Number of BG in the Britcon armies

[petedalby]

I'm sorry to hear that things continue to decline in Spain Julian but I'm still struggling to understand the problem in the UK?

Numbers are up and formats are being varied. The Derby Teams features a 900 AP Eternal Empire theme which - and I agree with Nik - is great news - more toys on the table!

Event organisers have the opportunity to mix things up if they wish to.
Yes - change the table size if you want - why not?
Change the break conduitions if you wish to penalise larger BG armies - why not?
Oxford a running a comp which can't feature Skilled Swords. The US are running an 'Average Joe' event. Farnborough are running a scenario based comp in September.

All these things can be achieved wiithout changing the rules. Stop worrying what the perfect army / number of BGs is - just enjoy the game.

[hammy]

thats sounds like a game I would liked to have seen! blood and snots left right and centre! good stuff!

jon

If that's what we want then you need to incentivise bloodbaths over stalemates somehow. Previously there has been concern over fixed results in that arena, because it isn't a zero sum game.

Bloodless draws are hardly a regular occurence in FoG. 10-10 draws are far more common than actual nothing per side lost games.

I might be able to get hold of the scoresheets from Britcon and see how many games were bloodless. I doubt it will be many.

[philqw78]

Bloodless draws are hardly a regular occurence in FoG. 10-10 draws are far more common than actual nothing per side lost games.

I might be able to get hold of the scoresheets from Britcon and see how many games were bloodless. I doubt it will be many.

But that isn't the point Hammy. Scoring just for killing, not what you keep will force attacking play.

A scoring system of 1 point per 5% of enemy AP destroyed, routed, forced of table would be great. +5 for destoying enemy if you are not destroyed. Giving a possibility of a 25-19 win, 19-25 loss. And Olivier/Pete may well have been a 19-18

But the way people believe people are these may happen too often. I don't suspect they would as others would see the game, but there is the possibility.

[peterrjohnston]

I did some playing with your numbers – as posted vice the raw numbers. I noted that your percentages are percentages within each BG size series. It would be useful to know how many data points there are in each series as the graphs actually look quite variable making it questionable whether or not there are statistically significant differences between the series – with the exception of the 11 minus BG series.

There are certainly less armies in the smaller and larger BG series. I would say this is for obvious reasons. First, at 800AP, the intention of a lot lists seems to be that armies should be around 11-14 BGs (Nik?). And there are many popular lists for which it is hard to get above 12/13 BGs, especially in the medieval time period. 100YW English is one, Dailami another, although see Dave's Santa Hermandad Nueva Skythian on madaxeman.com for an example of how to get to a lot of BGs in late medieval... The second is that now players realise <=11 BGs is a bad idea at 800AP.

However, there are some good armies which can easily make 16+ BGs and function well. Dominate is obviously one, Christian Nubian another, and many of the skirmish steppe armies.

When I did the original analysis on the Italian results, I was expecting something like a normal distribution of scores, but with the added complication of a rise in the 20-25 scores because of the +5. (Something close to a third degree polynomial). The number of results in each range is:

<= 11 82
12 - 13 328
14 - 15 204
16 - 17 44
18+ 50

(I misremembered the total, that makes just over 700 scores; I do have another 80 or so to add in sometime).

I normalised the data for the obvious reason of being able to compare the ranges on the same graph.

Like I said when I did my original graph, I would like to see more results (I just don't have any more to add!), so Hammy's Britcon numbers are interesting. More is always useful. For this reason I didn't start doing anything more sophisticated in the analysis (as well as the fact I don't have a great deal of time to devote to this! ).

There may also be some obvious bias in the data. I can think of at least two. First, it's one UK competition only, and only Italian results, although I did include the Roma European meeting last year. Second, some players prefer certain lists (Graham with Dominate for one), which might skew the results.

Anyway, using the midpoints of the scoring categories and 22.5 for the top category I computed averages and slopes per series.

What do you mean by "slope per series"? If it was a linear regression, that would be wrong, the relationship is not linear, the +5 complicates the relationship. Without this I would probably expect a normal distribution of scores, not a linear one.

