Pages

Tuesday, March 5, 2024

Location, Gaming, and the Magic Quadrant

The last time we looked at the Wargames, Soldiers, and Strategy's Great Wargaming Survey, the topic examined the possibility that survey respondents' location is an influencer in driving gaming preferences.  That analysis focused on a series of pairwise comparisons of location against table size, game venue, group size, gaming frequency, game length and favorite wargaming period.  Almost more than a handful of attributes.  Each attribute had multiple levels (treatments) too.  Only those survey respondents who classified themselves as primarily historical wargamers were included.  As a refresher, these pairwise comparison analyses are found at Does Location Influence How We Game?. The survey results demonstrated that location not only plays a role in how we game but also what we play.  

The blog post summary suggested that a further "next step" in the analysis would be to combine these pairwise comparisons into a multi-dimensional, holistic examination of group association.  Today's analysis is an extension of the pairwise comparisons seen earlier and addresses the multi-dimensional question.

Multiple Correspondence Analysis (MCA)
Multiple Correspondence Analysis (MCA), briefly, is a statistical technique from which any underlying structure in the categorical response survey data may be detected. MCA is an extension of Simple Correspondence Analysis.  The results are then presented in graphical form in two-dimensional, Euclidean space.  Interpretations are understandable and intuitive without knowing the underlying technique.  Interpretation, though, can be as much art as science.  Interpretations are only tendencies and associations based upon the survey data.

Data Reduction
Plotting the analysis outcomes over seven attributes each having multiple levels makes for a messy plot.  In order to make the resulting plot more manageable and (hopefully) more intelligible, data reduction techniques are introduced.  The main technique utilized is simply re-binning levels of each attribute to reduce the number of levels under consideration.

After re-binning, the levels for each of the attributes are:
  • Location: UK (and Ireland), USA (and Canada), Europe (Continental Europe and Scandinavia), Australasia (Australia and New Zealand).
  • Wargaming Periods:  Only the top periods remain.  They are: WWII, Napoleonics, Ancients, Medievals, 18th Century, ACW.
  • Game Frequency: Weekly, Monthly, Less Than Monthly.
  • Game Length: 1-2 hours, 2-3 hours, 3+ hours.
  • Game Type: Scenario, Campaign, Pick-up.
  • Game Venue: Solo, At home, At club, At game store, Friend's home.
  • Table Size: Small, Average, Large.  
Now consider the interactions between the seven wargaming attributes.  When analyzed using MCA, the seven categorical attributes with their accompanying levels are plotted in two-dimensional space as illustrated in the graphic below.
Some of the levels are clustering near others while some levels are seemingly plotted as outliers.  Like correspondence analysis, attributes plotted along the origins exhibit little differentiation.  Those levels farther from the origin demonstrate more separation and distinction.  At this stage, inferences can be difficult to make.  Adding in the origin reference lines help to accentuate these differences.  Reference lines also allow distinctions between each of the four quadrants.
Magic Quadrant
Once the plot is divided into quadrants, each quadrant is examined to discover any associations between the levels of the various attributes.  For ease of visual identification, each quadrant is color-coded.
Does each quadrant present a specific collection of levels and attributes?  Are levels of an attribute unique to a particular quadrant?  Is there overlap?  Can any useful inferences be gleaned from the MCA?

Since gamer location was the motivation for these studies, let's examine Location.  The Magic Quadrant shows that each location level appears in one quadrant only.  There is no overlap of two in the same quadrant.  That result suggests that location does influence how gamers game.  Do other general tendencies surface from the analysis?  Are there tendencies peculiar to a specific region?  Summarizing by quadrant produces the following tendencies and associations:
  • UK (Orange Quad): Tend to meet weekly at a club for two to three hours per session.  Clubs tend to have ten or more gamers.  Perhaps a slight tendency to game 18th Century and WWII more than other regions.
  • Australasia (Yellow Quad): Tend to game monthly at either a friend's house or at a game store.  Games tend to be played on large tables with between five and nine players.  Game sessions last three or more hours.  May feature Napoleonics more than other regions.
  • USA (Green Quad):  Tend to game less often than either UK or Australasia, at home, and with one to four players.  Tend to game ACW more often than other regions.
  • Europe (Blue Quad):  Tend to game solo, on small tables, with games only one to two hours in duration.  They tend to play Ancients and Medievals more than other regions.     

