GentlemanJon — Before i46 there was a strange sense of malaise in European TF2. As usual a string of legendary players were threatening to quit the game forever, and complacency over the inevitable crushing of the American interlopers was rife. It ended with partial mix Euro teams more focused on getting pissed being wiped out by the highly motivated yanks and sowed the seeds of the reverse for i49.

Last time the North Americans were complacent sending only one team and were swept aside by the twin giants of European TF2, the ultra-professional Epsilon uber team and the band of brothers Broder. It was time for the NA team to make the excuses, it was the Quick Fix, Cyzer sucks, etc…

So what state is EU TF2 in this year? Epsilon, stripped of core members, have come out on top again but in far from convincing fashion that leaves more questions than answers. How big is the  fl1p-shaped hole in the AwS squad? Did Epsilon really improve or were AwS just weakened? Is there even a decent EU team this year? Was Darn! any good?

Baseless hyperbole along these lines doubtless follows.



The graph shows damage per minute in blue, K/D ratio in red and the player rating in yellow. The K/D is measured against the top axis while DPM and rating against the bottom axis. Onto the outrage…


Calm down kids, Kaidus seems to have morphed fully into a kind of roaming Demo, or channelled the spirit of Exon, while his strong man soldier team mate eats up the medic’s attention, a scenario familiar in the NA game. You have to wonder if the relative lack of spam support may have hurt his team but this really only illustrates that his main role was not to be the team’s battering ram and choke spammer. He was the demo most likely to get his kills on the enemy medic, as you might expect with his style and lack of pocket role, closely followed by War and Hys. Ryb and Smirre got the highest share of their kills against opposition demomen.

In fact all the big name Demos posted lesser stats this season compared to some demo led teams. The perennially underrated Huhy posted excellent stats based mostly around enormous spam damage, but only played 4 games although one was against Epsilon. His style seemed to suit Rockit better, at least for generating his own stats. Hildreth posted a similar profile but wasn’t able to quite match the damage output and KD, despite his calling duties allowing him the luxury of baiting his team.

Despite playing a sticky nerf affected game Death still managed to post top stats but if any player is the centre of their team’s whole strategy it’s him. His KD is clearly well ahead of his damage as he prioritised survivability, playing the cleanup demo, but was no slouch under direct pressure either as I thought he might be this season. Ryb and Kaidus also had their KD well ahead of their DPM, although possibly for different reasons.

The Epsilon revolving door candidates Hys and War are as evenly matched as they have been in other seasons with War coming out a little ahead despite also playing in a sticky nerfed game. In previous season though they’ve both been spam heavy whereas perhaps playing a little against type they both exhibited a greater focus on survival this season. In War’s case this may be partly down to Rockit’s sluggishness in transitions and pushing advantages.

At the opposite end of the stylistic spectrum are GDK and Smirre, aggressively spam heavy racking up damage but at the cost of survivability.


This illustrates the relationship between damage, kills and healing for the individual players. The yellow bar is the average healing they received per minute and is measured against the bottom axis. The blue bar is the amount of damage they had to cause to register a kill or assist and is also measured against the bottom axis. The red bar is the number of kills achieved per minute, which is measured by the top axis

Hys and Death record the fastest practical application of their damage posting by far the fastest kills per minute. Death we know is very much a clean-up man for 4-25 but as well as being formidable in death match in his own right. Was Hys baiting his team out a little or was he benefitting from being part of the Epsilon machine?

Kaidus and Smirre inhabit the 2nd tier of kill rate. From his heals received you can clearly see that Smirre is the most pocketed demo in the league, almost received more heals per minute than damage per kill, so was able to be aggressive. Kaidus as the least healed demo gets his kills from extremely precise damage focusing clearly on deathmatch, posting the lowest damage per kill & assist. Of course this may also be a result of playing near Stark as his disdain for wasted damage spreads by osmosis. It is important to remember the extent of Kaidus’s pursuit of well protected targets in assessing these stats.