[peteratjet]

Hi Julian,

It's a risk to comment on someone else's circumstance but if this is the case it would seem that the problem is you either need to find another group of find a way to shake up this group. You say, in an earlier that they are losing players (i.e., numbers are down) but that they fear tinkering as it might lose them players. Maybe if you invited some more aggressive, good players from somewhere else to participate in a tournament that might open some eyes. Best of luck with trying to invigorate the group.

They've all drifted to FOW and they ain't coming back as things stand.

Having seen Flames of War being played at a couple of shows, the thing that stands out to the casual observer is the quality of the terrain, which ranges from nice to exquisite. In contrast, the "three pieces of felt" minimum standard for ancients games is a bit sad. Apart from the visual appeal, the pre-set terrain poses different challenges to the players. It's clearly more work for the organisers, requiring a lot of preparation and placing more constraints on match-ups as the competition proceeds if they want to avoid "I already played on this table".

[shadowdragon]

quote="peterrjohnston"]There are certainly less armies in the smaller and larger BG series. I would say this is for obvious reasons. First, at 800AP, the intention of a lot lists seems to be that armies should be around 11-14 BGs (Nik?). And there are many popular lists for which it is hard to get above 12/13 BGs, especially in the medieval time period. 100YW English is one, Dailami another, although see Dave's Santa Hermandad Nueva Skythian on madaxeman.com for an example of how to get to a lot of BGs in late medieval... The second is that now players realise <=11 BGs is a bad idea at 800AP.

However, there are some good armies which can easily make 16+ BGs and function well. Dominate is obviously one, Christian Nubian another, and many of the skirmish steppe armies.[/quote]

I was thinking about this too as the choice of BG is conditional on the choice of army (or is it the other way around? ) which complicates things as we now have to separate the influence of army from BG size from the data. I suppose if someone paid you money to do it one could set up controlled experiments with a sampling of armies , same set of players with armies designed with large / small BG's but that'd be no fun unless it was a lot of money.

When I did the original analysis on the Italian results, I was expecting something like a normal distribution of scores, but with the added complication of a rise in the 20-25 scores because of the +5.

I wouldn't expect normal distributions in general since a number of things are conditional (e.g., game designed to suit a particular range of BG, influence of army type on numbers of BG, prior success with a particular army, etc.) and not independent.

I normalised the data for the obvious reason of being able to compare the ranges on the same graph.

That’s understandable, but, as I’m sure you know, that means there’d be less confidence in the results for the bins with lower numbers….add to that the possibility of fewer army choices in those bins.

For this reason I didn't start doing anything more sophisticated in the analysis (as well as the fact I don't have a great deal of time to devote to this! ).

That’s understandable – see “lots of money” statement above. One of things I’d consider doing is a principle component analysis using a variety of factors (e.g., like the suggested fields above). However, I think the bigger issue is player perception and enjoyment. If people are really concerned about this as an issue feedback surveys are an idea…at least in Spain.

What do you mean by "slope per series"? If it was a linear regression, that would be wrong, the relationship is not linear, the +5 complicates the relationship. Without this I would probably expect a normal distribution of scores, not a linear one.

Yes, I did use linear regression but I did not mean to imply the relationship was linear. I only had the graph above to work with and just wanted to get an idea of the trend. Estimating a trend and de-trending data is, for example, a common practice with something like Fourier analyses – but that certainly doesn’t imply linear relationships. And, yes, the +5 should be removed first. I’m not sure if a normal distribution is right. There could be “fat tails” as a result of the factors you mention that skew results or because of changes in time with changing experience, preferences, etc. of the community.

I think it’s a real challenge this case (i.e., disproving a null hypothesis such as “numbers of BG has a significant influence on game scores) as it is complex, which BTW is what makes the game enjoyable – for me anyway.

Still I think it was interesting and useful to see the graphs. However, all I can see that’s reasonably substantiated by the data is that “11 or fewer BGs” is a bad idea.

Cheers

Paul

[hazelbark]

They refuse point blank to deviate from the rules. There is a great respect for orthodoxy and a fear that tinkering with anything might lose them players.

There is I think a common view here, maybe less intense, but one of the fears is that the rules become different at every club.

Orthodoxy is what permits international events without recriminations.

