Come with me now, on a journey through time and stats!

 

 

I like completing sets, I must confess. I’m the sort of person than genuinely enjoys seeing ship masteries count up on the ship tree. I dream of one day having a TQ(Tranquility) ship tree that looks like my TD(Thunderdome) one. As such I have been made immensely happy by the Solyaris Chtonium team on Sunday Night, #095, then mad, then happy, then a little mad again, but that’s a different issue.

What they did that made me so happy was bring the little green sliver of this doughnut, T1 Frigate Logi, specifically in the form of the Navitas. Surprising given its T2 sister, the Thalia, has been so utterly shunned for the Deacon that’s seen 17 pairs picked. What then played so much with my emotions was the rest of the teams fitting. With the T1 frigates putting out only half as much rep power as a pair that a T2 Logistics Cruiser can solo before we get into the reps lost in falloff or the fragility of the ships, they cannot be leant on like even their T2 counterparts can.

 

 

With this considered the SLYCE team set up their Battleship heavy team and in a beautifully pragmatic set of choices went on to hull tank their pirate battleships including their flagship. This ehp buffer proved key to their victory over Templis CALSF, with their flagship clinging on to its low hull to take the win on points.

Worth noting on these graphs I have always assumed a full high rack of T2 reps of the appropriate size. The Oneiros and Exequror are chosen here as a baseline as trying to get into the rep amounts of the logistics with 5 or 6 high slots as how much rep can be achieved, or sustained by a reasonable fit becomes a complex and subjective topic. The choice of T2 reps can even be argued as fits will often make compromises in the reps to allow more tank or mobility, however, this gives both a ceiling and a good idea of repping power.

Brief errata from last week's blog: I must admit my fallibility, the Rabisu has the same rep ranges the other T2 Armor Logi as its bonuses multiply rather than stack. The Etana is the only T2 Logi to have have a non-standard range at All Skills V, however as this is a 2.5% range increase it is not shown.

The divide between the longer optimal and harsher falloff range of Shield and Armor is of course obvious. We also see minimal differences between T1 and T2 Cruiser options but Frigate options have a drastically reduced effective range. All worth considering when watching the AT; especially as we see so many frigates and so much play with MJDs testing the very limits of rep range.

 

Now to find the antithesis of logistics: explosions! We’ve had a fair number of them, with an average of 5.55B ISK and 11.85 ships destroyed in every match. Putting the average price of an AT team at 4.7B. Worth remembering that all these ships are real Tranquility ships that are burning for your pleasure. In fact nearly half a million baseliner New Eden Citizens have been killed in the course of the AT so far.

Worth noting that the Battlecruiser Meta we’ve seen so much of this year has done wonders for bringing the cost of the AT down, at least in terms of baseliner lives. A BC cored team, especially if it omits a Battleship, as we have often seen them do, has as little as 15% of the crew requirement that a Battleship core team from previous years would have.

We Demigods of New Eden have little time for the baseliners however, we can throw them a pittance of ISK for their lost loved ones and pay no mind to it. What we do have time for is the blinged out wrecks of flagships. The second weekend found 5 more flagships burn up in glorious conflict and set the benchmark much, much higher for expensive losses with 3 of the flagships lost becoming our new top 3. In fact weekend 2 accounted for 5 of 12 flagships lost, but 66% of ISK lost in flagships killed. 52% in fact is in just the most expensive two.

A huge note to be put alongside this is that these killmail values are estimates at best. Due to the limited supply and uniquely intense demand for the officer modules that these ships can fit as well as the more private market through which they are typically sold, the listed prices for them and the prices paid can be rather disparate. Estimates for the actual costs of these ships can run to 2 or 3 times the listed value.