On the subject of spam GDK is most guilty of directionless spam needing to cause over 450 damage before registering a kill or assist which goes hand in hand with the slowest kill rate. Huhy and Hildreth who also top damaged similarly didn’t match their damage with their kill rate, Hildreth being a more efficient killer whilst receiving a little more heals than Huhy.

War and Ryb are twins in this chart, appearing rather anonymously in the middle of the pack, neither death matching particularly strongly or spamming that hard either.

Weapon Use


As usual War has the highest share of grenade kills getting around a 3rd of his kills via the pill. Hys, Ryb and Huhy also showed a strong bias for an alternative career in plumbing, and this group may have the best chances for continued relevance in the wake of a future Valve sticky nerf (you know it’s coming).

GDK is by far the most reliant on stickies but Smirre and Death show a strong preference for the win bombs as well, with Hildreth and Kaidus sitting anonymously somewhere in the middle.



As usual it’s an Epsilon soldier in the top spot but not perhaps the one people might expect as KnOxXx received considerable censure for his play during the season. He does of course have the advantage of being the uncontested overlord of Epsilon so naturally calls as he sees it giving him the advantage of fleeing certain death first. Even so that is a considerable lead in aggressive damage.

Mike played roamer for several games this season and even with his superhuman soldier skills it’s unreasonable to expect him to post top stats, even so he wasn’t far behind where he is now when he played roamer, and he matches anything Gear had previously achieved.

Rounding out the Epsilon trio Tek crashes the party for a few of the lower ranked pocket soldiers but doesn’t match the exploits of Gear in past seasons.

Zebbosai is the only soldier matching the Epsilon pocket pair but he took every last drop of uber and every healing point he could get his grubby rocket launcher on to do it.

Amongst the best of the rest Klar and Zoob shared similar damage, but Zoob enjoyed better survivability and of course much smoother relations with the rest of his team.

The old school pair dragged rudely from the cosy fire of their retirement home back to the white hot heat of battle for Rockit acquitted themselves pretty well. Hocz managed mid table amongst pockets but Darn! was one of the better roamers still able to put those pesky kids in their place despite a total refusal to jump in on Viaduct, or in fact play the map seriously at all. In a team bereft of effective calling his contributions from roamer helped Rockit’s play too.

Lazy Pandas soldier experiment importing Muuki from Highlander Prem (and up about 4 divs from the previous 6v6 seasons) proved to be a success, the mercurial medic killer proving he belongs at the top level of European 6v6 posting good roamer stats whilst continuing his style of relentlessly going after the high value targets.

Rag! at least bets his own roamer but otherwise it’s a season to forget for the Saints pocket, and cAPS too.

Celebrity roamer ShaDowBurn did well but wasn’t decisively separated from the pack in terms of production. There are no points for jump map skills here unless they get results, although the result of killing Hys repeatedly on Granary middle before it even started was worth seeing.


Looking at the healing it’s striking just how many Zebbosai took on, but he hit the highest kills per minute of any soldier, his damage being used more efficiently than KnOxXx. In fact KnOxXx’s damage efficiency is on a par with a number of roamers and struggling soldiers revealing what the criticism of the Epsi legend was all about. The margins at the top are very slim.

No pocket soldier used their damage with more efficiency than Zoob who posted the 3rd fastest kill rate. The only soldier of any stripe to do so was Darn!, it’s like a look into TF2 pre-history where constant spam didn’t dominate the playing field. Despite this the only roamer to get a faster kill rate than the Finnish fossil was Tek. Look and learn youngsters.

Hocz on the other hand struggled to get a good kill rate as his low damage output combined with mediocre efficiency doesn’t amount to many frags, although as with War this is probably at least partly down to Rockit’s general problems transitioning aggressively enough.

Forsak3n plays an unusual game receiving more heals than a roamer but fewer than other pockets. It was postulated in commentary during the season that he got bored of playing the steady damage tank and in terms of aggression and durability he only ends up slightly ahead of his roamer, and behind in terms of kills per minute.

It’s notable that Muuki did his work on next to no healing, Shyyou got the most healing behind Zebbosai but didn’t even have a positive KD, and Salmon struggled to shake off the rust pocketing for FNX.



