Friday 18 October 2013

Raffazza Presents: ETC Team Selection - Bulgarian Style!

One topic that never fails to get those interweb juices flowing every single year is ETC Team Selection. The outcries and/or grumpy mumblings on this subject is an annual sound emanating from communities as disparate as Canada, our beloved upside down colonial brothers and, without fail, the leaders of team selection grumpiness, England and the United States.

The clarion cry of grumbles signals the start of a new year where, along with inevitable international grumblings about comp (which will be resolved when the teams go to the Denmark ETC Warm-Up and end up copying Germany's lists) and the more localised (unless you’re American, that dirty laundry gets aired publicly) issues with selection.

This had me thinking, let’s have a look at how some countries do it, see if there is anything the wider world can do it. Team selection varies drastically country to country – some do it directly through rankings, some on past performance, some select their friends (allegedly! – seriously, I jest!), some take the previous year's team plus the winner of this year's Masters (ok, I'll stop now :) ) whilst some take any that qualify, and others focus on “team fit” etc.

To have a look at an interesting take on this I thought I'd get Bulgaria’s Captain to give us a rundown of their (impressively comprehensive!) selection process. A relatively new team, they are unencumbered by historical precedent and got to design their system from the ground up. And I think countries could learn a lot from it.

To explain the whole thing in more detail, here is their proud Captain, Hristo who… well, he introduces himself. Safe to say he is one of the nicest Dwarf players you are likely to meet (not to damn the man with faint praise or anything……) and he can be contacted on twitter on @fnarrr 





It seems like ETC 2013 has only just gone by, so it seems early to talk about organization and preparation for ETC 2014…but it’s exactly what is happening at the moment for many of the attendants!

Those of you who know me will do so either from my “10-from-the-back-take-it-off” shenanigans  resulting in picking up best in race dwarf trophies to a chorus of boos, or from my TWF/Twitter trolling exploits, second only to the great daddy troll Mr Sewell (all hail!). I’m also the captain of team Bulgaria WHFB for 2013, and so have the exciting task of designing, implementing and overseeing the team selection process.

A little bit on team Bulgaria – we are a relatively new team, hailing from a very small (by UK standards!) community – 2012 was our first year attending, where we managed our first performance target (“not last”) and even our second one (“above Wales”); in 2013 we reached the dizzying heights of midtables. As a trend, this should set us up as one of the good teams next year, and with a podium by 2015, but we’ll see how that goes.

So what is an optimal selection process for a WHFB ETC team? In many ways, working with a small community is a blessing – my expectations were for between 15 and 20 applications. Having witnessed some of the drama accompanying selections for Team England, I had a few requirements for the design of the selection:

1)       Process must be fully transparent
2)       Process must be fair to all involved

In addition, since we are both a very small community and working hard to expand the hobby back home, I had to keep in mind the following:

3) Process should not be putting people off participating due to excessive commitment                        requirements
4) Process should aim to allow/encourage new players to get involved, even if they understandably have less of a chance to make the team

After having some issues with hobby commitments in the previous year (understandably, when you have 9 people going for 8 places, you don’t feel under a lot of pressure), I had one last requirement in the design specification:

5) Process should penalize those who fail to put in the commitment needed (and remember design spec point 3!)

So what is the solution?


Team Bulgaria WHFB Qualifiers League 2014

I worked together with the previous captain Pavel to try out a selection system for 2013, and he suggested an everyone vs everyone league as something simple to implement and to work with – and even though there were a few hiccups, the core concept was sturdy enough to resurrect.

The core concept of the 2014 qualifier league means that each applicant to the team will play each other applicant once, under ETC comp on ETC maps using pre-submitted lists; the game will be scored on the 20-0 system, and the 8 people who are at the top are the new team. The process lasts 3-4 months (so about a game a week required) and serves both as team selection and as ETC practice – everyone brings filth, and the best rise to the top, selection done Dark Elf style! Easy! We did however add a few extra bits in there:


Unplayed games: last year we had issues with incomplete games; so this year a penalty system is employed for every missed game as follows:

1 missed game = -2 points2 missed games = -5 points (-7 cumulative)3 missed games = -10 points (-17 cumulative) 4 missed games = -15 points (-32 cumulative)5 missed games = disqualification!

The game itself is recorded as a 10-10. When disqualifying people due to missed games (which I hope I won’t have to do!) the ones with most missed games are removed first- along with their results – so if you have 3 unplayed games against people who didn’t turn up at all, by the time it gets to you those won’t be there! It also means that if person A takes person B off with a 20-0, and person B drops out, person A won’t keep that score – harsh but fair, everyone gets an equal chance at the punching bags!
Wanted the punishments for 1-2 missed games to be fairly soft as RL happens, but if you are on the verge they could easily cost someone a spot.