We have destroyed a mere 12 of the 44 flagships that were declared, by comparison, 17 Flagships were declared and never fielded before their teams left ATXV. Of the 20 remaining flagships, across 24 teams, only 8 have actually been fielded. Only 1 of those fielded has been in the winner's bracket, where all teams still hold their flagships and only VYDRA RELOLDED have fielded theirs. The other odd-one-out of the winner's bracket flagsheeps is the A Band Apart. Armageddon, sitting in a field of fluffy Bhaaaaalgorns. Okay, that might be more stuck in my head than it should be.

 

The rules this year have been praised for producing a diverse meta, and the measure that I have always preferred for measuring this is ship pool use. With nearly 49% of the overall pools used ATXV has surprised me and we’ve still got up to 43 matches yet to be played. We are excluding Uniques here as they distort the numbers and many are incredibly unlikely to be fielded in any AT. I would still say some will never be seen, such as the Stratios Emergency Responder, but after ATXIV’s first match from Pandemic Legion, fielding the Raven State Issue, little can be absolutely ruled out. The other major issue is that Uniques are exactly that. They are species that can and will become extinct through destruction of the ships on TQ, within the AT or, harder to track, because the owners leave the game. Of course if their owner has left the game they exist only in a state of lockdown and aren’t lost but some of these lockdowns are more than likely to outlast Tranquility itself.

A quick note before I go on, on the Gnosis, it is technically a Unique ship in that there is a finite supply of them. However, this supply is so large and has already been refreshed by CCP with additional large giveaways of the hull. Combined with the fact it’s often overshadowed on TQ, (Alpha uses to one side) and so isn’t so widely used as to be risking extinction soon. All of these points can also be applied to the Sunesis, the Gnosis’ Destroyer brethren. To a lesser degree these points apply to the pirate faction corvettes, however, at 2 points, they would only see the light of day in a style based move and never for strategic reasons.

So, with these put aside as they distort the pool we instead look at the ships you can actually go out and buy. We have seen nearly 49% of the pool used, this comes down to two major factors. Feeders and the PLEX auction changes are one contributor, they have seen us enjoy a new crop of teams showing interest and getting into ATXV itself. The extent of this can be seen in the fact 31 of our 64 teams that started ATXV did not play last year. Bringing in so many teams many of which are fresh to the AT and so bring in a unique and more experimental form of theorycrafting. This can be seen in some of the stranger comps or unique tech choices we are seeing this year.

An aside here; of those 31 new teams, 8 remain, with 3 still undefeated in the winner's bracket. With only 24 teams remaining and tournament pedigree often being pointed to as a strong indicator of strength. An aside to this aside, however, is that a lack of ATXIV seeding does not mean a lack of tournament pedigree as the HYDRA RELOADED team are one of the most storied AT teams to be found.

Returning to the pool use however the sheer number of ships used is in the second part a product of how many valid strategies there are going into ATXV. We’ve seen all forms of Logi and lack of used, we’ve seen teams based on Command Ships, Navy Battlecruisers, Battleships or even large numbers of Cruisers. Weapons from RLMLs(Rapid Light Missile Launchers) to Medium Turrets to Drone Swarms and all of them picking up wins.

Over 100 ships are inevitably omitted from this graph for conciseness as they haven’t yet been picked. For those curious the total ship pool for the AT stands at 255 this year, 113 picked. Sadly for my need to complete sets, we are never going to see all 255 ships, the more achievable goal of all 38 ship classes is also likely beyond reach, even though 34 are already ticked off. The issue is some ships and even classes are simply unpickable, um, 2 point pirate corvettes... or only have certain rare options that mean the entire class is unpickable on points. *cough* Covert Ops *cough*

The way we have so many meta styles in a competing space, built on different logistics and support techs, leads to a huge number of different ships being used. Despite this, the top 15 most picked ships account for a little under 50% of all the picks this year. This comes back to the popularity of certain support ships even within a diverse meta. The Vexor, for example, is the most picked ship, it accounts for 5.9% of all ship picks, appears in 34% of all comps and 55% of all matches. This makes it the only ship you are more likely to see than not in any given ATXV match.