Kiler4fun completes his rise to the top of ETF2L by confirming his position as top scout this season. However he didn’t enjoy total statistical supremacy with Yangpie playing insanely aggressively racking up the highest DPM of any scout but at some cost to his KD.

Sheepy’s position is slightly exaggerated by dodging the top dogs during the season as is Skappa’s.

Fl1p and Stark are very evenly matched revealing how strong the scouts were for Aws, and also illustrating the size of the gaping hole fl1p’s sudden departure leaves AwS dealing with for Lan.

Bash on the other hand posts only mediocre stats, his role not as successfully executed as in previous Epsilon teams. It seems that for most of the season AwS enjoyed scout superiority, although this may be down to claims that Bash spent half the season adjusting his sensitivity.

Smzii and gf18 round out the top tier of scouts with Smzii playing the slightly more aggressive style.

The Hildreth Highlander promotion experiment fared significantly less well in this area with Tracker bailing out after one game and Lazybear eventually replaced after posting lacklustre stats, resulting in Lazy Pandas scrambling for some more experienced players to save their Premiership status.


Stark claims supremacy in kill rate on almost no healing and as you’d expect is highly damage efficient while doing it, although unexpectedly kensi has the most efficient use of damage (also on very low healing).

Bash’s contribution looks a bit better in this regard but he’s still lagging behind the elite players and indeed several eastern Europeans. However not shown in these stats is that he got a far higher share of his kills against the enemy medic (and the highest against the demo although not by the same proportion compared to other players), which indicates that high value kill suicide missions were significantly higher on Bash’s agenda than most.

Cookie flexes the old dm skills a bit more clearly in this chart.

Lazybear was the most wasteful with his damage and got the lowest kill rate. I feel like I’m beating up on the guy a bit, but I kissed his arse in HL articles so he’ll have to take the rough with the smooth. Yangpie, wltrs, atomic and alba also didn’t finish off the damage they put out as efficiently as their peers.



Yay, something a Saints player is good at! Dennia manages to rack up the sniping stats by having plenty of still living enemies throwing themselves at him after his team went down.

Tek is the best player in a team that actually kills things, also racking up the fastest headshots per minute. This prowess could be critical at i52, as he can’t be expected to hold the flank Gear style, and there’s an Australian Ovis Epsi will probably have to face at some point.

Kensi is the other outstanding head shotter.



In this graph the blue bar shows the kills and assists per death and is measured against the bottom axis, the red shows healing per minute and is measured against the top axis and the yellow shows assists per minute also measured against the bottom axis.

Mirelin has clearly the highest assists per minute but that’s expected having his heal beam stuck to the fastest killing soldier this season. Similarly “Can’t milk it” Raymon matching him most closely is down to KnOxXx’s similarly high kill rate although it seems Mirelin’s heal rate suffered either through total focus on Zebbosai or possibly his calling duties (or dicking around with his crossbow from half way through a 5-0 win).

Raymon’s massive KAD advantage is probably down to playing GGWP and the two bottom teams this season so he rarely died.

Ondkaja gets the award for most thankless role played this season, and also best reading of The Very Hungry Caterpillar. Unfortunately he’s still waiting for his team to transform into a thing of beauty.

Kissakala’s step up to Prem doesn’t seem to have been a problem, but then very few of the key medic stats rely on the scout’s performances which is where his team suffered. Stevepander’s holiday from Pyro didn’t go so well.


Stevepander finds himself in fine company, with Raymon and Mirelin dropping uber almost as frequently as he did.

The thing that confuses me here is that Whiteglow NEVER drops. Some mistake surely? Other than this superhuman feat (or probable evidence of my incompetence processing uber drops) Skeej actually drops least with KnOxXx close behind – no doubt he was robbed in the final.

KnOxXx, Admirable and Raymon stand atop the ubers per minute although the differences are so miniscule it doesn’t really mean anything.

