Tiger main gun nerf too much? (Winter patch)

Though I haven't had the time to do it for all allied tanks yet, this comparison for the IS-2 should outline the general trend at least...

Once again, thx for crunching the numbers. I’m a bit surprised that the S2K a 4 man squad is only 10% faster. From these numbers it seems like the IS-2 pretty good balanced across squad sizes.

One question though: When calculating T2K are you starting the time at 0 shots? If we disregard the time to rotate the turret then engagement basically starts when the first shot is fired. With that in mind, should the T2K not be counted after the first shot instead?

In order to get back to the main topic and also to weigh in to the discussion with more numbers:

While I haven't been able to generate any statistically meaningful in-game data yet (obviously), I've run some theoretical simulations to estimate how the incoming changes to the AOE profiles impact the anti infantry performance of the affected heavy tanks (details on the methodology can be found here).

The comparison is based on the number of shots (S2K) and required time (T2K) to kill a full-health, 6-man squad in either "wide", "stacked" or "clumped" formation (Fig 1) at ranges 5, 20 and 40 m.


Fig 1: Formations used in the simulation, numbers below each image indicate average (AV), minimum (Min) and maximum (Max) distance between entities

The resulting average numbers for S2K (Fig 2) and T2K (Fig 3) over 5,000 iterations for the live and WBP-versions of the Tiger, IS-2 and Pershing, as well as the King Tiger and KV-2 for reference can be seen below.


Fig 2: Shots required to kill a full-health, 6-man squad.


Fig 2: Time required to kill a full-health, 6-man squad.

Expectedly, the S2K or T2K values show significant dependence on the squad spacing and range. However, and more important for the actual discussion, you can clearly see that the effect of the AOE changes on the overall anti-inf performance of the Tiger (as well as IS-2 and M26) is rather limited, and the tanks will be nowhere near as "neutered" by these changes as people suggest.
The actual magnitude (see Table 1) is in the range of only 5-10%, and both Pershing and IS-2 seem to be impacted more by the WBP than the Tiger, in particular for more "clumped-up" formations.

	Clumped	Stacked	Wide
Tiger	5.3	6.9	5.2
IS-2	7.0	4.7	4.9
Pershing	11.1	10.4	8.1

Table 1: Percentage of T2K increase for WBP-versions of Tiger, IS-2 and Pershing depending on squad formation

TLDR: Tiger will still be going strong against infantry in the upcoming patch

But, are you using the correct scatter values?

Once again, thx for crunching the numbers. I’m a bit surprised that the S2K a 4 man squad is only 10% faster. From these numbers it seems like the IS-2 pretty good balanced across squad sizes.

One question though: When calculating T2K are you starting the time at 0 shots? If we disregard the time to rotate the turret then engagement basically starts when the first shot is fired. With that in mind, should the T2K not be counted after the first shot instead?

Yeah the small difference between the S2K values seems surprising indeed, given that most people tend to assume a 4 men squad would have to die in 2/3 the time a 6-man squad does. However, you have to keep in mind that this heavily depends on the AOE profile of the weapon and squad formation/spacing.
In the examples above the difference is highest when the average distance between each model decreases between the 4- and 6-men squads, as it is the case for the "stacked" formation (here the difference is up to 25%). Also, the IS-2 has a pretty large AOE radius (5) and minimum AOE damage (28), meaning that even in a fairly spread-out formation each model is likely to receive at least 28 damage even if not directly hit. Hence the low variation between the "wide" 6- and 4-men formations. For a weapon with low scatter and relatively small AOE I would indeed expect the S2K difference in the "wide" formation to be much greater.

You're absolutely correct, the T2K value is calculated after the first shot by taking (number of shots - 1) x (reload + windup/winddown + etc).

But, are you using the correct scatter values?

Well, that's obviously what I've aimed for at least. If you mean whether I'm using the values from the patchnotes or plugged from the editor, then it's the latter. That of course is not to say that the scatter model itself may not be 100% accurate, but it should be very close as far as I can tell.