List changes: last year those were unregulated. Whilst in theory this is one big long tournament, so the same list should be used, in practice making someone stick with a list they realise they don’t like for 3+ months of gaming is neither conductive to good ETC practice nor user-friendly to newer players. So, you can pay some points to change your list – the cost is the same regardless of whether it’s a single item or the entire army. -5 points for the first change, -10 for any subsequent ones.


Minimal Results:  this is after all marking people on a curve, so we needed some minimum standard – which is 4 points per game average after penalties. If for some reason there are less than 8 people above this score, the ones that are select their teammates by vote.

So how does this design cover the specification I mentioned earlier?

       1)       Process must be fully transparent – Check. It’s a league; you win games, you get on the team. Simples!
      2)       Process must be fair to all involved – Check. The everpresent luck element of WHFB should average out reasonably well over 15+ games. New players and vets compete shoulder to shoulder, and all you get scored on is your game skill and some minimal commitment.
        3)       Process should not be putting people off participating due to excessive commitment requirements – Check. A game a week (excluding any practice you may want – but hey, that’s not mandatory!) is pretty relaxed for someone wanting to be on an ETC team.  As a side effect of people living abroad/in process of building armies, everyone applying has been told they need to be able and willing to use Universal Battle
       4)       Process should aim to allow/encourage new players to get involved, even if they understandably have less of a chance to make the team – Check. Booking someone to play a league game with you is easy, since those are the games people prioritize their time for; and you have access to the best gamers the community has to offer. It is possible to change your list every time if you don’t really care about actually being on the team and don’t mind the penalties.
       5)       Process should penalize those who fail to put in the commitment needed – Check. Penalty system should take care of this.


Lessons from the previous league

I’ve already covered non-attendance and list changes. There is one other item which was an issue last year, and that was new books – aside from the list changes they prompt, they also mean that I need to comp them in line with the current ETC comp pack.

Luckily, this year the start of the qualifiers is right after the DE release, and the end date is shortly after the release of the next book (fingers crossed for Dawi early next year!) – just before the new comp pack draft 1 is released. The obvious solution was to write comp for Lizardmen and DE quickly, and put a blanket ban on using any new books if such happen to be released in that time.

Writing comp for WoC last year, I severely underestimated their restrictions – so for Lizardmen this year I called on some help from the experts (including the owner of this blog) to end up with the following:

Unit Specific Restrictions: max 1 Salamander Units max 3 Skink Skirmishers max 2 Mounted Scar Vets max 3 Stegadons
General Restrictions: max 3 Flyers max 6 non-character units under130pts & 20 models
Magic Restrictions: Cube of Darkness, Becalming 1 DD each
FAQ: Primal Fury does not work on supporting attacks Pirahna Blade only works on close combat attacks
I didn’t quite have the timeframe (or expertise to consult, since it’s a brand new book) for DE, so after some discussion from teammates released the following:


DARK ELVES
0-3 Dark Riders 0-3 Hydra/Kharibdyss Combined 0-4 Pegasi/Sea Dragon Cloaks combined 90 Shots over 12" max  count as 5)

Engine of the gods: burning alignment targets units in Arc only
Other plusses of the qualifier league

In addition to meeting my design spec, selecting with a league has a few other benefits. I’ve already mentioned it acts as practice for the ETC itself, under similar pressure. It allows opponents to book games with each other at times convenient for both, lowering the barrier to entry for the team. Last, and more important, is that it should result in what is objectively the best players for the team – wunderkinds are given a chance to shine among the veterans, and has-beens are ruthlessly removed. Even as captain, I’m not guaranteed a spot – I’ll be playing in the qualifiers along with everyone else!

Problems with the qualifier league

The process isn’t fault-free; luckily a large proportion of those issues can be easily policed in a small gaming environment like ours and dealt with if necessary:
-          There IS a luck element, and if you are not a candidate for a podium it could knock you out of the team. Not a massive issue, since I’m not sure how that is different to WHFB in general, but it is there.
-          System isn’t really applicable with any more than 20 candidates, 25 at a push. It can be used as a second round of selection after other preliminary requirements in larger communities though.
-          Penalty system is abusable – I will be policing the below examples (among others) with warnings followed by penalties, but it would be better if such loopholes weren’t present:
o    If players collectively decide to remove someone from the team, they can refuse playing them – incurring small penalties for themselves, but effectively disqualifying them from the team
§  As games are set up throughout a central Facebook group, it would be fairly obvious if this happens, as you can see a person’s “anyone up for a game” posts go unanswered
·         We are all good people and this won’t happen in practice ;)
o    List changes can be made tactically to play the meta – selecting certain opponents to play before others, etc
§  Again, easy to see if someone is excessively dodging people. List change penalties are harsh enough to stop this to a large extent

Afterword
As a cherry on top of the best selection process ever, the 8 players who qualify get to nominate and elect a captain. The qualifiers began as of 14th October – if you are curious, Raf can post up the lists


Hristo








An interesting and well thought out system, I think. One that could, in my opinion, be used as a secondary round of qualification for bigger communities and remove a lot of the "x player is not good enough to be on the team" mumblings you hear every year. You say no to anyone you think is going to be detrimental to the team and then get the remaining players to confirm they deserve to be there.