The Vexor hull itself has proved to be insanely popular, with it's 3, well 4 if you’re being picky, it makes no difference, variants accounting for 14.6% of all picks. They fall at 1, 3, 6 and joint 115th respectively on the list of absolute picks.

The Guardian-Vexor here, of course, lacks any real merit for the price tag and thought it’s one of the truly unique, unique ships, with its game breaking ability to control 10 drones, its limit on bandwidth, however, leaves it doing less damage than the VNI with no drone tracking bonuses for the same points. Not to mention the less favourable slots, reduced tank and truly painful price tag.

Overall the Gallente prove to have one of the shallower used ship pools of the Empire factions, not quite as bad as their Minmatar allies but surprising given their overall share of picks. These have been focused however on a tight core of drone ships, with one of the year’s break out compositions; the Drone based All-In being composed often almost entirely of Gallente ships and even the more traditional compositions with logistics taking a similar route.

No team however has actually fielded an all Gallente composition, though 8 have been fielded that involved no other Empire factions. All matches so far have involved at least one Empire faction ship. The other factions have seen less preference, with no teams picking only Minmatar, 1 picking an Amarr Empire only team and 4 Caldari Empire only teams occurring. Impressively 14 teams have been seen taking a Rainbow of all Empire factions, with 9 of them also mixing in at least 1 Pirate faction ship.. Pirate factions fair less well with 59 of the 192 fielded teams utterly foregoing them.

In a similar vein, only 6 teams fielded so far entirely omit Tech 2 ships, the lure of Logistics, 90% of which has been T2 so far has proved too great. By comparison, 9 team compositions have foregone Faction ships, both Navy and Pirate entirely. Impressively 19 compositions have been built without Tech 1 ships, no doubt the availability of the Daredevil at 2 points has helped in this. At the other end of the spectrum Tech 3 ships saw space in only 16 compositions, with 14 of those being comps with T3Ds, (Tech 3 or Tactical Destroyers).

So, sadly my dream of an all Gallente team is not to be. Excuse me while I go try and convince my captain to run this:

 

Eos
Eos
Oneiros
Ishtar
Vexor Navy Issue
Vexor Navy Issue
Keres
Keres
Vexor
Vexor

•  

•  

•  

I’ve been told I’ve more chance of a single Empire faction team with this:

 

Scorpion Navy Issue

Scorpion Navy Issue

Basilisk

Ferox

Ferox

Stork

Hawk

Manticore

Manticore

Raptor

 

But alas; teams do not make choices based on my desire for fun stats. Another example of this has been that, despite my requests, the absolute pick rate of the Navy Battlecruisers have actually gone down. Now 11.6% after being 13.2% of absolute picks at the end of Weekend 1. My wish had been that it increased past the 13.7% which was the combined Battleship and Destroyer pick rate after the first weekend. By now that’s grown to 14.0% as well, my dream is far from gone, but we are heading in the wrong direction for now.

The sheer popularity of the Navy BCs is beyond a doubt, however, more popular than Destroyers or Battleships as an entire ship class. With BC Cores being used in 28% of all compositions they are seen in huge numbers in these as well as being picked into All-In compositions, hull tanked BNIs are more than just a solo staple. The only other ship class to claim to have been picked more than all Battleships combined is the humble T1 Cruiser.

Tech 1 Cruisers are seeing increased use this year with a points reduction. This combined with an overall points squeeze has made them, or mainly the Vexor, a go-to heavy tackle choice.

With a 65% share of its class and a different role to its closest competitor the Vexor will likely remain a dominant pick through the rest of ATXV; already it holds a place as the most picked ship of ATXV. The Caracal that is so close behind is a pick often used as a cheap source of another RLML platform, lacking sturdiness or particular speed and tackle capability it still warrants 4 points with the damage it brings. The Moa merits its share by providing the tanky shield tackle option that the Caracal so distinctly isn’t.

