It is a very straight sleeve, and when drafted out according to the instructions, very wide. The hem proportion in the draft is quite small especially in comparison to the width at and above the elbow.
In the instructions they calculate 1/4 of the Ad in two different points. Point C and F in the picture on the right.
I took that to be the armhole width from the body pattern but it wasn't specific as to whether that meant just the basic armhole width proportion of 1/8 chest plus 3.5 cm, or 1/8 chest plus 3.5, plus 2cm more, which is what you use in the drafting of the body.
I decided to take it as just the proportion (which I suspect is wrong) and when I pinned it into the jacket it did look too long.
The photo on the left (above) is the first try at it - very wide -and the photo above in the middle is a modification of the calculation for the width.
So, in the first I used 1/2 armscye measurement plus 1 cm and in the second 1/2 armscye minus .5 cm. The wider sleeve ended up having 6 cm of ease in it- as compared to the armhole- which I felt is way too much (that can depend on the fabric, style, and making up methods) The narrower version had a more reasonable 3 cm. Maybe using 1/2 armscye would be best.
Of course it is impossible to shape a muslin sleeve, so here it is pinned into the jacket as is.
I'd prefer a narrower elbow, and an altogether shaplier sleeve, but to take out another cm at the front at the elbow level might be a problem. The forearm seam here is displaced towards the undersleeve (as in most modern sleeves) which limits the shaping you can put into the pattern unless you are able to work the fabric with the iron.
The other option is to increase the sleeve hem circumference at the front, which I think would improve the shape in general- don't you think it is a bit too straight and abrupt there?
You can see that the back sleeve seam is too long, so I think the calculation of 1/8 chest + 3.5 + 2 is better and that decreases the back seam length by .75cm by dropping point C circled above.
That also raises point F slightly if the calculation is consistent.
I guess that means another try.