A question occurred to me though: this system is clearly geared to finding the “best players” rather than “best players with armies x,y and z”. This is an approach that divides opinion globally, and I thought I’d check if it was intentional.

Connected to this, I wondered if there was chance that results could be skewed by less skilled players picking up a “net list” (for example the German DoC list) and rolling over players who are maybe using their normal armies? And is this an issue if so? 


               
               
Yep, well spotted. It was intentional, and heavily supported by UB use - since there you can use whatever you want in terms of models.

I do believe specific army expertise is very valuable, but I don't think this discussion is applicable in this case for a few reasons:

- army selection comes after team selection in terms of timing, due to AR release schedule, and having a team of army experts locking down the selection before comp release is a problem.

 - ETC comp, as hated as it is, does produce the best army balance I have witnessed in a comp pack. It’s got its issues, and you can complain it restricts your army build, but there is little ground to complain you are being pushed out of an army book as a whole. I.e. the army you like = netlists in powerlevel.

 - it is awkward to select the 'best player of army X' for your team after army selection, because among other things, your selection metagame would only contain the 8 armies you have allocated team slots for - so I doubt I'd go down that path even if I had the final comp available to me right now.

My personal opinion on whether person A (best with army X, but midperformer average) is a better player than person B (better performance, goes for latest filth & netlists) is probably highly biased :p luckily its irrelevant due to the practical concerns above.


In terms of netlisting - not a massive concern; partly as I mentioned the comp keeps all armies at roughly the same power level; and secondly the standards for success are higher. Formula can be boiled down to:

Success = Player Skill + List + Luck, with PS being relative to the List used.

Cutting Luck out (hopefully law of averages over time), anyone using a netlist they are inexperienced with is compromising their PS, which if they were new is low in the first place anyway; as only the top 8 qualify, the standard they presumably set will be of max or near max performance lists (be it netlists or own concotion) + the max PS they can bring - so neither the netlister nor the person bringing a terribly suboptimal list they are good with should make it in. In reality, with a small pool that isnt the case, but after all list design is a necessary skill - so if you design cack and get taken out by the netlist, its your own fault!




So, there you have it – ETC Team Selection Bulgarian Style!

Let us know your thoughts as ever either on here, Twitter, in person, or by howling at the moon – whatever floats your boat really!


Until next time


Raf

2 comments:

  1. Interesting approach and one which probably could have been adopted by Scotland over the last couple of years. We've typically had very few 'serious' players willing to spend the cash to travel to foreign parts and commit the time for all the ETC nonsense which surrounds actually playing 6 games of toy soldiers over 3 days.

    That said I don't really know if it would add anything to the selection process. I imagine if we were to ask Hristo he could likely tell you who the top 8-10 players were on that list. And given that knowledge and the knowledge of which armies they typically play with you could pretty quickly find a shortlist of 8 without playing a single game. That said having it decided through a league format removes any issues around 'old boys network' or similar cronyism complaints. Play well = get in the team certainly has some merit.

    Scotland's selection process may lack some of the transparency of the Bulgarian approach but what it lacks in transparency it more than makes up for in Simplicity. The ETC is founded on the basis of the teams that have been before get to vote etc so we've basically stuck with that as the underlying principle. The team who have been the year before elect the captain for the next year. That captain then gets to pick the team. Simplez!

    ReplyDelete
  2. I've certainly got my predictions; to what extent they are accurate can be reported here in a few months time. One issue with such opinions is that, being opinions, they are grounded in personal bias; another one is that they do not respond well to the power fluctuation in a small but rapidly growing community - with many new players entering the scene, I don't know who is better than whom; and us traditionally having just enough people for a team, any dropouts from 'vets' will result in newer players on the team.

    So, the 'wunderkinds' get their chance much earlier - rather than having to win against the 'vets' for a year before they start being considered - and this keeps the previous players on their toes. I'm certainly worried about not managing a spot!

    From a captain's perspective, I'm not sold on the idea of handpicking players; and sure as hell wound't be happy with it if I was a player who was overlooked! I think having it as a tradition helps, as everyone knows what the deals is, and there is a history of it working successfully; are you however not concerned that players will be inclined to select captains who would pick them, and vice versa?

    For small communities, in all honesty the Scottish approach would probably work for us without too much hassle and drama; but I wanted to lay some groundwork which can be applied to a larger pool of players, where 'old boys networks' whispers become more of an issue. Hopefully, as we adapt our methods to more and more applicants, they will all stay 100% transparent, even if a full-on league is no longer appropriate.

    ReplyDelete