Kirb your enthusiasm!


"Pink isn't a color. It's a lifestyle." - Chumbalaya
"...generalship should be informing list building." - Sir Biscuit
"I buy models with my excess money" - Valkyrie whilst a waitress leans over him

Monday, February 20, 2012

3++con - The Initial Statistics Dig

So I'm sure many of you are tired of hearing about 3++con by now - hell I'm tired of looking at it :P but we might as well start looking at some of the statistics we uncovered and a lot of the subjective interpretations one gets having been there for the entire event (unlike when I did the statistics for NOVA or 'Ard Boyz).

First, let's look at the army break-down with Wins & Losses.

Armies Number Wins Losses
Black Templars 0

Blood Angels 3 11 7
Chaos Daemons 1 3 3
Chaos Space Marines 1 2 4
Dark Angels 1 4 2
Dark Eldar 6 21 15
Eldar 2 7 5
Grey Knights 9 27 26
Imperial Guard 4 12 12
Necrons 3 8 8
Orks 6 22 14
Sisters of Battle 1 2 4
Space Marines 5 13 17
Space Wolves 5 13 17
Tau 2 5 7
Tyranids 2 5 7

In total, 51 armies. I have not included the stand in player and unfortunately, one of our Necron players did not show up Day 2so they don't have a full array of statistics. What we see here is the majoirty of the 5th edition armies are the most popular - IG, SM, GK, SW, DE are all vying for the most number of participants with Orks thrown in for good measure. Necrons and Blood Angels both seemed to have a lower representation and this was perhaps due to Necron's newness and the smaller sample size available to us. The rest of the armies come in 1's and 2's and most of these armies are pre-5th edition or armies which are commonly regarded as poor 5th edition books (Tyranids & Sisters of Battle).

So, before we look at the Win/Loss statistics, let's look at some overall statistics from this grouping. A common and understandable complaint of 5th edition is the Marine prevalance - with so many of the Marine books being good when many of the Xenos books are not, it's pretty easy to understand why and despite their vast differences, it can be a bit boring playing them. Let's look at such a breakdown:

Marine armies: 23/51 (45.09%) - this is quite low for 5th edition where many tournaments see 60%+ of Marine armies. Even if we include Sisters of Battle and Chaos Space Marines that only takes us to just under 50%. This indicates quite a bit of variety in armies seen at 3++con compared to other tournaments we've looked at and is perhaps something more specific to the Australian comp-heavy tournament scene rather than an indication of a moving away from Marines. Whilst more Xenos armies are certainly relevant than before (Necrons & Dark Eldar), we're seeing a pretty even spread here. With that in mind, let's look at 5th edition book usage.

5th edition armies: 38/51 (74.51%) - a much bigger margin and a relatively decent indicition of the popularity of the new books. Whilst there are more new books than old now (9 vs 6), there is a greater proportion of player's using the newer books. Whether it's a case of newer is better, recency effects, simple chance or some other factor is largely impossible to determine. All we know is they are used more often than raw statistics indiciates.

So let's look at the raw army stats and keep in mind our much smaller sample size than normal. With NOVA we often had 20+ armies to draw statistics from - we don't have a grouping of even 10 armies from 3++con so results are far more likely to be skewed and thus invalid.

What we quickly see are the older books and disregarded 5th edition books (Tyranids & Sisters of Battle) lamenting down the bottom of the pack but there also seems to be relatively small differences between the overall army winning percentages. Let's remove the sample size = 1 populations to get a more accurate picture.

Again, we can see very little difference overall between the armies (around 20% from the best to the worst). A random statistics I have always maintained is that a balanced army should always have between a 40-60% chance of winning against any other army, so overall army statistics falling into that category for me indicates something has gone right.

That being said, there is an obvious difference between the armies doing better than 50% and doing worse than 50% yet what is surprising is Space Wolves and Space Marines being in the under 50% pile and Orks and Eldar being in the over 50% pile. Eldar only ran with two armies and both generals attend tournaments quite regularly and are very experienced with their armies. We could therefore label this an anomoly due to the small sample size and greater than average experience of the players behind the armies. Orks however have now shown themselves capable of competing at NOVA based events consistently in terms of winning. However, whilst their win percentages have been relatively good, these wins are generally done in the middle brackets (at NOVA) and even when multiple Orks made it to Bracket 1 for 3++con - none of them made it to the final.

With the continued and documented relative success of Orks across multiple tournaments and regions, it does lend strength to the claim that Orks are capable of trading blows with the 5th edition books but over several games against good opponents, are more likely than their 5th edition counter-parts to fall apart simply due to lack of codex choices and options on the tabletop.