  • Interesting results.  MCA can provide a data-driven dimension to uncovering tendencies of gamers across the globe.  Does location influence what is played and under which conditions?  The survey results suggest, that yes, it does.

    Of course, these are broad generalizations of the tendencies provided by respondents to the survey.  We are individuals.  Still, does this present a picture of the wargaming hobby that is familiar to you?

65 comments:

  1. Interesting analysis Jonathan. I think in the UK, homes tend not to be large which drives the club scene and gamers tend to live closer together which makes frequency of gaming much easier. About 90% of my own club live within 5 miles or so if the club. Less than 30 minutes travel time.

    ReplyDelete
  2. I foind it interesting that there is so much gaming at game stores in Australasia, but comparatively less in the USA. I would have thought this would be reversed, primarily because outside of Warhammer shops there just don't appear to be that many game stores that I have noticed either in Brisbane or elsewhere, and yet the analysis has been performed on mainly historical gamers.

    ReplyDelete
    Replies
    1. Glad you find the results interesting. From my perspective (and region), there are not that many game stores period, whether with or without gaming. If there is in-store gaming, it typically caters to non-historicals. Keep in mind that the results are relative to the other regions. Perhaps the other regions see even less in-store gaming?

      Delete
  3. Interesting topic, I am glad of your summary as I got a bit lost amongst the charts.

    I can only speak for the UK bit of the data. It doesn’t particularly represent me, but I think it is a fair description of the UK. We are a relatively small island with a high density of what can be described as clubs and some of these are very active and well organised, because are largely responsible for our vibrant show scene.

    I would have thought that there are just as many gamers meeting with a friend at home or playing solo, but they don’t show up in your ‘trends’ as being equal, so perhaps not.

    It would be interesting for there to be a poll at the entrance to one of our decent sized shows that asked one question - are you a club member? The proportion in that sample of footfall would be interesting, though of course the location still carries the vagaries of population density and transport links etc, though when I attend a show, I am often travelling in the 80 - 160 mile radius and so venue does not correlate to factors falling out from my home area.

    ReplyDelete
    Replies
    1. Sorry for losing you early on, Norm. I see from responses below that you are not alone.

      I tried to lay out some of the groundwork without diving into the theory and mathematics. That discussion would have lost you for sure.

      On the "at Home" v "solo" presentation of game venue, the placement in each of the quadrants is relative to the tendencies in other groupings. What we can conclude from the graphic is that Solo/@Home gaming are more similar than going away for a game.

      With so many wargames show in the UK, I wonder how respondents classify gaming at those venues? At a Club?

      Delete
    2. I'm like Norm, I got lost in the lead up but could understand the summaries.

      Delete
  4. The UK result seems to reflect the gaming scene in Bristol, with I think 3-4 clubs within the greater Bristol area. I would have thought Napoleonics would have been ahead of the 18thC, but happy to see the latter getting into the mix though.

    I and my friends generally buck the trend, by being solo gamers or at friends houses, due to geographical location and circumstances. These days I do wonder whether I would enjoy going to a club, as I'm very happy with my current set up.

    Thankfully your colours coded chart at the end made sense of everything as to be honest you lost me at the start!

    ReplyDelete
    Replies
    1. Thanks, Steve! The placement of levels is relative. Notice that '18th Century' is near the origin. That suggests little distinction but the UK tends to see slightly more gaming in this period than other regions.

      Sorry to have lost you early on but happy to see that you picked it up later.

