Gevlon's changing post...

Well looks like even he knew he was wrong, but was just too proud to admit it. Today I wake up to another new set of comments on yesterday's post (thank you) and then went to read my morning Eve info dump and noticed the quick mention of Gevlon changing his post from yesterday. Well I just had to go take a look, I mean after all, I'm sure it was just minor changes....

So, his changes.... let's take a look.... Hmmmm, a mostly new paragraph close to the top with actual numbers replacing his fo-statistical analysis the had at the bottom of yesterday's post. Honestly I am surprised he didn't break them further down into percentages.... what else?

He removed a few graphs that didn't really show anything. And added a few more lines to his analysis on the graph I pointed out yesterday... but just to make it clear that the reason WH players have a higher average loss isn't because they have more expensive ships. OH NO! It's because they get caught in their bling PvE ships and get ganked more often! Of course, how could I have been so dumb?!

I mean other than that whole "High isk losses are always PvE" losses mechanic that he has thrown in there, that isn't, hmmmm, whats the word.... true.

So let's quickly delve into why WH losses are on average more expensive. Where I fly, I tend to do a lot of small gang warfare, but not really on the same "level" as WH small gang warfare, where you need to make every module, every single percent count because WH's limit the mass you can bring through. In WH space it actually does make sense to pimp as much as possible.

This turns a usually 2 bil isk Bhaalgorn into easily a 7 bil isk Bhaalgorn because those extra few percent more cap drain or percent more resists can, and do, more often turn the tide of a small gang battle. Gevlon likely doesn't understand this because he's never been in a small fight where those kind of percents matter. His eyes are so focused on Nullsec fights that he may not yet comprehend that there is a whole other level of fighting going on out there that he is not aware of.

So then spread that across the most used hulls in WH space, Tech 3 strategic cruisers, whose hulls and subsystems alone usually cost 500m isk, unfit, then you quickly get above the 1 bil isk mark on these ships with all that extra bling packed int here, be that bling for PvP or PvE for that matter. In short WH ships are more expensive because by FUNCTION they need to be more capable. Hopefully that's not hard logic to follow.

The other funny assumption that is still left in Gev's article....? That all of highsec is more dangerous than WH space, you know, because living in backward ass Amarr highsec, with 2 other people in local, at MAX is scarier than say, any WH system. When a "level of dangerousness" scale implies that the above is true and then is claimed to be fact. Well let's just say that doesn't hold up too well..

Also, lest I not forget, here were some great points made in my comments yesterday -

Azual ( of the blog The Altruist )  - "Jita sees about 6x the number of kills per 24 hours as Rancer (a notoriously permacamped lowsec pipe system), but you're far far more likely to get killed in Rancer than in Jita. It would make much more sense to analyse kills per resident or kills per jump, at which point I expect WH space would rank much more highly."

Anonymous - "I would like to see people like Gilgalaad or some of the folks in the Eve Uni trade chat to publish their knowledge about the market in a blog.

The knowledge itself is largely available, those guys give it for free, however a blog is a much more accessible way to learn about it."

I too would like to see more blogs about trading since I find that stuff pretty good reading. I did some station trading, what seems like ages ago now, but mostly meta 4 module trading, and made a huge profit while never leaving Jita 4-4. Lazy in my old age I suppose.

There were tons of other good comments, but mostly about how shitty Gevlon is as a source of information for things other than Market trading, and wishes that more people would take up that mantle. Even one great comment from the Gevvie himself. But I'll let you go read that off the article if you haven't seen it already!

A response to Gevlon's poor analysis.

@Gevlon post here. I know I shouldn't even respond as he has always shown a none too bright attitude towards how eve works and I could find fault in near every post he makes, but this one just stands out....

Here's the real baseline of the issue. Gevlon spent so much time trying to come up with data to support his position that he failed to actually interpret the data he was given. He disregarded or failed to notice data that did not support his notions, and then formed fallacious assumptions around those numbers.

To recap, he says...

"WH people are both extremely rich and that WH space is extremely dangerous. I used regional statistics from killboards to prove that WH ship losses aren't that expensive. They claimed that because they go to PvP in null (because they are so awesome that no one attacks them in their WHs). So I went to look for more data."

I ask you, to look at the "Average loss value MISK" graph again. The statement he is trying to disprove is, in short, "WH pilots, on average, pvp in more expensive ships and more dangerous area's."

First - Do WH pilots fly in "more expensive ships?" Yes, using his own graph, in every loss category, both in few losses and lots of losses, "WH people" aka Dark Blue, lead the curve. That alone, ALONE, is enough proof that indeed, WH people do fly and lose more expensive ships, on average than any other "Area," as defined by Gevlon.

Second part ("Dangerous area") - this is where it gets tricky, how do you define danger? Isn't this just a judgement call? WH people will say that WH space is pretty dangerous because of the lack of Local, or any wide ranging, local based intel networks found in low/high/null that can warn of approaching hostile fleets/gankers (by way of setting standings, or seeing criminal flags, long before they are a threat.

Now for WH space there is another wrinkle, it's the only true part of the game where fighting the PvE content is the only means for existence. Now what do I mean by that, I mean that in order for it to make sense to live in WH space, you must either be actively doing PvE or at least devote some of your time to PvE to make a living out in WH space using traditional methods.