However, none of this begins to explain why Space Marines and Space Wolves did so poorly. Space Marines didn't have a good showing at NOVA either which could be due to several factors, again. Due to the release of lots of Space Marine armies - the more serious competitors might have decided to move onto a Marine army which is better competitively or due to the Space Marine's often 'bland' nature within their armies, something which suits them more personally. However, it does seem that Space Marines are lacking a bit in relation to other 5th edition books (other than Sisters and Tyranids) and might slowly be regressing backwards as they begin to age.

Space Wolves though... well it doesn't make any sense really! Let's dig deeper and look at the top three Brackets, their army breakdowns and how the armies went there overall.

Armies Wins Losses
Space Wolves 9 3
Orks 17 7
Blood Angels 8 4
Dark Angels 4 2
Eldar 4 2
Imperial Guard 8 4
Dark Eldar 15 9
Grey Knights 14 10
Daemons 3 3
Necrons 6 6
Tyranids 3 3

And of course the first thing that jumps out at us is Space Wolves are now on top - and one would expect so with the winner of the tournament being a Space Wolves player. We still see the majority of 5th edition books doing well (though Space Marines are still absent) and surprisingly the Orks are still up there in terms of winning percentage. You'd have expected them to drop down the field whilst looking at the top brackets but they maintained their strong presence.

This is where some subjective analysis must come in I'm afraid. Firstly, the Orks never played each other in the top bracket, so they always had chances of three wins. Secondly, Julian Miller (8th place, Necrons, 3-3) who was undefeated on Day 1 and a very good player, was very hungover and under the weather. He played all three Orks and lost all three games (after stupidly pointing out he'd never lost to Orks in 5th edition). This is not taking anything away from the Ork players Tim, Janinda and Luke as they were the better players on the day but if those three wins had been turned around into losses, we might see the Orks and Necron places in the above graph swapped around. Again, we're still seeing Orks creep into the top so there is obviously something there, but they aren't able to consistently crack good armies run by good generals to win these type of tournaments. This is just pointing out their impressive standing here might have been, in part, to a factor one simply cannot accomodate for.

First Conclusions

So what is this all telling us - particularly the non-number oriented of us? 5th edition books are, by and large, a very good step in the right direction. We saw this at NOVA, we saw this at 'Ard Boyz and we're seeing this at 3++con with a much smaller sample size - 5th edition books (excepting Tyranids and Sisters) all do relatively well in a tournament setting with roughly equal winning percentages. The older books are often left behind and find it difficult to gain consistent wins against the newer books in such an environment as evidenced by all the brackets (bar the bottom one) being won by a 5th edition book.

Further things of note:
  • Space Marines seem to be regressing - whether this is due to the player's using them (not moving to better books, playing fluffier armies, playing for the sack of nostalgia, etc.) or the book itself is hard to determine but we're not seeing Space Marine players crack into the top brackets very easily.
  • Orks are clearly the best of the rest and often slip up onto equal footing with the good 5th edition books but still have trouble pulling out the tournament wins in these situations.
  • Grey Knights are still not being shown to be the overpowered codex people think they are. Whilst a lot of people play them which pulls their winning percentage back due to poorer player's using them, even in the upper brackets we're not seeing Grey Knights dominating and sweeping all before them. In the above graph we see Grey Knights are 8th out of 11 armies represented in the top three brackets and if we remove the sample size = 1 armies, they are 6th of 8th. They are good certainly and are slotting in nicely with the other 5th edition armies but are not OP cheese.
  • Dark Eldar and Necrons are carrying the 5th edition Xenos torch. Although Orks are clearly the Xenos army of 3++con with three players in the top bracket (four in the Top 3 Brackets overall), Dark Eldar shown to do consistently well in both the middle and top brackets across multiple tournaments. This is highlighted by Bracket 2 & 3 wins at 3++con. Whilst Necrons don't look to have done well on paper, with a small sample size and external factors, this is hard to judge but two of the three did make it to the top Bracket. More information needs to be collected to see if Necrons fall into the same boat as Orks or if they are like Dark Eldar and are legit.
 I'll be doing some more digging later but I haven't inputed all the data yet (i.e. who went first, objectives won compared to lost, what armies did best on which scenarios, etc.) so I will get to that. What I'd really like to do though, is somehow run a matrix which can determine a player's competitive score ranking or something so we can compare how difficult a player's run was. I.e. Rupert Pupkin played the four of the top six players in his last four rounds and was himself ranked 4th indicating his 3-3 status would be more significant than someone further down the order.

If anyone has any ideas on how to do this, please let me know.

Follow us on Facebook!

Related Posts Plugin for WordPress, Blogger...