I assume we both agree (from another thread)
that the 2A5 is fine being the same BR as the M1A2 / M1A1 HC / Click bait.
Let’s compare the Leopard 2A4 VS M1 Abrams, and the Leopard 2A5 VS M1A2.
Armour:
I’ll be using 120mm DM23 for the RHA values.
2A4:
The 2A4 has decent hull armour, but nothing too crazy.
The LFP is easily penned (like with most MBTs) and is around 240mm thick. (Covers around 40% of the hull)
The UFP is around 305mm thick, which is okay but nothing crazy. (Covers around 35% of the hull)
The upper UFP, however, can auto-ricochet APFSDS:
(Covers around 24% of the hull)
The turret area is also nothing too impressive, but the left cheek is around 400mm thick, and the right cheek around 435mm thick.
The breech is around 295mm thick.
The turret cheeks can block some weaker rounds, such as 120mm DM23, and 105mm DM33 / M774, but the breech is quite large, and has a ~300mm thick weakspot where the gunner optics is - killing both gunner and commander.
vvv Against M774:
The turret ring is also fairly small. (Covers around 1% of the Hull)
You can hide the hull by going hull down, due to the -9 degrees of gun depression, but rarely the case in city maps.
I’d give the hull a 1.5/5, and the turret a 3.5/5, giving the entire armour profile being 5/10.
Abrams:
The LFP of the hull is actually not too shabby for its BR, which is around 390mm thick. This can sometimes get 105mm DM33 / 120mm DM23 to non-pen, but is easily penned by other rounds, such as 3BM42 and DM63. (Covers around 45% of the Hull)
The UFP of the hull is also not too bad, as it covers most of the hull and is an auto-ricochet:
However, a problem with the UFP is that it can ricochet the round into the turret ring / breech / crew compartment for the upper 90% of it, albeit the damage varies from not doing anything, to injuring some crew and turret modules, to killing them out right:
(Covers around 55% of the Hull)
The turret area is not too impressive either.
The turret cheeks are a bit weaker, around 395mm on the left, and 440mm on the right.
The breech is around 385mm thick.
vvv Against 120mm DM23:

This can stop weaker rounds, like 105mm DM33 and sorta stop 120mm DM23 (at least with the right cheek), and it doesn’t have the gunner optics issue found on the 2A4… nor does not have as big of a breech as the 2A4. However, it does have small wedges on each side of the cheeks that are worth around 245mm of RHA.
The Turret Ring is a noticeable weak spot, and is worth around 62mm of RHA.
This allows autocannons to kill the Abrams frontally, as well as the any MBT with APFSDS - irrespective of BR.
You can hide this turret ring weak spot to some degree while hull down in hilly maps with the 10 degrees of gun depression (less so than the 2A4 though), but rarely the case in city maps.
I’d give the hull a 3.5/5, but the turret a 3.0/5, giving it a total of 6.5/10.
However, if we include the turret ring issue, it’s more like a 4.5/10.
Survivability:
2A4:
One-Shot area:

Since it has 3 crew members on the left-hand side, most corner approaches from that side is quite dangerous, as they can severly cripple you or easily 1-shot you.
The other side however, is much less dangerous, especially if you do not take 2nd stowage ammunition (Hull Ammo). This means that you can side scrape somewhat better in the right side than the left side.
The ammo is kept safe in the turret bustle.
The 2A4 doesn’t have nearly as big of a turret ring, and so it doesn’t really have a 1-shot area there.
Do keep note that at close range (let’s say <20m), the UFP’s angle is lower, so the easier it is for a round to not ricochet.
This means that it would look more like this at close range:
Abrams:
One-Shot area:
Instead of the driver being left of the tank, it’s now centre.
This makes the hull shot smaller, but still possible.
The turret ring is much larger, and so it’s more devastating to the crew.
The modules you hit with that shot can also cripple its ability to flee, shoot back, or even do either.
This is not an issue when getting shot at the 1-shot areas in the 2A4 - other than maybe the engine and turret ring.
At far ranges, and for flicking shots, it’s easier to shoot centre mass of vehicles than not, which I assume you agree with. So I believe trying to go for one-shotting the Abrams is easier than trying to go for one-shotting the Leopard.
For example, if both vehicles were an array of pixels 7x wide, and 7x tall, we may get this:

I can’t speak for you, but I believe it’s easier to hit the one on the right than the one on the left, if you had only a couple seconds to do so.
If you do miss the Abrams’ 1-shot area, you’d either hit the UFP (nothing), or the breech.
If you do miss the Leopard 2A4’s 1-shot area, you’d either hit the UFP (usually nothing too), or the LFP (which you will get the driver, engine and maybe the gunner).
So I’d give the 2A4 a Survivability rating of 7/10, and I’d give the M1 Abrams a Survivability rating of 4.5/10.
Maneuverability:
2A4:

Abrams:

I know that HP/TON doesn’t say everything, but they have near the same acceleration.
Although like Necron said, the Abrams is slightly better in neutral steering too.
The Abrams has a 6% higher top speed, and a 23% higher reverse speed.
Make that what you will, but I would say both are quite good, and not too dissimilar.
So I’d give the 2A4 a Maneuverability rating of 8.5/10, and I’d give the M1 Abrams a Maneuverability rating of 9/10.
Firepower:
2A4:


Abrams:


So the Abrams’ has ~40mm less flat pen (not really worth comparing), and around 32mm - 22mm less angle pen.
That isn’t that much, but penetration affects whether or not a round goes through, how much spalling the round makes after it does penetrate, how much the spall can ‘penetrate’, and how much damage the spall fragments do.

(This source is around 6 years old: https://www.reddit.com/r/Warthunder/comments/9m338k/ingame_apfsds_postpen_performance/,
but I assume they didn’t change the code since then, and I couldn’t find any other sources), although this video: https://www.youtube.com/watch?v=JdzZIDtLp2A,
demonstrates that mass doesn’t seem to really matter - although some APFSDS (that aren’t classified as long) may take it into account).
So from what I understand, there is a min residual pen spalling, which is <20mm.
Then there’s a spectrum from 20mm to 150mm of residual pen for spalling.
Then it’s max spalling past 150mm of residual pen.
As for the video, (which is more recent), the min spalling residual pen is actually <50mm.
Then there’s a spectrum from 50mm to 400mm of residual pen for spalling.
Then it’s max spalling past 400mm of residual pen.
So unless you are shooting at a flat plate worth 10mm, you are not getting max spalling (possible in-game) with the 2A4, and you are never getting max spalling (possible in-game) with the M1’s APFSDS round.
As for the 32mm - 22mm more angle pen with 120mm DM23, let’s say you’re going for a 1-shot against a Christain II:
410-331 = 79mm of residual pen, so you will get 79mm residual pen of spalling, penetration, and damage.
Now let’s say you’re doing the same 1-shot with the Abrams against the Christain II:
372-322 = 45mm of residual pen, so you will get min residual pen of spalling, penetration and damage.
For example:
The 2A4’s round (from testing) can more reliably 1-shot than the Abrams’, which is to be expected I guess.
How often does it really matter?
I couldn’t tell you for sure, but the slight increase in lethality and slight better penetration of 120 DM23 over M744 is nice - I guess.
You can find other random wacky ways in which the difference can matter, such as when shooting at angled sides - especially against Russian side ERA.
The reload on the 2A4 is 6s, but the reload on the M1 Abrams is 5s.
So the 2A4 is around 20% slower (in terms of reload) than the Abrams.
When they were both 10.3, them being able to face the abundant squishy tanks at 9.3 made it so that reload rate was by far more worth it than the extra pen and damage.
However, I’d say that the better firerate still helps a lot, but the Abrams’ firepower not nearly as out-right better than the 2A4’s firepower anymore, as it’s less useful in uptiers than more penetration, but still has uses in terms of killing multiple targets at the same time and 2-shotting opponents quicker than the 2A4 (although the 2A4 may not even need to shoot that second shot).
Let’s look at the PT-16, though.
It has a 4s reload, so the 2A4 is around 50% slower than it, and the M1 Abrams is around 25% slower than it.
This is more apparent with the Abrams VS the 2A5/2A6/2A7 VS Type 10.
They both have the same turret rotation speed, but the 2A4 has a 67% better gun elevation speed, although 24 degrees is more than enough in most cases anyway, so I won’t take that into account.
Nevertheless,
I’d give the firepower rating for the 2A4 a 7/10. (What used to be a 7.5/10)
And I’d give the firepower rating for the M1 Abrams a 7/10. (What used to be a 7.5/10)
2A5 VS M1A2:
Armour:
2A5:
Now with that out of the way,
the improvement in armour with the 2A5 is just with the turret.
For the BR, the hull is even worse (especially because of the BR and the fact that nothing changed to it).
So I’d give it a 1/5.
The turret, however, had improved drastically.
The turret cheeks are impenetrable, and the wedges makes the breech smaller.
However, the lower breech itself is only around 105mm thick, instead of the 2A4’s 385mm.
So I’d give it a 4.5/5.
Overall, it’s around a 5.5/10 - but could be 9/10 if hull down.
M1A2:
The improvement in armour with the M1A2 is only the turret.
For the BR, the hull is even worse (especially because of the BR and the fact that nothing changed to it).
So I’d give it a 2.5/5.
The turret, however, had improved by quite a bit.
Most things cannot pen your cheeks frontally - not even the left cheek with DM53:
The breech is also the same as the M1 Abrams, unlike the 2A5.
However, if you try to wiggle around your turret to avoid getting shot in the breech, even the slighest of angling can make it easily penned by long-barrel DM53 - let alone short-barrel DM53:
So I’d give the turret around a 4/5.
Overall, it would be a 6.5/10, but it could be a 8/10 hull-down.
It still has that same turret ring issue, though, so I’d put its overall score down to 5/10.
Survivability:
The survivability of both are effectively the same as the previous ones, although the 2A5 has spall liners all around its turret (other than the rear), which drags the 2A5’s survivability up.
It also may be worth mentioning that the 2A6 has a longer barrel than the 2A5, so it’s more easy to take it out. The 2A5’s and Abrams’ barrel length are similar from what I can tell.
You cannot take out the gunner and commander from the left cheeks of either vehicles with all rounds now, too. So both their survivability gets dragged up.
But since there are now much higher penetrating rounds than before, there will also be more spall fragments, more spall damage, and penetration of those spall fragments, which drags both of their survivability back down.
So it would be an 8/10 for the 2A5, and a 5/10 for the M1A2.
Maneuverability:
2A5:
M1A2:

The 2A5 accelerates slightly better than the M1A2 (around 3% better), although the M1A2 may still neutral steer a bit better.
They both have the same top speed, but the M1A2 still is 23% faster in reverse.
Although, they both still go fairly fast in reverse, so make that what you will.
I’d give the 2A5 a mobility rating of 7.5/10, and the M1A2 a mobility rating of 7.5/10 too.
Firepower:
Now here’s the interesting part:
2A5:


M1A2:

They have similar penning rounds, but the 2A5 reloads every 6s, whereas the M1A2 reloads every 5s.
This means that the M1A2 has a much better firepower than the 2A5.
They both have the same turret rotation speed, but the 2A5 has a 67% better gun elevation speed, although 24 degrees is more than enough in most cases anyway, so I won’t take that into account.
I’d give the 2A5’s firepower a 7.5/10, whereas I’d give the M1A2’s firepower a 9/10.
=================================================================================
So overall, I’d give the 2A5 a rating somewhere around 71% , and the M1A2’s rating somewhere around a 66% or a 68%(with gimicks included - ignoring stuff like optics, ESS, smoke grenade amount / pop count, thermals, inability to look behind due to NATO hump / near the rear due to new fix for Leopard 2s…etc, which may boost more points towards the Abrams (maybe by +1 in total)).
Of course, the 2A5 is usually better at hull down and long range due to the better turret armour and lack of turret ring weakspot, whereas the M1A2 is usually better in the flanks due to the better reload, same round, and slightly better hull armour - despite similar mobility.
As for the 2A4’s rating, it would be somewhere around 69%, and the M1 Abrams at around 65% or 66% (+1 for the gimicks that it also has over the 2A4).
Same story as with the 2A5 vs M1A2, but the fact that it has similar firepower as the 2A4 makes it so that it feels more of a light tank with above average firepower than an armoured light tank with great firepower (like with the M1A2).
I hope that kinda makes sense.