Team of the season

It’s an interesting season with many teams in flux as there were a lot of changes in personnel and styles which gives me a chance to pick someone out of leftfield on Demo and go with Death. Last season he wrecked Div 1 and I thought his style would be found out in the top division but he’s still flourished. Yes, he gets the benefit of his team playing around him but teams should play around their best players and give them the chance to shine. The usual suspects weren’t so impressive stats wise, plus I like a nice simple name based on a concept. War, Death, anything along those lines.

If I was picking an alternate it would be Kaidus, it’s incredibly hard to post competitive stats playing a looser style and if you ignore damage his kill rates (arguably more important) were right there. You need extremely fine judgement and it opens up heal hungry soldier options.

The soldier position is a little tricky. You can’t pick against Mike but his stats make him look poor on pocket and the fact he ended the season playing pocket makes his roaming performance untrustworthy. Add to this that I’m looking for an excuse to shoehorn Darn! into the equation and it’s tempting to avoid the obvious selections but it’s got to be Zebbo on pocket and Mike on roam.

On scout I think a straight up choice of the top two players Kiler and fl1p would be perfectly good, but I’d be tempted to switch fl1p out and let Stark play with no heals. Although I’d champion Sheepy as one of the stronger scouts playing at the moment his position in the stats isn’t truly representative so gf18 and Szmi would be my choices behind the top players.

Skeej is sneaking into contention for Medic, very low drops, good assist rate (with no top kill rate player to heal), good heal rate, impersonated expertly by Mirelin. Fuck it – Skeej it is, unfortunately there’s not even a pixel trophy to celebrate the feat.

Most competitively challenged team of the season

Although I don’t usually indulge in this kind of negativity, it’s extraordinary that so many members of the same team, Saints, qualify in this area. Smirre manages to look half decent on Demo (although he had massive heals) so gdk is probably the best candidate there and the unfortunate Lazybear could play instead of Dennia. But uncontested are Ondkaja, rag, cAPS and alba. Loss of motivation forever.

So… i52 then

It seemed that AwS were on target to at least take Epsilon to a very close finale, but the sudden loss of a top scout was never going to be easy to deal with and overall Zappis’s performance in the final was mediocre despite showing flashes of brilliance. Fl1p held his own better in the regular season games (although admittedly they played different roles) without being spectacular so you have to assume that the final would have been closer with him playing.

Given this perhaps the Epsilon resurgence wasn’t quite as convincing as we might have been led to believe. However it looks like they will still be considerably better balanced than the Epsilon lineup that lost at i46 (i.e. the demo can do the rollouts) and while it’s not quite super team territory their new found keenness should see them turn up ready.

On this occasion Mixup should find themselves faced with at least one well drilled professional unit forged in the heat of league battle ready to play rather than a bleary eyed hung over mix team trying to remember where they are, an experience some of Froyotech got at i49. However with clockwork back in the mix Froyo will be confident of going further than last year, as he regularly plays one game per ETF2L season and posts what would be league leading stats.

AwS it seems are heading towards a low point just at the wrong moment, but their roster is still stacked with talent and just like a couple of the NA i52 teams their struggles can be overcome if they make the right decisions. At the very least at Lan they should be immune to the hardware and internet problems that have plagued them recently freeing up two of their top performers to play to their best. Judging from interview comments they may be vulnerable to the incubus of doubt interfering with team structure and calling discipline under pressure, whereas Epsilon are completely solid in that area.

It seems despite the inconclusive recent results there is a dangerous air of expectation around the EU scene as online specialists (hax anybody?) XTS cause chaos amongst the top NA ranks. I don’t expect it to represent the likely performances at i52, both teams will be hungry for different reasons – Mixup to set a record that will probably never be beaten and Froyo to even up the NA score at international Lans – while the domestic leagues will be something of a distraction or an opportunity to experiment.

Also I hear there are some plucky colonials trying their luck again. Oddly it appears the logic for dismissing their challenge this time around is that we know they’re good, so “we’re ready”. Admittedly I don’t see them winning it, but if it’s true that they’ve significantly upgraded scout and demoman then that’s all that would have been required at i49 to push them much closer to the final. Take them lightly at your peril…