[ketzo]

Oh man this is an awesome talk, thanks for linking!

Quick excerpt:

> The player said “I lost a 2:1 battle, and I get that, I know I should lose that sometimes.”

> I said, okay, so what’s the problem?

> He said “well, I lost a 20:10 battle — what’s up with that?! 20 is so much more than 10!”

[drc500free]

It seems reasonable to assume that's it's instead something like 10 simultaneous 2:1 battles, and you need to win a majority of them. That's very different odds than a single 2:1 battle.

He seems stuck on interpreting feedback through the lens of a linear-odds, one-shot model. The player feedback is that it shouldn't be linear, and there should be less randomness for larger numbers.

That all makes sense to me... and I suspect makes sense to him when he's not giving a talk for comedic effect.

[tedunangst]

What do you do when it's 15:7?

[Dylan16807]

Then you probably want to calculate the attacks between two big blobs of units without separating them. But the result is still going to be "something like 7 simultaneous 2:1 battles", and not at all like 30% odds of upset.

[igiveup]

Is that wrong? I don't know the logic of the game, but a 20:10 battle can very well have different odds than a 2:1 one.

[kemayo]

They're talking about ratios of unit-strength, which correlate directly to odds of success in a battle, not to numbers-of-units or anything more fuzzy like that. Given which, 20:10 and 2:1 have identical odds of success.

[waterhouse]

Do they, though? It depends on how the combat works.

Suppose, for example, that each side has 5 hit points, and repeatedly you roll a 2:1 die to decide who gets 1 point of damage, until one side reaches 0 hit points. The chance of an "upset", where the weaker side wins, is not 1/3; I compute it to be roughly 14%. If both sides start with 10 hit points, I compute the chance of an upset to be 6.5%. The law of large numbers means that, the more die rolls the combat involves, the less likely an upset is.

Or. Suppose that, at each step, one side has N soldiers and the other has M, and repeatedly a random soldier gets a kill; so that's an N/(M+N) chance that the first side gets a kill, and M/(M+N) that it's the other side. This would make advantages compound within the battle. Then I compute that a 2:1 initial matchup has a 5/6 chance (83%) of victory, and a 10:5 matchup has a 98.8% chance of victory.

(edit) I guess you could say I'm challenging the idea that "unit strength", such that when strength A fights strength B it's decided in one step with probability A/(A+B), makes sense as an abstract concept.

  (defmemo meh (a b p)
    (if (is b 0)
        1
        (is a 0)
        0
        (+ (* p (meh a dec.b p))
           (* (- 1 p) (meh dec.a b p)))))

  (defmemo nub (a b)
    (if (is b 0)
        1
        (is a 0)
        0
        (+ (* (/ a (+ a b))
              (nub a dec.b))
        (* (/ b (+ a b))
           (nub dec.a b)))))

[kemayo]

Sure, you could design a system where it's more complicated, but Sid Meier didn't. In the video he's talking about Civilization Revolutions, in which combat is just "attacker's Attack stat vs defender's Defense stat" to form a probability-of-success, which is then rolled to see who won the battle. There's no hit points or anything like that, just those stats.

More than "players don't understand math", this might be a UI or tutorialization issue. I.e. presumably it was unintuitive because people imagined more complicated ways it might be working behind the scenes, causing large absolute stat-disparities to feel like they should work differently despite being in similar ratios. It's a case where showing an explicit odds-of-success display might have helped, though XCom famously showed how that can backfire...

(Revolutions was a deliberate simplification of the Civ formula, so they could try to appeal to console / mobile gamers rather than the traditional hardcore PC audience.)

[feoren]

> presumably it was unintuitive because people imagined more complicated ways it might be working behind the scenes

I think this is exactly it. And then Sid Meyer calls his players stupid and irrational for assuming the game had more depth than it actually had. For assuming a celebrated game designer would put even a modicum of thought into making a combat system that was balanced, made sense, and felt good.

It's like selling a gallium spoon and then calling people stupid when it melted in their soups. Sure, if you know a lot about gallium, you wouldn't be so stupid and irrational as to put it in your hot soup. But it's a metal spoon that you bought from a reputable vendor. Spoons go in soup. They were being completely rational; it's just that they were tricked into assuming a product was less crappy than it actually was.

[pxc]

> I think this is exactly it. And then Sid Meyer calls his players stupid and irrational

I think this is overstating what Sid Meier says in the talk. His original goal was to make his simple combat stat system clear to users by describing its odds as odds conventionally are described.

> For assuming a celebrated game designer would put even a modicum of thought into making a combat system that was balanced, made sense, and felt good.

That's exactly what he did, through player testing! Through practice and player feedback seems to me like a perfectly reasonable way to uncover an intuitive notion of unit strength. It's not like he said 'they're odds, stupid! learn how to understand odds.'. He recognized that player intuition and fun was the real goal, and his team gradually made the combat system more sophisticated.