Can you source me the results you're using please, I can only find 3 games with the VII being compared against the 2080. Battlefield 5 (DX11, 1440p), Forza Horizon 4 (DX12, 3440x1440) & Strange Brigade (Vulcan, 4K). I'm pretty sure you can't be coming to this conclusion based on those three results.
https://www.eurogamer.net/articles/...-first-radeon-7-benchmark-results-in-25-games
You can also look at the claimed performance increases over Vega64, and find this average also puts it behind a 2080
https://www.techpowerup.com/reviews/Zotac/GeForce_RTX_2080_AMP_Extreme/31.html
At 4K, V64 is at 65% and stock 2080 is at 94%. AMD claims R7 is 29% faster on average than Vega64, so 65*1.29 = 83.85 , 83.85/ 94 = 89.2% of a stock 2080 performance, or over 10% slower.
When compared to the higher clocked 2080 models available, well you are at 83.5% of the performance of the review Zotac model.
As I said, 10-15% slower than the 2080 for the same money, less features, but more vram.
Another aspect is AMD claims the 7nm process offers 25% more performance at the same power. Power usage is likely to be the same as it is already at 295w. This performance comes form increased clocks. R7 has no new features or architecture difference that would significantly help beyond bandwidth. On the flip side, although R7 might have gained around 25% on clocks, being a salvage chip there are only 60CUs, so 92.75% of the compute per clock. Which means in games that arn't bandwidth limited performance increases will often be under 20%. Conversely games and settings that are bandwidth limited might enjoy the 30-40% gains reported. If you are not running 4K maxed then you wont see these benefits. Which is why AMD's numbers come form 4K max settings where the Radeon 7 actual fails to maintain a 60FPS average.