Transfermarkt is run by a network of German volunteers. They admit themselves that it's all very subjective...
https://www.theguardian.com/footbal...ing-on-transfer-valuations-made-by-volunteers
Using understat.com to dig deeper shows that the underlying numbers aren't wildly different. We're talking about a drop from 9.4 shots per game to 8.8 shots/game. The number from open play is almost identical (7.0 this season, 6.8 last) but it's non-shot set pieces where we're struggling (ie indirect freekicks and corners) - down from 2.5 shots/game to 1.6.
If you look at xG (as good a measure of quality of chance as we have available) then you can see that the
quality of the chances we're creating this season is the problem. We're going from 0.14 NPxG/shot to 0.10 in both open play and from corners. That might not seem a lot but what it means in practice is that we should expect a goal every 10 shots this season compared to one every 7 last. When we had such fine margins last season it's no surprise that the regression to the mean we've seen this season has hurt us so badly.
There is an issue with the lack of finish - same as last season - but the issue is the creation of
good chances, particularly from open play.
On a player note, the NPxG/90 mins vs goals/90 for our strikers is as follows (again, taken from understat) and shows that we have an issue with all of our strikers except McGoldrick.
McGoldrick 0.29 xG/90 vs 0.31 goals/90 (Didzy is performing slightly better than expected)
McBurnie 0.27 vs 0.12 (McBurnie is considerably worse)
Sharp 0.22 vs 0.15 (Sharp is considerably worse)
Burke 0.21 vs 0.10 (Burke is considerably worse)
Brewster 0.20 vs 0.00 (yet to score... if he plays 90 against Liverpool and scores he'll move to 0.11)
Sheffield United detailed xG stat for the 2020/2021 season. Data for players, different formations, situations, game states and etc.
understat.com