      Delete
  5. I think the results are potentially influenced by those who responded.
    The UK has an active club scene; my experience is it tends to focus on the main / most frequent attendees interests with the odd venture into new fads / popular game systems. Again, in my experience, Nappys, Ancients and WW2 predominate; I'm surprised to see C18th as for years this has been a minority period.
    Increasingly, as an alternative to clubs, there are informal groups who meet at each others homes or sheds or occasionally hobby stores; the latter are often tied to a system (eg Billhooks at Boards & Swords Derby).
    Parallel are the fantasy / Sci-Fi gamers - some clubs and a lot in stores.
    Clubs often have distinct "cliques" if large enough; often tied to a period / rule system where only some members want to play.
    Frequency; a lot are weekly, otherwise monthly or randomly infrequent.
    Good one for the VWC as we have UK, US and people with experience of Aus?
    Neil

    ReplyDelete
    Replies
    1. Neil, as in surveys of this type (non-random, self-selection, etc.), results are influenced by who actually responds. Still, I find the results interesting and perhaps even useful to gain a (weak) pulse of wargaming. Given those caveats, the results remain consistent year in and year out. With a turn-over of roughly 50% new respondents, perhaps these results present a kernel of truth within?

      Delete
  6. As with other respondents so far Jon, I would say your tendencies for Australasia correspond quite well with my gaming experience in New Zealand over the last 30+ years.
    I have played with the same basic group the whole time - we started off at a wargame "store" that was run by Mark (1866) and when he left the role and the business morphed into FoW, I moved on to playing at friends' houses.
    Although I don't attend as often as I would like, the majority of our group meet fortnightly at Barrys place and play an all-day Sunday game 10-1600 roughly on a 6' x 18' table.
    Do we play more Napoleonic than any other period - debatable - the group (not me so much) might actually play more ACW than any other single period (that will please Stew if he reads it!)
    There are clubs here too, of course (we are an ex British Dominion, after all!) but none of our group have been involved with an organized club for many years, probably because we are lucky enough not to have to rely on scout or church halls for a large enough gaming venue.

    ReplyDelete
    Replies
    1. Thanks for your perspective and observations from New Zealand, Keith! Doubly good to see that the results reflect (to some extent) your own experiences. Interesting that you play ACW more than other periods especially when I see so many other periods out on your table including a lot of Napoleonics games.

      Delete
    2. Stew is pleased. 😁
      (i always read Keith's (and several others) comments on different blogs.

      Delete
    3. Stew, your reputation follows you. We now ALL know which CW is the Best CW and some can even hum a few bars.

      Delete
  7. Another interesting analysis. I was surprised by game store result - I'd always associated it (obviously wrongly) with US practice. But then, if I think on it, I read various US-based blogs and they normally game at home, or virtually, so that fits.
    It's a long time since I was in a club but my recollection fits the sort of scale of turnout. However, although we did have club big game days occasionally, a regular meeting usually featured clusters of people playing one-to-one, or in small groups (4-6). This may have changed, but I don't think so.
    Another blurring factor would be that people form gaming groups that are not formal clubs and tend to meet in homes, perhaps of a member who has more space for a permanent set-up, like a dedicated room or shed, rather than a public venue, like a hall or function room.

    ReplyDelete
    Replies
    1. Thank you, Anthony! Out in the Pacific Northwest, I see no in-store historical gaming. Then again, I have my hands full gaming via other means.

      While you may game either at home or at a friend's, remember these associations are relative. The US/Canada and Australia/New Zealand appear to game at these venues more often than the UK by comparison.

      Delete
  8. Thank you for showing us the data in such an organized manner. I admit to being lost until the scaffolding of quads & colors was added. I would agree most of the US players have been monthly visitors at a host or a store Store gaming is newly added and several stores have tried to capture a winning combination. That ACW is more popular here than elsewhere...No accounting for taste.
    I won't bore everyone with how things have evolved here, but it is much h easier to game than just 40 years ago. Not to mention 50 years ago.

    ReplyDelete
    Replies
    1. You are welcome, Joe! Again, happy to see you found your way back to understanding the results after leading you astray for a bit.

      Delete
  9. The results make sense. As you point out, the more the result diverge from the center, the more significant the associations are.