Lowsec, Nullsec and Highsec are not limited in this manner. Lowsec does have FW plexes, or other PvE items, similar to those available, but much less profitable than, the plexes out in Nullsec, but you can still use PI and stations to evade, change or modify the combat. Nullsec has moon mining, reactions, drug, cap and supercap production just to name a few things that do not require active in-space participation. Highsec I shouldn't even have to cover because station trading alone is so profitable.

Meanwhile WH space stands alone with the massive majority of it's ISK making ability coming from sites that need to be run and thus can be scanned down and pounced. This is such an important distinction because it goes to lending towards a much more risk based, high danger atmosphere to gain the benifits of living in WH space than anywhere else in the game.

Then, Gevlon tries to pass off the majority of expensive losses by semi-to low activity pilots as "PvE" losses, or ganks. Now while I don't at all agree with this idea, except in the case of highsec, let's look at things through "his" lens. Expensive "PvE" losses actually help prove that WH space is more dangerous than the average other area in the game.

Why is that? Well simple it's as simple as my above point about use of the content and also risk-aversion. Rarely will a Eve player just jump in their most expensive ships, and fly around in high-risk situations. With the possible exceptions of the super-rich. But really we are trying to focus in on the "average player" of each demographic.

I will say that the average Nullsec resident does not fly their pimped out ratting/botting ships unless they have the means to ensure it's safety. In WH space there is no real way be "safe" as one is in nullsec. Anyone with a probe or an anom scanner can jump in on their ships. Yet WH players continue to field and lose expensive ships, even though it is more dangerous.

If that is not your definition of 'danger,' Then maybe you lay more weight to the number analysis of Gevlon, however those numbers are flawed and easily refuted when looked at by a few simple factors.

For instance, Gevlon does not factor in the all important data about the 'opportunity of loss.' In WH space with it's MUCH lower population density, proved using both CCP's stats and the Gevlon stats, reposting a graph from Gevlon,

Clearly, the number of people living in WH is the smallest of all the "area's" people choose to call home, by both the CCP and Gevlon killboard based metrics. Gevlon attempts to provide some kind of "ratio" but this ratio value is completely useless. Like comparing apples and oranges.

Gevlon takes his ratio by using his 9.2% of the polled PvP losses are "WH" based pilots, and then he compares that to the supposed total population of WH space (5.5%). It's because it's a bad ratio... Look...

The problem with the ratio is, Gevlon is taking a full sum value (a percent value that adds up to 100%) and comparing it to ANOTHER DIFFERENT full sum value and coming up with a ratio. CCP based their population values not on PvP activity but where players spent the majority of their time. Gevlon is just looking at PvP values, comparing the two values does not reveal the % of pvpers or a ratio of PvPers to non-pvpers. The ratio is the difference between his PvP fed values and the values that CCP provided. Not his assumed idea that it shows the ratio of active pvpers in a group of players. That's value is just total statistical nonsense.

Back to my point... the importance of the 'opportunity of loss.' If you only have a relatively small number of inhabitants, that makes both fighting and losing ships more rare. Now THIS is where Gevlon goes COMPLETELY off the rails. He breaks down WH space into C1-C6, but then leaves Null, High and lowsec combined. Either he did this because the data when combined, ie C1+C2+C3+C4+C5+C6 didn't support his theory, or because he was just lazy. Now because I don't have his exact figures all I can do it make a guess.

If you add up the values for the different classes of WH and provide a combined WH stacked graph, it would show the true value of all WH systems, and that value would be much higher than either Null OR highsec, second only to Lowsec... But like I said that is just a guess.

Let's not fail to point ou here that by his own graph Nullsec has a lower 'value' than Highsec. Sure it's damning when a C6 doesn't live up to highsec but then Gevlon glosses over the fact that Nullsec also doesn't have as high a value. Also that's a simple explanation  Highsec has more people than either Nullsec or WH space, in fact more than both combined. This gives a much higher sample size than either AND much more opportunity for loss.

One last thing, Gevlon bases the majority of his assertions on the idea that in short "a player with a high isk loss but low losses per year is essentially a PvE player." This completely disregards the "casual" eve player like myself and many others. I'd be lucky to get 45 FIGHTS in a year I'd say, let alone 45 losses.

Using myself as an example... While not a very active player, I am certainly NOT a PvE player. One of my losses easily pushes me over the 50m mark in a year. Now I am not exactly a huge loss player, IE when I die in stuff it's rarely all that expensive, but still I lost a carrier this year, that was a 2 bil isk loss or so, Plus other assorted losses, and now according to Gevlon, I am either an "outlier" despite the fact that I know of many other players just like me, or a victim of being ganked doing PvE. Seems a rather stark contrast.

To remove the true PvE gank losses, Gevlon should have called high isk losses with high damage NPC values, indicating plexing or belt ratting, as PvE losses. And then taken steps to exclude those results from his findings, I'm sure the appropriate SQL command could easily do this... This would quickly provide a more accurate and stronger overall graph. I'd imagine with those results stripped out than Gev's assumptions would not be so easily passed off as fact.

So, in short, the statistical analysis offered by Gevlon is flawed, incomplete and uses flawed logic in an attempt to "proof" his conclusions.