Thrust on a stand is not the same as thrust on the plane. Some planes, with fat cowl, will reduce the effective thrust because the area of air that's blowing onto the cowl immediately turns into drag. Here's a little equation I came up with to estimate real thrust:

Find the prop disk area: A = (Diameter/2)^2 * Pi
Calculate the cowl profile area, say C, as viewed from the front.
Effective thrust = Bench thrust * [(A^1.5 - C^1.5) / A^1.5]
The term on the right is the approximate efficiency factor

For example, 19x8 Mejzlik would have A = (19"/2)^2 * Pi = 283.5 sq in.
Estimate the Extra 300 XS' cowl profile area to be a rectangle of 10" * 8" = 80 sq in

Effective thrust = 23 lb * [(283.5^1.5 - 80^1.5)/283.5^1.5] = 23 lb * 0.85 = 19.55 lb

So a 19" prop on a 10"x8" cowl would have efficiency factor of about 85%. Of course, a more streamlined cowl would have higher efficiency than a brick wall cowl, way, the Sukhoi's cowl.

So when we use bench thrust or calculator thrust to predict its performance on a plane, take it with a grain of salt, because we will lose some due to the cowl drag. That's why profile planes and pattern planes tend to require less power than the scale planes to get the same performance-per-weight. Hope this helps.