I was inspired by a thread over on Dakka about what the top tournament armies are in 40k. I think it's an interesting debate that is skewed by a lot of flawed data or assumptions. I think a major fallacy in the argument is citing anecdotes or personal experience about who is good with what army. Generalship is never relevant in these arguments. We're talking about codex prowess here. Assume in the discussion that all players are equal and then make the predictions.
The other logic flaw is citing tournament statistics. Volume of wins doesn't mean a whole lot because some armies are a lot more popular than others, and just because they are necessarily good. Yeah, quality codices can be popular, but other factors are involved: Finances, aesthetics, fluff, availability. These all play a factor. Chaos Daemons is an army with lousy model support that proves expensive and difficult to build. Space Marines have extremely popular fluff and wide plastic model availability. They show up at tournaments a lot more frequently.
The other issue is that tournaments all play by a different set of rules. Some are comp heavy which basically gives weaker armies with a handicap. Some are soft score heavy which can often skew results. Even generalship awards sometimes don't mean much because of bizarre pairing methods or massacre scoring. I have repeatedly pointed out the flaws of those kinds of systems here.
As an exercise I decided to put compile some data of my from our own tournaments. I aimed to get a handle on what the top tournament codices look like so I put together a table like so:
Army Entries Points Average Tournament Wins
Tyranids 3 8 2.6 1
Blood Angels 5 11.5 2.3 1
Orks 5 11.5 2.3 1
Chaos Daemons 5 11 2.2 0
Tau 2 4 2.0 1
Chaos Space Marines 6 11 1.8 0
Black Templars 2 3.5 1.8 0
Eldar 2 3.5 1.8 0
Space Marines 6 8.5 1.4 0
Imperial Guard 5 5.5 1.1 0
Space Wolves 3 2.5 0.8 0
Dark Angels 1 0.5 0.5 0
Dark Eldar 1 0 0 0
Witch Hunters 0 0 0 0
Daemon Hunters 0 0 0 0
Necrons 0 0 0 0
This illustrates data from our 4 tournaments this year. It shows the number of times an army entered a tournament and gives their results and then we create an average score.
Does this tell us what the best armies are? Not at all. One major problem is these armies are piloted by different people. Generalship is not equally stated here, and I think that's an important point when looking at tournament results. On top of that the sample size is very small. Only 4 tournaments and 15 rounds of 40k.
Still the data is there to be viewed and we can learn a bit from it. We see what armies are popular around here (Space Marines, IG, Orks, Daemons). We can see what isn't (Necrons and Inquisition). We can also see what is posting results at 1500 points in a soft score free tournament. Worth a look anyway. I'll start putting together additional data as more rolls in. The better the sample the more valid the results.
You know, you could probably factor in the generals with the armies, you have ELO ratings for everyone. It would change the results, since some people like me and Ryan play different books in different tournaments.
ReplyDeleteAlso, you keep track of the army lists that people bring, and what missions - do you keep a database of what armies played what in each mission and what the result was? You could make this pretty detailed and interesting with that information.
I do have all that info Mike. Most of that is actually published on the tourney result pages except for maybe the first one we ran in February.
ReplyDeleteIt would be cool to consolidate that data into some useful information though. Completely agree with you there. Question is? What data do I use and how do I filter it into something meaningful?