    ReplyDelete
  10. Statistically not very significant but the Hong Kong scene has some interesting aspects. Like Neil's comment on the UK scene, all members live fairly close to each other, given the very small size of the place and the good transport links. Full club meetings are not frequent, though, because finding affordable venues of sufficient size is very difficult and we will be losing our only reliable venue in June (we are discussing a huge game of Waterloo to finish up with!). It looks as if we will be shifting to holding several meetings in smaller venues or members' homes, trying to co-ordinate via social media.

    ReplyDelete
    Replies
    1. Thank you for the report from Hong Kong! You are correct in that this region does not see a significant number of survey responses. Did you complete the survey in 2023?

      Delete
  11. hmm, I'm trying to sort out "well we used to do X 40 years ago" from what I'm aware of what little I know now of local gaming (say 200 km) as well as keeping mind of various diverse practices I'm aware of now.

    The common NA habit of small groups meeting at someone's house is probably the most common though rare for me these days for personal reasons. I also know some informal clubs that game days in public places, usual as a part of a wider games venue (eg a quarterly board game event at a community hall or similar where miniature wargamers make up about 1% of the attendees).

    The other thing that seems to be most common in NA is medium to large conventions where volunteer "game masters" people run multi-player miniature wargames for others to play (along with vendors to give people somewhere to spend money while they're in the mood like at most any show or event ). A LOT of gamers attend these events, some making drives of hours or days (no name/no packdrill etc ) and in some regions, they various venues are close enough that some people will go to 2,3...5....10+? shows a year.

    Lastly, while SciFi & Fantasy shops for a certain retailer have been hosting in store game playing facilities (good for sales). That marketing scheme has now hit WWII, esp for Bolt Action from what I've heard.



    ReplyDelete
    Replies
    1. Excellent summation, Ross! Like you, I am likely out of the loop with what the local trends may be. I found a niche in among a handful of local gamers and games are always at someone's house. I have not been to a con in about 20 years.

      Delete
  12. I feel like the USA is somewhat lagging behind the other regions. though I fit the model, game less often, but when do game do it for longer (less than 4 hours is waste!) and mostly ACW!
    😁

    ReplyDelete
    Replies
    1. Why do you suppose the US is a laggard? Population dispersion, independent spirit, no time/no money, anti-social, or something else?

      Delete
    2. Couldn't say with any authority; My money would be population dispersion. miniature wargaming is not a particularly mainstream hobby so us gamers are spread all over the place and not close enough to play regularly / weekly like other places.
      I also think that's part of the reason why US conventions are bigger deals, with being a whole weekend and lots of participation games; because people travel from far away and want to get in their gaming fix.

      Delete
    3. Thanks, Stew! No one says we must speak from a position of authority. Wargamers speculate on almost everything. Seeing the UK wargame show scene from afar, it is hard to believe that a show can run from only 10am to 4pm for one day only. Even then, gamers are heading out before closing time.

      Delete
  13. Very interesting analysis Jonathan. Like many others I am a bit lost in the underlying reasoning ( I like the sentence - "Interpretations are understandable and intuitive without knowing the underlying technique. " - that is me 😁)
    The comments suggest that the results do seem to ring true for the various countries.
    Thanks for performing these analyses.

    ReplyDelete
    Replies
    1. Thanks for your comment and for trying to understand the analysis. Meaningful wargaming inferences must be coaxed out of the survey data, sometimes with effort. The MCA allows one to see these associations graphically that where otherwise they remain hidden.

      Delete
  14. I think that makes me an Anglo-European-American, but it all sounds plausible, Jon. Seems to tie in with the reports one sees online. Wondrous what crunching the numbers can do ✔️✔️✔️

    ReplyDelete
    Replies
    1. Martin, you wear a coat of many colors. It is a wonder what interesting inferences can be teased out of the data.

      Delete
  15. Hi Jon. Something that interests me is the type of games being played, which is one of your "re-bins". I can't seem to find it on the diagrams though. Does it show variations by region?

    ReplyDelete
    Replies
    1. Good eye, Anthony! Yes, the attribute is missing from the graphic. In an earlier iteration, Game Type was present but the impact of inclusion was not significant in a statistical sense. Therefore, the variable was jettisoned. All three levels plotted near the axes suggesting little differentiation among regions.

