crash testing in labs and see the difference in numbers. The forces on vulnerable parts of the body (the head, spinal cord and abdomen being the main three areas focused on) are much higher in FF seats compared with RF, and for older children (4-8) much higher in seatbelts compared with car seats or booster seats. Roughly 3-4x higher loads to the child's spinal cord is seen in FF seats. That doesn't mean anything unless you know where the threshold for injury is. That's what the Swedish Plus Test aims to look at, so if you want to know that a seat is crash tested and the dummy experienced loads of less than the threshold which would cause spinal cord injury, you want a Swedish Plus Tested seat and they are all rear facing. Not all RF seats pass the Plus test but all Plus tested seats are rear facing. It is possible that Cybex Anoris (which uses an airbag) also gets numbers less than this, because they very quickly changed the test criteria when that seat came along to ensure that only RF seats were allowed to be tested, and that the seat has to pass without the aid of airbags which could fail to inflate, which I find a bit suspicious. However older data does show a clear difference in neck loading between RF and FF seats in general. This is an interesting read with some graphs: http://erfmission.com/crash-course-crash-testing/
The other way it has been looked at is long scale studies looking at real life crashes where children died or suffered serious injury in FF seats and using expert analysis and data from the crash scene (if recorded) to work out whether the injuries could have been prevented with use of a RF seat. The first study of this nature (in 2008) found it was likely to have been possible in about 3 out of every 4 cases and was part of the basis for increasing the FF age to 15 months (though the outcome of the study was to recommend 4 years), the more recent one (in 2024) found it was likely to have been possible in about 2 out of every 3 cases. This probably tells you FF seats have got better and usage has increased, rather than RF has got less protective. Again this refers to all children under the age of 4, and the span of accidents used as case studies tends to go back about 20 years because there just aren't that many fatal accidents involving children, thankfully, so the figures are always somewhat playing catch up. But it does still happen even in recent years with current seat models that children die or are injured in FF seats and a RF one would have helped. It's unlikely that FF seats will ever fully catch up to RF. But using a FF seat is still a good choice if RF doesn't work for some reason.
There will always be crashes where they are so severe no child seat would help. There will always be crashes which are so minor nobody is injured. The place where RF makes a difference is the group in the middle.
driving slowly
Some people have commented about "pootling around town" or driving at 20mph. This is still a fairly significant speed, it's not a bumper car (they drive at about 4-7mph for reference). Because velocity (speed) is squared when calculating impact, it's a larger difference than you'd think in impact force which is why a car accident at 20mph can cause injury whereas bumper cars tend not to - it's nearly 10x the amount of force (49 vs 400) and then a jump up to 30mph isn't 50% worse than 20mph, it's over twice as severe (400 vs 900). That is only an extremely rough calculation, because other numbers go into calculating the severity of a crash, but a helpful rule of thumb is if you're thinking about speed, always square it to understand the difference between different speeds in crashes.
All car seats legal for sale on the UK/EU market are crash tested up to 30mph, and this is not an easy test to pass. The speed is chosen because if you have designed a product which can withstand a crash at 30mph and protect the child occupant, it's a reasonable chance that it will provide at least some protection in a higher speed/impact crash. ADAC/Which?/NCAP test at a higher speed of 40mph and indeed the vast majority of child seats tested do well on this, although occasionally a seat will fail dramatically - often the extreme budget end of the market, although not always. They also clearly show better results from RF seats.
Will a FF seat protect adequately in slower crashes? Quite possibly. But don't assume that RF makes no difference at all - it still does. "Slow" crashes are still violent events which are hugely impactful on the body and car seats (and car safety features in general) are primarily designed around that kind of 20-40mph kind of crash, because they are the most frequent, because they are impactful enough for the results to transfer to higher speed crashes and because they best represent this "window" where the design of the car seat makes a difference to outcomes.