I only have personal experience. I understand that the plural of anecdote is not data, but the effect of these assertions is so striking that I would bet money on this being true. And sorry for the wall of text. I just don’t know how to make this any shorter without just saying, “trust me.”
Autotargeter: I had heard that an autotargeter can reduce drone agro, but was skeptical. So, I went out into the same combat sig with and without the sig amp on, and the difference in drone agro was like night and day. I went the entire first pocket without any agro, and I believe I only got drone agro in the second because I got too far away from the rats with my orbit.
Damage/Ewar: In incursions, we have what we call agro magnets. They are guys whose combination of damage and Ewar help them to consistently draw agro switches. Thus, we add them to watchlists and/or keep them pre-locked. Hell, we even give them Teamspeak badges for being an agro magnet.
Spider tanking: I’ve heard people complain about drone aggro in content that I had personally experienced no drone agro in -I attributed this discrepancy to the fact that I was spider tanking. Regardless, WTM teaches logi pilots not to use all 4 reps unless in emergencies, because doing so will greatly increase the chances of logi ships drawing agro. In fact, I over rep in some situations in order to intentionally draw agro. For example, I like to draw agro during the 1st wave 3rd room of a TPPH in order to pull the rats towards the DPS anchor spot. I will get at least some of the agro 99% of the time (no exaggeration, it’s almost guaranteed). And can get most to all of the agro about 50% of the time -which seems to depend largely on the presence of badass dps boats in fleet. Long story short, I’d bet money that remote reps have an effect on threat generation.
Anyway, I could go on, but hopefully you get the idea. I don’t have a source, nor have I conducted scientifically rigorous research or anything, but I have had gotten good results by operating under the assumption that these things are true.
Now, for the matter of your test results.
I think the way agro works is that a multitude of factors each contribute to an overall agro score, which determines your place on the agro table. And then the higher you are, the more likely you are to get selected. My belief in this is based on my experience with agro magnets in incursions. Instead of one guy always getting it, the same 2-5 guys will draw agro switches wave after wave, site after site -most of the time, that is. Sometimes, however, “a lesser agro magnet” will draw agro -which I believe is the result of them being high enough on the agro table to have a small chance of getting it.
Moreover, things like range to the rats does appear to have an affect on agro. Thus, it needs to be controlled for. For example, when running EC’s, I would usually warp to an asteroid at 70 in order to land near the spawn point, set up my spider tank, and get to work. In such instances, I hardly ever got drone agro (I’m talking I might lose 1 drone every 80 hours or some ■■■■). However, I would sometimes accidentally warp to the asteroid at zero, and wind up further away from the spawn point. I found that when I did this, my drones would be significantly more likely to draw agro.
Anyway, all of this is to say that all confounding variables need to be controlled for. For example, an autotargeter might dramatically decrease agro for someone who has good DPS and who stays within range of the rats, but might have little effect for someone who is orbiting 50k away with no weapons fit. Is this making sense? Perhaps a hypothetical chart might illustrate what I’m asserting.
Of course, I just pulled these numbers out of my ass, but I suspect that agro works like this. Anyway, in this example, adding an autotargeter to the domi can make a big difference to drone agro, while adding the same exact mod to a VNI orbiting at 50k might make no appreciable difference.
I’ve also heard that threat generation can work differently depending on the A.I. of the rat in question. So your mileage might vary depending on your content. FYI, most of my experience has been with incursions, Invasions, Anoms, Combat Sigs, Escalations, live events, and the abyss, to a lesser extent. IIRC, things like drifters and diamond rats have different A.I.'s, but don’t quote me on that. I also usually brawl in PvE, and have been spider tanking trig ships a lot since they came out.
Alright, that’s enough. If you want to go out at conduct actual research, be my guest. I get noticeable results from my assumptions, however, and will continue to act as if they are true until I see strong evidence suggesting otherwise. I may not have proven this is true, and there may even be other confounding variables that I havent’ considered*, but I’ve like the results that I’ve gotten.
Fock. I know this is already a wall of text, but this is interesting, and may even be relevant. So, I once read an article about an MMO that had finally found and fixed a long standing agro bug. Certain players had complained that they were significantly more likely to draw agro, but the devs initially thought it was in their heads. Turned out it was true. Something about their character id’s, or some ■■■■, was causing them to get placed higher on the agro table than they were supposed to be. Thus, there is the remote possibility that we might straight up be getting different results due to a bug. Anyway, sorry I can’t find the article I read, and don’t even remember the name of the MMO.