EVE General Discussion

Clearly what's he's saying, and he was the lead dev from its inception, is it was designed to be safe. It was not supposed to be easy to kill people in high sec. If you wanted to be hard to do. High sec piracy was not encouraged or supported by CCP.

Crucially he finishes with" That's the WHOLE point of high sec", my emphasis, as in it was designed NOT to be dangerous.

Makes the idiots that constantly spout BS about EVE always being as toxic as it is now and that the system we have now was designed from the start to be this way look like spaz's. As they are.

And he considered it quite safe...safe when you could tank CONCORD. Safe when players (enough of them) could kill CONCORD (at least I have seen claims to that effect). Gate guns would not fire on you unless you engaged in aggression in range of said guns. And over time CONCORD has gotten stronger (they are now invincible, in effect) respond faster, and look at what has to be done to kill people in HS, you have to work in sizable groups or incur fairly substantial ISK losses. And the removal of insurance when killed by CONCORD has made ganking more costly. The changes to watch lists and war dec mechanics have made war decs hilariously easy to avoid.

And these things are basically relative. No section of the game has ever been designed to be totally safe. HS is the safest and it clearly is. The vast majority of players in HS do their stuff and do not get ganked. And if you take some moderate precautions your chances of getting ganked are going to be even less.

What people are complaining about are things that are easily seen. A big fat freighter getting ganked (because the player was imprudent). What we don't see are all the freighters going through Uedama and not being harassed. Everyone is talking only about what is seen and ignoring completely what is unseen. This is a flawed form of thinking. It was pointed out around 150 years ago by Frederic Bastiat. His point was basically the concept of opportunity cost. Everyone here gets that, but then they fail to get a similar concept when discussing ganking. All they do is look at the ganks and then whine because "there is not enough risk for the gankers" ignoring the fact that the ganked made choices (horribly bad ones) that in large part lead to them being ganked.

So, everyone who complains about ganking should log in and get in a ship and go sit in Uedama, say the Sivala gate, and watch how many freighters go through, then go look at zkill and see how many got ganked in the same period of time.

Well without defining what "more likely" means and if it reaches statistical significance levels, it makes it only an inaccurate observation. CCP at ~1:44 says it too, that they only got limited ability to analyze this data

No, not quite correct.

Okay, what his your hypothesis: That ganked newbies quit sooner. [strike]at a higher rate than players not ganked[/strike].

At the very least this tells us that we cannot reject the null hypothesis (that there is no difference between ganked and non-ganked newbies).

Think of it this way, suppose the confidence interval for ganked and non-ganked is (12,38) and (20,50) where the values are days. And that the mean values are 20 and 35. Clearly the means are not statistically significant in terms of their difference. The mean value for ganked players days is 35 which is inside the interval for non-ganked players. In this case, we fail to reject the null hypothesis and thus we do find support for the alternate hypothesis. Even if the confidence interval for ganked players were wider it would not help in that we'd still fail to reject the null. Because the mean value for the non-ganked players would be inside it.

Logically, the results from that presentation should quite clearly weaken your belief/support for the hypothesis that ganking new players leads to them quitting sooner than if they were not ganked. At worst it has no effect, at best a beneficial effect.

Which you failed to do, IMO. Worst case interpretation ganking players in their first 15 days has no effect.

While I actually agree that CCP probably got it right in relation to ganking in the first 15 days and retention rates.

The statement "Also even an inaccurate observation is better than no observation at all and an uninformed opinion based on a gut feeling." is a complete pile of crap. Take the cold fusion experiments of the late 1980s. They received an inaccurate observation, that they had achieved cold fusion. Everyone's gut feeling was that this was wrong.

It turned out the equipment was malfunctioning. So NO an inaccurate observation is by its definition WRONG, while a gut feeling may be right.

Well, I'll go with John Tukey,

Killing people in high sec has never been easier. Highsec is not safe, its extremely dangerous. I've lived in all areas of the game and by far the most dangerous is high sec.

As for freighters if someone wants to kill you you are dead. No amount of tank, inertial stabs, scouting, planning, webbing alt will help you because you cannot warp out when your scrammed or pointed and they don't align fast enough to avoid being locked and scrammed. After that its bump and gank till your dead.

To put it in perspective, the drop in required isk to gank a freighter is enormous, it used to be 30 or more battleships now its 30ish cats, or 7 stealth bombers. Very very cheap.

What we are dealing with are anecdotes. Some here use one set of stories about how dangerous HS is, others here (myself included) have other stories about how safe HS is*. (Not that I'll go to Jita any time soon)

What CCP is dealing with is data. And it's quite clear from the monthly reports, just how much data.

*apart from camps and bubbles, most places are fairly safe if you're careful.