Crazy that the fees made such a huge difference to the balance! That's quite a substantial amount spent! Shocked
Yes...not cool. Do you think @HCP there was anyway to escape this "substantial amount"?
About the only thing you can do is use a wallet that allows the setting of custom fees and try to be more aware of the network status at the time you go to send a transaction...
https://mempool.space/ and
https://jochen-hoenicke.de/queue/#BTC,24h,weight give good overviews of the current state of the network and with a bit of practice you can determine a "decent" rate that will suit your purposes (ie. cheapest option, don't care about how long it takes to confirm vs. must be next confirmed in next block etc).
A lot of wallets try to simplify fees to things like "low", "normal" and "high"... but don't give any indication as to what those rates actually are (or mean)... and because the network can be quite volatile, you can end up over (or under) paying by a substantial amount!

I'd have to look at all the transaction details to see exactly why your total fees paid were so high, but the normal causes are:
- Using legacy UTXOs as inputs. Legacy UTXOs are much larger than SegWit UTXOs, so using older Legacy addresses will mean that when you go to spend the coins, your transaction will have a larger total data size. As the total fee is (size * fee rate), then logicially a larger total data size == higher total fee.
- Using multiple inputs to a transaction. Having multiple inputs (Regardless of SegWit or Legacy) will result in a transaction with a larger overall data size. As above, larger data size = higher total fee. This is usually the result of collecting a lot of small amounts of BTC from mining or faucets etc and then trying to spend them all at once. Can be somewhat mitigated by occasionally
consolidating your inputs when the network is not busy and the fee rates are low. That way, when the rates are higher... you'll likely only need 1 input and the smaller transaction size helps offset the higher fee rate required.