      Delete
  16. Well I for one totally understood the preamble but got lost in the simplified graphs and clear explanation of results. ;-) FWIW you nailed the European historical gaming experience. There isn’t one unless you’re in a major city like Paris.

    ReplyDelete
    Replies
    1. Oh I meant to ask…is remote gaming prevalent enough to warrant a separate analysis?

      Delete
    2. Great to see that the preamble did not lose you! Remote gaming counts were low so I did not include remote games as a separate attribute. A shame, really. I think remote gaming quite popular.

      Delete
  17. Always interesting to see how analytics break things down. As you say, these are understood to be very broad generalizations, but I'm guessing that even though "correlation is not causation" there is some validity underlying the trends.

    ReplyDelete
    Replies
    1. Ed, with consistent results year after year, in my mind at least, there seems to be some validity to these underlying tendencies.

      Delete
  18. I think the Rejects buck the trend for the UK. We game normally monthly, not including remote games, play at home, well in Posties shed anyway. We play for 6 to 8 hours a game, we game with between 4 to 7 of us at a time and most of us live within a 20 minute a drive of the shed.

    ReplyDelete
  19. Good topic! The results reflect my experience in the Northwest USA. For 25 years in Alaska I only found one historical gamer and we infrequently played ancients exclusively. All other gamers I knew of played fantasy. There were no clubs other than D&D. For the last 26 years in eastern Washington, any gaming has been sparse to nonexistent. I attended only one organized convention (no longer held) and have connected with only one other historical gamer. The only clubs I know of are over on the coast, a long way off. The game stores I know of do only Warhammer 40k.
    The UK has more gamers per capita than any other region. Every gaming friend I have made there belongs to a club.
    So the results make a lot of sense!

    ReplyDelete
    Replies
    1. Thanks for your feedback, Vol! Compared to the UK, we are much more isolated. I have not been to a wargaming con in almost 20 years and that was on the other side of the state, 5-6 hours away.

      Delete
  20. I have no idea how you do the analysis, or what the numbers mean on the dimensions. I dropped Maths at 16. Years of age that is, not counting. But I do follow the explanation of what it all signifies. It makes sense from a UK historical wargamer's point of view.
    Looking at the original post on the subject, we Brits seem to have a lot of solo gamers, as well as folk who game in the largest groups. The former is counter-intuitive given population density and relative number of gamers - but then I'm one of those cussed b's who does game solo despite being in the largest metropolis in W.Europe!

    Another thing that springs to mind about the UK is that it has a massive club culture. Not just wargaming; everything. All manner of hobbies, and interests and activities lead or led to the formation of clubs and societies. Sports, politics, gardening, crafts, gentlemen's, model railways, stamps, youth clubs, working men's, book, real ale, rambling, youth organisations (Boys Brigade etc), guns, and even sports club supporters clubs. Clubs with their own clubs! There can't be anyone who hasn't been a member of at least one club. We love to join things, and even better, break-out and set-up our own clubs! I don't know if this a peculiarly British thing.

    ReplyDelete
    Replies
    1. Good to see that the explanation at least makes sense! From the graphic, the UK seems to dominate gaming at clubs with large numbers of players. Very interesting that "club culture", likewise, is robust in the UK. I don't see that same level of club culture in my region but, then again, perhaps I have never been invited or have no mainstream hobbies. Even getting fellas together for a F2F game can be challenging.

      Delete
  21. Nice analysis Jonathan. I am surprised that there is a high score for playing at games stores given the shortage of games stores in NZ. I would have thought clubs would have been the stronger element..

    ReplyDelete
    Replies
    1. Thanks! Keep in mind that the analysis is based upon relative associations. While clubs may be strong in NZ and Australia, the UK sees a stronger association based upon survey responses.

      Delete
  22. I have seen on occasion the suggestion that half the world's wargamers are in the US, yet you seem to have great difficulty finding one another : ) . But then, I am also aware that Brits always underestimate how far apart things are in the US compared with here. I wonder if you get more club and gamestore play in the big cities and more friends' homes and solo in the country/wilds?

