Is the EIP-1559 burn rate on Ropsten representative of the expected rate when deployed to mainnet?

Watching watchtheburn.com it looks like the testnet burn rate is around 4,000 ETH a day.

From etherscan's total supply chart https://etherscan.io/chart/ethersupplygrowth we can see the total supply increases about 20,000 a day.

This means, if what we see on the testnet is what we might expect when EIP-1559 is deployed to mainnet, we would still see a daily increase of about 16,000 ETH a day.

ETH 2.0 and POW to POS conversion aside, this seems to show pretty starkly that this new protocol will not cause a decrease in ETH supply over time.

I'm sure I'm missing some pieces to this whole picture, can anyone help enlighten me to any other aspects of this I may be overlooking?