When a platform makes an SAR scan it generates a very narrow EM beam for ~2 seconds. What that means is the only thing that will detect the active SAR is the unit being painted, and they only have a couple of seconds to do it. Thus the (lets say) SAM battery may detect a fleeting EM source, which it can not detect or track on radar, and in a matter of seconds to minuets a weapon will impact. Assuming the SAM battery has very capable ESM. Thus in real terms the F-22A betrays nothing by using SAR, except giving the crew the option to abandon their posts and run.
But the missile fired at the SAM unit, itself can be engaged and shot down, and a modern IADS will have no problems intercepting it. If it also happened to have caught the short EM burst, it can then put 2+2 together, and at least have more or less confirmed detection of inbound hostiles.
The other thing is that in a properly done IADS, the only permanently hot radars will be general airspace control radars, and only when a threat is detected will the SAMs actually go hot and start searching for targets. Even the Georgian SAMs (SA-11) during the recent war (which is hardly an exemplary ADS employment) operated in this manner. An S-300P battery can go from march (cold) to hot in 5 minutes. This includes time to set up the unit. It would be reasonable to assume that if it's already deployed, and just needs to flip the switch, it could go from hot to cold in notably less time. The Buk M1 proved it had the capability also. I think it's safe to assume that many of the SAMs currently being actively proliferated around the world are capable of this type of employment, which immediately changes the game for the penetrating platform. It doesn't know where the GBAD is until the GBAD is ready to open up on it, and thus can't judge threat levels.
It seems to me that most of the scenarios being discussed here are against an inferior opponent. I understand the contextual relevance, but it's hardly worthwhile to analyze a situation where the opponent is inherently inferior in capability. We all know the USA can bomb third world countries into sawdust. If we are trying to truly see where the limits of the F-35 vs the F-22 are in penetration, then we have to push them both to the limit, to the point where their differences in signature management, RCS, etc. become relevant. Against a Syria-level IADS either one will do just as well, fly in, drop the payload, and fly out without anyone noticing.
EO DAS is actually a development of the F-22A's MAWS, it just provides the ability to project IIR imagery to the pilot and target A2A missiles. Therefore in this scenario, dealing with GBAD threats, EO DAS will only provide a slight improvement over the MAWS. IF the GBAD system itself is close enough for the EO DAS to see they your in trouble, and if there is a missile launch the only difference is the pilot will be able to see an IIR image of it (the MAWS will still provide a threat bearing). The HMD is planed to be integrated on the F-22A in time. The only real difference in 2020 SA will be the EOTS, which will not provide any real additional benefits over SAR when targeting J series weapons.
Except that an IADS does not consist solely of GBAD. It also includes airborne components. So the EO-DAS still has significance in that regard. The true objection here should be that once you're down to fighting the enemy, instead of a stealthy penetration, you've already defeated the whole point of a VLO platform.
In a purely legacy network, I would say you are right... but all that changes with our stealth assets.
For example, if a GBAD has a low-level detection range of 50 miles for legacy aircraft, then it's stealth detection range is about 5-10 miles. If these assets are placed 50 miles apart in order to have 100% overlap in a legacy environment, then there is an approximate 30mile corridor between them for stealth assets to exploit. Even if the missiles get launched, they will have a very difficult, if not impossible, time locking onto the stealth asset due to it's reduced RCS.
The SAMs will be cold, and their locations not known if the IADS is operating on wartime footing. Being mobile they can change locations every 3-6 hours if need be to avoid sattelites spotting them. The corridors may exist, but you don't know where they are. You also have air-space control radars, and AEW&C which may not be able to provide targetting data on a VLO target, but is likely to provide detection info at least, which would then enable all the SAMs around the area where hostiles were detected to go hot in unexpected places. They could go hot for no more then a minute, to make sure they don't have targetting data (or take a shot off if they do) and then go cold again and relocate immediately. This would also enable AD fighters to vector in at least an approximate intercept, if the VLO targets are detected. If the AD fighters are datalinked to the SAMs, then we have a situation where the pentrating platforms are in for a very messy penetration.
Just look at GW1 and the success of the F-117. Even the F-35 has a much better RCS than the F-117.
The Iraqi IADS was not operated competently. Nor was it's airborne component worth the cost of gas in their fuel tanks.
Again bombing third world countries into sawdust, with complete capability overmatch, is not true test of capability.
