Let me preface this that I am one of the people here that wants MP in Elemental games. While I ultimately approved FE not having MP, it was with considerable consternation. If you guys knew what % of the WOM budget was put into MP you'd be shocked (>30%, mainly because of the dynamic, worldwide virtual machine server infrastructure that was developed -- every game is a "pit boss"). So I can assure you, it is literally impossible to be more disappointed than I am.
Now, that said, someone asked for some numbers. Here are some facts that help provide some context to the discussion.
77% of the Demigod user base after its first year had never attempted to play a single multiplayer game. And that was a game designed as a multiplayer centric game. Chris Taylor and I double checked that number repeatedly because we were certain it had to be wrong. It wasn't.
Over 90% of the Sins of a Solar Empire user base at the time we were about to release Trinity had never attempted to play a single multiplayer game.
At GDC, friends of my from Firaxis told me that only around 4% of the Civilization IV user base had ever played a single MP game (not finished a game, simply tried it).
Fewer than 1% of the WOM user base attempted multiplayer. I don't mean played a game. I mean simply went to the MP lobby (it adds an active flag to your SD account if you simply went to the main MP screen).
Now, I want to emphasize that we have no problem developing features that only a fraction of the market is going to use. Heck, we used to make OS/2 software. But at the end of the day, decisions have to be made and sometimes your options are all non-ideal.
But as my first paragraph should hopefully make clear, it is not something we're giving up on.
Can we agree that quoting numbers from two of your own titles and one other 'that a friend told me' is a million miles from an exhaustive canvassing of the TBS playing demographic?
Can we also agree that without the methods of collecting being cited, the numbers are of very limited value? I could say that 98% of players want MP based on my own data collection but without detailing how I collected it that figure is worthless (and absurd). I'm not accusing you of that, of course, I'm sure your indicators are fairly sensible but they can't really be evaluated otherwise.
Granted, clocking users as they come into the lobby is a seemingly viable way to measure such a thing but I have little faith in polling and the like. Although even then, bear in mind that myself and my circle of friends are die-hard multiplayers and the reason that some or all of us never entered the lobby with WoM is because we already know that the multiplayer just wasn't there. Thus the figure is skewed and unreliable. As impressive as your 1% sounds I think it says a lot more about the state of multiplayer in WoM than it does about demand.
These concepts must be considered abstractly, too. It's fine to say, historically only a quarter of people played Demigod but then Demigod was a title of very varied reception. Many people don't go beyond a review score. Why to assume that the multiplayer crowd were actually drawn enough to your title? Multiplayers are loyal beasts and rarely jump ship for anything less than the best. An MP-centric title with rave reviews has no trouble selling. You present the Demi-god figure as evidence of the fractional demand of multiplayer but given that it's a multiplayer centric game could it not be presented equally as evidence for falling at it's central objective?
Also, most multiplayers will learn a game in singeplayer before they bring their game to the lobby. Is there anything to disprove that 60% of users didn't buy Demigod with the intent of playing multiplayer, but didn't like it enough to continue beyond the 'campaign'? Only 23% of people entering the lobby does not actually prove conclusively that only 23% of people were interested in multiplayer. Would you not agree?
Furtermore, who is likely to be playing Demigod today? The 77% who have rinsed single player or the multiplayer community? How do you attribute the value of demand? If 8/10 people play a game alone for 6 months and then shelf it but the remaining 2/10 play it for several years should the value be assessed on a 1 for 1 basis? Many older games still exist in the public mind on the weight of their multiplayer community alone. Different genre I know, but where would SC1 be now with no MP component? The lesson is valid.
Demigod is also three years old. Do you not see a gaming world that hinges exponentially around multiplayer? Tell me if I'm insane because that's what I see. Also, even for the sake of argument we allow for the fact that only one quarter (still a fairly large 'fraction') are historically interested in MP in a 4x TBS that statement completely ignores a dynamic and changing world. Who's to say the people buying 4x TBS games tomorrow entertain exactly the same breakdown? In fact, all reason suggests the trend only has one way to go, does it not?
Soundbites are all well and good but in reality they barely scratch the surface. That 77% means very little to me when it fails so spectacularly to reconcile with the world I see all around me every single day.
At the end of the day, to be clinical about it, you're here to shift boxes and I appreciate that. There are also practical limitations to just how exhaustively such an issue can be explored and I understand that too. But shifting boxes is well and good but the games that are years, even decades old but still have strong communities and lasting legacies and whose originating studios enjoy prolific status usually share two characteristics: A moddable aspect and a multiplayer aspect.
To say one final thing. None of these figures seek to address the fact that the box said multiplayer and what was in it, didn't have multiplayer (in any functional state). A decision for no MP may disappoint. Broken promises enflame.