You only basically calculate deprecation into loss, not the transfer value of player. This is how things work in bookkeeping everywhere. If a company invests in, say, a machine, cost of that machine is not loss. Instead they add the value of that machine to assets. But the machine loses value and this can cause a loss if it is not offset by increased revenue.
In football, usually player's value is deprecated during his contract. And many of our expensive players are still on their first contracts so their values need to be deprecated. On the other hand we will see quite a drop in deprecations because both Sigurdsson (~9m/a) and Tosun (6m/a) are removed plus their wages (and that of Delph as well, his value deprecation was only 2m/a).
But as has been said, FFP is calculated over three seasons and we made a big loss in the previous two. Last season should already have been much better since Walcott (5m/a deprecation + 5m/pa wages), (Bolasie, 6m/a + 4m/a) and we saved a lot of money on wages on James and Bernard (around 15m/a in total). Selling Digne also looks good in the books as his book value wasn't much anymore having signed in 2018 (book value has to obviously be deducted when a player is sold so transfer fee is not directly just profit).
But I suspect we will still be far from profitable considering how big the losses have been.