Returning to the Vexor however you find a ship that has a lot of merits. Particularly it is a drone bonused ship, with good bandwidth, 75mbit/s, very reasonable for 4 points. Comparatively the cheapest 125mbit/s is the VNI (Vexor Navy Issue) at 8 points. The ability to forego all highslot weaponry and still bring reasonable damage frees up the fitting room of the VNI. This allows a solid armour tank, though the slot layout is flexible and we have seen them picked into shield comps in 9% of their matches. The large number of mids more typically lets them bring utility such as ReSeBos(Remote Sensor Boosters) or EWAR. Finally, the released highs allow options on smart bombs and capacitor warfare. In all this makes the Vexor a very flexible ship around its core of projected supporting damage that can provide a given team comp additional potency or flexibility on the fly.

The prevalence of Tech 1 Cruisers has lead to an increasing trend of using purely cruiser based support wings. Through present in only about 5% of all compositions, much more will effectively function with just cruisers playing tackle and mobile support. Often these teams will drop a Command Destroyer or two into the support wing to bring a team’s links and though they can offer a comparable hard tackle role they will leave the majority of the duties to a Vexor or Vexor pair.

The efficacy of this is solid with cruiser support wings chalking up a 66% win rate, well above the 48% of Vexors independently. However only 55% of the purely Cruiser Supported teams used Vexors, others might forego such tackle and rely on EWAR cruisers, where others have chosen to use options such as the Vigilant as heavy tackle bullies. In general, this trend seems to be independent of the pure T1 Cruiser surge we are seeing but has no doubt evolved from it in comparison to previous years.

 

Those of you watching over the weekend could not have missed the voting tool and the prizes given out with it. This is a tool that’s been developed by the wonderful Veetor Nara, who was also responsible for adapting my graphs for the twitch screen. So big thanks to the man for both this tool and putting up with my demands for bigger and bolder graphs.

Veetor has also been able to give me access to the data from day 4 of the EVE_NT coverage so we can compare the public perception to the results we saw.

So, well, I kinda sat back after this and realised that in contrast to my usual walls of text this is just a wall of fairly difficult to parse. Sorry, not sorry. What you have here, in alternating rows are the vote shares and then the match results for the 16 matches played on Sunday. We can see that 10 matches went with vote and 6 against. The most one-way vote was L A Z E R H A W K S vs Bright Side of Death with 82% of the vote going to the victorious P O W E R D U C K S.

The biggest vote upset, and reasonably one of the biggest upsets of the day was Pandemic Legion vs The Initiative. With a 76% vote share Pandemic Legion were the odds on favourite before finding their comp outmatched in the arena and losing 100 to 32 to The Initiative. The 76% vote share was the second highest proportion of the vote that any team achieved that day. With the next being fulfilled by HYDRA RELOADED over Hard Knocks Citizens, 75%-25% and 100-0, and the next after being let down by WE FORM VOLTA against A Band Apart, 71%-29% and 54-80.

Match titles are long, okay? But we can see the usual trends, a slow pick up as people watching tag onto the idea of the vote, a surge for popular teams, such as Pandemic Legion and a dip around the break, just after L A Z E R H A W K S vs Bright Side of Death. Pandemic Legion vs The Initiative is the highest point of the graph, with 1093 votes, 834 of whom were wrong. Knowing EVE players, however, there were no doubt players voting both ways.

 

Now, as I was told on each weekend, 17 times on the first weekend and 11 on the second, by 4 and 8 different individuals respectively; people like graphs of graphs. So here for a brief entry, I will include The Stats Blog of Stats Blogs. And we will kick this off with a Graph of Graphs.

We can see here that the Feeders Blog was a little all over the place in terms of the graphs it used, including-eww, must have been a mistake-a pie chart. The Week 1 graph then focused, apparently into going a little nuts on the doughnut graphs.  Week 2 evens this out and reintroducing the area graphs, with a better reason this time.

