The extended barrage range the Lefh gains at vet 3 doesn't affect the counterbarrage ability, which is not informed towards the player. It stays the same as a vet 0-2 Lefh while still showing the longer arc on the map.
The counterbarrage range of a vet 1 Pwerfer is the same as a vet 3 Pwerfer (33% more than normal), this is also not informed towards the player.
Indeed, the good measurement accurary on the pen chances suggests that the real miss and scatter-hit chances are fairly accurate as well.
While the test will always work out in favor of the Churchill (since StuG always hits), I did not think that it would be that severe, especially since the StuGs model is not that big.
It surprised me too.
I guess it's partly because the Churchill has decent scatter. I forgot how it works exactly, but the Churchill's vertical scatter is 6, while most generalist tanks have around 6,5-7,5. Besides that it has a scatter offset of 0,3, which means the scatter area is displaced 30% further, while most generalist tanks have a scatter offset of only 0,25.
I posted it purely in the context of, on paper discounting scatter, the StuG and MK VII take similar shot counts to kill each other. Which is a vague point I did not originally bring up.
Somebody said it was BS and I laid out the numbers as to why it is not.
That was it. That was the whole context of my post. The only conclusion I put out about it was that a pair of StuG, who cannot miss so scatter doesn't matter, are ample to counter a Churchill. Which they are.
You did defend Hannibal's "vague point" by posting your calculation which confirmed what he said, while only mentioning scatter as a sidenote. You didn't give your own conclusion, but supported his, comparing a single Stug to a single Churchill, not a pair.
My point is, you can't properly compare the two just with hit chance if 57,5% of the shots the Churchill fires (read: the misses) are affected by scatter.
Otherwise agreed that two Stugs should be enough to hold back a Churchill.
I thibk it should be clear that the theoretical calculation does not account for scatter and is therefore somewhat flawed, but it is a way better and more comparable model thean just crying 'XYZ OP!'.
Testing is better if you have a sufficient sample size, and the theoretical calculation can give you an idea of how big your sample size needs to be to get reliable results.
It's more than somewhat flawed, to the point the results are close to useless.
In my ingame test the shots that hit & bounced the Stug were 30,88%, which is reasonably close to the 29,17% the Churchill statistically has to bounce it. -> Good!
Meanwhile, 15% of the ingame shots missed, which is far better than the 57,5% chance to miss without accounting for scatter would suggest. -> Bad!
I think we can agree in this case my small sample size gives far more accurate results than comparing the two statistically while ignoring scatter as a factor.
like how you just casually ignore this. You cant just read some statistics and expect it to be identical ingame.
Do you know the exact penetration of the Stug at range 40 or how much scatter influences the hit chance of the Churchill? (Well the first one can be calculated I'm sure, but laziness doesn't help )
Best way to know the shots to kill is to let them face each other off sufficient times in cheatcommands on the test range map, making one of the two invincible each time.
For whoever is interested, I've tested this a few times, both at 40 range.
So 80 shots total for 12 kills, which means 6,667 shots on average to kill a Stug at range 40.
15% of the shots were misses, 26,25% of the shots were bounces (or 30,88% of the shots that hit were bounces, which sounds about right if it statistically has 71% chance penetrate), 58,75% of the shots hit & penetrated.
Some clarification: 13 shots corrected for accuracy and pen chance at the units max range.
Stug only needs 4 shots to die, but Churchill has horrible accuracy against a Stug and will miss at least half of its shots, while Stug will always hit the Churchill.
You can't simply correct if for accuracy, because each shot that misses becomes a scatter shot which has a decent chance to hit (Churchill has slightly below average scatter). Best way to know the shots to kill is to let them face each other off sufficient times in cheatcommands on the test range map, making one of the two invincible each time.
To add to my wishlist: I think the Ambulance should get a 50% damage reduction when in the base sector to make it slightly less vulnerable. The Ambulance is already very slow and the base layout doesn't help its cause.
The Ambulance only has 160 hp, while the Ostheer bunker has 480 hp for example. With this change the Ambulance would have 320 effective hp, but only when in the base sector.