Still it would be helpful to have the "official slitherine alternatives for tournaments" posted somewhere to help people in Julian's position. It would promote diveristy and probably promulgate a new round of interest.

[hazelbark]

Having seen Flames of War being played at a couple of shows, the thing that stands out to the casual observer is the quality of the terrain, which ranges from nice to exquisite. In contrast, the "three pieces of felt" minimum standard for ancients games is a bit sad.

Agreed. We definiately need better terrain. One of the advantages of FoG was less precise measurements than DBx, so in DBx having felt helped ease controversy. We can get better terrain in FoG.

For starters lets add more gentle hills with slope no felt.

[hammy]

Having seen Flames of War being played at a couple of shows, the thing that stands out to the casual observer is the quality of the terrain, which ranges from nice to exquisite. In contrast, the "three pieces of felt" minimum standard for ancients games is a bit sad.

Agreed. We definiately need better terrain. One of the advantages of FoG was less precise measurements than DBx, so in DBx having felt helped ease controversy. We can get better terrain in FoG.

For starters lets add more gentle hills with slope no felt.

Terrain is slowly improving but is still a looooooong way behind the terrain seen on FoW tables. Better terrain is however just a case of players being willing to bring and use it. There were a number of attempts to encourage better terrain by offering prizes but most of the time the terrain at the tournaments where these were offered was no better than normal and they were often not awarded as there were no deserving candidates.

As an example in one version of DBM Phil Barker declared that plain rectangles of felt were not allowed so one player I know just cut a corner off each of their rectangles so they were now irregular pentagons and kept playing......

[kevinj]

Terrain is slowly improving but is still a looooooong way behind the terrain seen on FoW tables.

Isn't it the case that FOW terrain is set by the organisers and not brought by the competitors? There is defiitely a trade off between quality and portability, plus the desire to be able to place a number of different errain options dependent on your opponent.

[jlopez]

Having seen Flames of War being played at a couple of shows, the thing that stands out to the casual observer is the quality of the terrain, which ranges from nice to exquisite. In contrast, the "three pieces of felt" minimum standard for ancients games is a bit sad. Apart from the visual appeal, the pre-set terrain poses different challenges to the players. It's clearly more work for the organisers, requiring a lot of preparation and placing more constraints on match-ups as the competition proceeds if they want to avoid "I already played on this table".

The terrain is actually bringing in a lot of new, young (and I mean 40 or less...) players to the hobby which FOG isn't. The organizers get around the "I already played on this table" issue by having a spare table or two.

[hammy]

Terrain is slowly improving but is still a looooooong way behind the terrain seen on FoW tables.

Isn't it the case that FOW terrain is set by the organisers and not brought by the competitors? There is defiitely a trade off between quality and portability, plus the desire to be able to place a number of different errain options dependent on your opponent.

It varies. At Britcon most of the terrain is brought and set by the organisers. At most other events (Challenge, Roll Call, Doubles for a start) most or all the terrain is brought by the players.

[philqw78]

The terrain is actually bringing in a lot of new, young (and I mean 40 or less...) players to the hobby which FOG isn't. The organizers get around the "I already played on this table" issue by having a spare table or two.

Or good draw software.
But, yes, ancients terrain is, on the whole, awful. I have some nice pieces. But I wouldn't like to lend them out to a scrum of people wanting to set their own terrain by the rules. And I wouldn't want to take them on slEasy Jet. Which leaves only pre-set tables as an option for those willing to give up their nice terrain for the weekend.

[philqw78]

It varies. At Britcon most of the terrain is brought and set by the organisers. At most other events (Challenge, Roll Call, Doubles for a start) most or all the terrain is brought by the players.

But Hammy, those players set up 1 or more tables and it is left set up for the weekend, or only moved by an organiser. People don't get to chop and change so the chance of damage is small.

[shadowdragon]

Thanks to hammy, I've had a very brief look at the data.

First, the hypothesis: I can't "prove" that army BG size has a significant influence on game results. What I can do is to see if the data disproves the null hypothesis, which is that "army BG size has no significant influence on game results".

