I know it's weird to talk about potential sequels, but hear me out a bit in terms of expectation vs what reviewers have said about this for me.
Based on the gameplay previews, impressions and my own assessment of interview materials, my expectations for Horizon were generally the following:
Combat/Enemy/Visuals = 9 out of 10
World/Activity/Quest/Story/Other = 7 out of 10
In short, I expected a good game that is still constrained by a lot of generic trappings of the open world formula without notable standouts, but will be elevated by being best-in-class in terms of combat & enemy design, with all the technical mastery of visuals from the art team at GG.
And some of the reviews, by Kotaku UK & Eurogamer, actually echoed that sentiment. I think in terms of pure checklist of design, Horizon is that game. A game of standout 9s, supplemented by other parts that are 7s.
What I did not expect, was that Horizon would apparently have a story quality of a 9, and executed the world/activity/quest/other aspects with the finesse and balanced design where even though the individual quality of those quest are maybe still a 7 or so, the way its laid out and executed as a complete package elevates it to an 8 in terms of the experience.
Of course, not everybody will resonate with execution, so some people will still see it as a 7, no matter how nicely its packaged, but the fact that a vast majority of reviewers though that the execution was brilliantly handled to the point that its comparable with the best-in-class open world games is quite possibly the biggest surprise.
To me, that is one of the hardest things to do in any game. It is not difficult for a game that is composed of a bunch of [8/10] mechanics/design elements to come together as a strong [8/10] piece. But its really really hard imo, to take a bunch of [6/10]s and [7/10]s and execute the balance of all those components to make it more than the sum of its parts.
Because, from where I stand, even iteratively, I see a ton of areas of improvement for a Horizon sequel that are complete no-brainers and is realistically, quite easy to implement given another 2-3 years.
Another additional 10-15 unique variants of machines would go a long way to make a truly complete animal ecosystem. 26 is good, but given the dynamic nature of interactions,I think every additional 5 would be exponentially incremental by leaps & bounds in terms of dynamism.
An actual melee combat system.
Multi-piece armor equipment system ala RPGs like TW3/DAI/etc.
Bigger settlements/cities/enterable buildings.
Improved human enemy A.I.
Improved side-quest writing/design.
Etc etc
Horizon, even based off all the glowing reviews, isnt at the same situation like many highly lauded games like Bioshock, TLoU, Red Dead, Mass Effect 2, Uncharted 2, etc where theres a very specific design choice in every part of those game that it isnt an easy +1 on everything to make an easy sequel. We saw how the +1 mentality didnt work with Uncharted 3/Bioshock Infinite/etc.
As it is, there are clearly many parts of Horizon that has a lot of room for improvement on an iterative level. Its exciting to see where GG can take the formula further.