    ReplyDelete
    Replies
    1. Based upon WSS survey responses, the UK and USA maintain nearly the same number of response counts year after year.

      You may right in that we (in the US) simply cannot find one another. As an example population dispersion, my state of Washington is 40% larger in area than all of the UK. Total population is not 8 million. In the US, 40% of the population lives along the coasts. It is a big country and I have not throw Canada into the mix which is included in the USA/Canada bin.

      Delete
    2. As you know, I am a bit prone to pursue intriguing idea rabbits down holes, so I did a quick skim of the internet. One popular proxy for popularity of wargaming in different zones is TMP, because TMP do a detailed breakdown of where members are located.
      http://theminiaturespage.com/statistics/members/?type=By%20Location
      They have about 39,000 trackable members, of whom approx 15,000 are US based. UK is second highest, with about 8,200. Now, big caveats on this (not least the bias toward English speakers) but gives some idea of relative distribution.
      Given your mathematical background, you are probably aware of this Finnish study from 2019
      https://journals.sagepub.com/doi/full/10.1177/1046878120929052
      Which suggests there may be as many as 0.5% of Finlands population playing wargames. Now, I don't think the data is anywhere near able to support that, but it does suggest that there may be hidden wargaming communities of some size lurking out there.

      Delete
    3. Thanks, Anthony! I recall hearing of this study but do not recall ever actually reading it. I never knew "miniaturing" was a verb. the link to the study is much appreciated.

      I would not count much on TMP as a valid gauge of size of the hobby. First, member counts are from the beginning of TMP time with obsolete (or dormant) accounts rarely (if ever) getting removed. No way to discern active accounts from stats alone. Second, users can have multiple accounts for many different reasons.

      As for lurking wargamers, we all know that that situation is very likely. These lurkers may actually constitute a significant subset of the population.

      Thanks again for going above and beyond to further this discussion, Anthony!

      Delete
    4. As I said, big issues with TMP data. But, looking at proportions many more US than UK members. WSS survey similar numbers of responses from both. WSS awareness v TMP awareness might be a factor? Whatever, there seems no way of really sound way of estimating numbers of wargamers out there, sadly.

      Delete
    5. Anthony, I get the question of "How many historical wargamers are there?" more often than you can guess. Based upon anecdotal evidence, WSS' survey, and limited industry data, I have a guess on this number but it is no more than a guess. TMP leans heavily toward a US bias, I think, which may contribute to the 2:1 ratio of US to UK members.

      Delete
    6. I wouldn't count on TMP UK numbers being an accurat picture. The Administrator is a known Anglophobe and has 86'd hundreds, if not thousands of UK gamers from the site. I got caught up in one of his purges several years ago, along with many others, attempting to defend my UK fellow gamers from just one of his anti-British opinion polls.

      Delete
    7. I was a product of a similar purge years ago dealing with copyright and IP.

      Delete
  23. I find the 18th century element the most surprising, otherwise broadly right, certainly now I'm a member of a local club, although I do of course continue to play in the shed. Fascinating post as always!
    Best Iain caveadsum1471

    ReplyDelete
    Replies
    1. Thanks, Iain! The 18th Century shows a big drop from the more "popular" periods but I included it anyway since I know many favor this era.

      Delete
  24. Fascinating analysis. Not sure I can add anything to the superb contributions. Though I must echo Ray's comment about the Rejects.

    ReplyDelete
    Replies
    1. Happy to see that the analysis is of interest. As always, reader commentary is often just as interesting.

      Delete
  25. Your analysis of these data is getting better and better Jon. I was amazed how well the clusters 'fell out'. So often such multivariate analyses add confusion rather than clarity. Brilliant job on the stats and the interpretation.
    I tend to find myself in some 5–10% grouping when it comes to trends or social things, but I fall squarely in this summary of wargamers in Oz. Still, wargamers ain't trendy, hey?!!
    Regards, James

    ReplyDelete
    Replies
    1. Thanks, James! The clusters worked out nicely, for sure. You are always trendy and in style. I appreciate your support for these analyses.

      Delete