I tried a few things but ended up dividing the game results from hammy's sheet into 3 sample populations, (A) those with 11 BG or fewer, (B) those with 12-15 BG and (C) those with 16 or more BG. Similar to Peter I then assumed that the distribution of scores for each population would be a normal distribution. The results are:

A: Armies of 11 or fewer BG (90 games):

Game score: Average = 11.5 +/- standard deviation of 7.8
Player ranking: Average = 10.8 +/- 9.4

B: Armies of 12-15 BG (390 games):

Game score: Average = 11.2 +/- standard deviation of 7.5
Player ranking: Average = 16.5 +/- 10.7

C: Armies with 16 or more BG (48 games):

Game score: Average = 11.9 +/- standard deviation of 6.5
Player ranking: Average = 8.6 +/- 5.9

Skewing: Both A and B have 57% of their games less than the average while for C it is 44%.

No analysis was done for the different themes, but it should be noted that the distribution of BG size within each was:

FOG1: 0 X 11- BG / 34 X 12-15 BG / 4 X 16+ BG (total = 38 armies)
FOG2: 5 X 11- BG / 22 X 12-15 BG / 3 X 16+ BG (total = 30 armies)
FOG3: 2 X 11- BG / 5 X 12-15 BG / 1 X 16+ BG (total = 8 armies)
FOG4: 8 X 11- BG / 4 X 12-15 BG / 0 X 16+ BG (total = 12 armies

Observations:

The assumption of a normal distribution for the scores for each population is questionable, which means many typical statistical tests are not applicable – so I didn’t try them. However, variability within each group is quite high while the difference in means was small across the groups.

This was not set up as a controlled experiment but as a tournament. This means that the armies of each BG size were not randomly distributed across the different player groups of FOG1 to FOG 4. Half the large BG armies were in FOG1 and more than half the small BG armies were in FOG4. More interestingly, small BG armies dominated FOG4 with 2/3 of the total number of armies.

Conclusion:

There is no evidence in the data to refute the null hypothesis above that army BG size has no significant influence on game results. Secondly, the data results more likely reflect a relationship between theme and army BG size (e.g., nature of army, player’s perception of advantages, etc.).

Result: The data set is an unreliable witness and is dismissed from the witness stand. No further questions.

[jlopez]

As long as you don't bring out the comfy chair.....

I find the comment that the players have moved to FoW interesting. what scoring system to they use for that? The reason I ask is that the Battlefront scoring system has one version where if neither player wins then they both lose. Perhaps you could use a FoG scoring system where you start with no points and only get points for what you kill?

They used the standard 6-1 system but found that too many players preferred to lose 3-4 by bumping off a couple of enemy platoons than risk a 6-1 defeat by actually going for the objective. As a result we now use the 1-0 system where you get one point for completing the mission irrespective of losses, ties are sorted using the 6-1 system. As a result players are being more aggressive.

I recently suggested something similar for FOG and...wait a minute...there's a crowd outside my house and they seem to be piling faggots.

.

It does sound rather like the issue may be the players. That said I will repeat that for me 800 point open (or 400 point open in DBM) is and was my least favourite type of Ancients wargaming.

That is also the unfortunate conclusion I have reached and am completing my Polish infantry company as a result. The issue with themes is that most players only have one, possibly two armies.

[philqw78]

I recently suggested something similar for FOG and...wait a minute...there's a crowd outside my house and they seem to be piling faggots.

Homophobic as well then? Or is that homothropic if they are piling them??????

[shadowdragon]

I recently suggested something similar for FOG and...wait a minute...there's a crowd outside my house and they seem to be piling faggots.

Homophobic as well then? Or is that homothropic if they are piling them??????

You do like to walk on the wild side, eh, Phil? Or is that just the pachyderm jeans...er, genes...ah, sorry! Forgot about the semi-naked part.

[azrael86]

[
The issue with themes is that most players only have one, possibly two armies.

Really? I find that quite surprising, although the plethora of themes and books undoubtedly forms something of a barrier to new players.

[nikgaukroger]

[
The issue with themes is that most players only have one, possibly two armies.

Really? I find that quite surprising, although the plethora of themes and books undoubtedly forms something of a barrier to new players.

Spain is a low wage economy for Europe - toy soldiers cost relatively more there than, say, GB or the US.