For any of you wondering these graphs are self-inclusive. As is the later word count, including The Stats Blog of Stats Blogs in its count.

Looking at the breakdown of graph types it’s obvious, and apt to see my favourite is the doughnut. Bar and column could arguably be bundled together but I often use them to represent subtly different forms of data based on instinct. Line graphs are ill suited to much of the data I track and the metrics I create as its continuous nature does not reflect the discrete data being fed into most of my graphs.

Finally, we see the comparative volumes of content in each subsequent stats blog. The increased word count of the Feeders can be related to the increased time I had to write it, while the increase in charts might be related to both the practice in doing graphs for the live broadcast and a need to break free from the confines of the stats tool, sorry Veetor.

 

Now before I wrap this up I feel the need to throw some more useful ATXV stats at you. We have seen 32 fielding of flagships from 24 teams. 8 of these have been from Flagships that remain in ATXV, across 7 teams, as said before only 1 is in the winner's bracket.

Of the 114 ships used exactly half of them have been used in more than 5 matches. Only 10 ships have been fielded without 1 ever being lost. Of these only 1, the Harpy has been fielded multiple times, seeing 4 team picks and 6 absolute picks without a single Harpy destroyed. Comparatively, 14 ships have gone undefeated, again most have only seen 1 match. The most matches seen of them is 4 from no other ship than the Rabisu, its status as an AT ship and so use by experienced teams with a solid plan no doubt helping keep it from defeat.

Here we see a breakdown of the 15 most contested picks, ordered by pick/ban rate. Typically the ships on this list make it in by a mixture of bans and popularity, however, some ships such as the Vexor, Daredevil and to a lesser degree Hyena appear simply due to their pick rate.

At the other end of the spectrum, ships like the Scimitar, Blackbird and Guardian appear simply down to being banned out. The Guardian’s bans come often from its potential role as a substitute for the Oneiros leading people to ban both in order to shut out armour logi. The Blackbird brings jams, that is all the reason and analysis needed for its high ban rate. The Scimitar, however, is an interesting inclusion, its counterpart the Basilisk is notably absent from this list, it is in fact 16th. Neither has seen huge amounts of love in ATXV.

With the Shield win rate at 47% compared to Armor’s 57% and use rates down at 26% compared to 60%, they are both less used and less threatening. This trend can be seen through most of the rest of the ships on this list with at least 13 of them fitting comfortably into Armor compositions.

We can take a look here at the performance metrics of the most contested picks. With most having strong win rates and survival rates, aside from the Logistics ships. The odd-one-out is the win rate of the Vindicator, 31.6% within an average of 53.9%, or 55.5% non-inclusive. Many of the ships on this list have their win rates dragged down by mirror matches, well not necessarily down but towards 50%, down for the better-performing ships on this list.

The Vindicator then is a great oddity, picking up a large number of both picks and bans for such a poorly performing ship. The two pick factors for this are its prevalence in poorly performing Hull All-In compositions and fear. Picked as a core ship for these All-Ins, it provides a huge amount of raw buffer and a stupid amount of damage, combined with the hard tackle needed to apply it. It is also feared perhaps beyond its value for that incredible damage output, the ability to take flagships, even in losses or otherwise throw in an otherwise unknown factor to peoples tidy theorycrafting will have it drawing bans in matches that it would be unlikely to win. In picks, there is also no doubt a factor of wishing to use the sponsored ship and a ship that is by many reckonings cooler to use than tournament staples such as the SNI and TFI, this trend is likely to disappear as we get to the pointier end of the competition however.

 

I will finish this off by briefly abusing my limited readership to send out thanks to them, a big thanks to all the teams competing in ATXV who make this all happen and everyone at EVE_NT for two wonderful weekends. There, I hope those 41 words can be justified. Or is it 50? Or 54? I need to stop now. Goodbye.