Random events on or off?

[logotet]

Thanks for the reply. I've enjoyed your "let's play" greatly. As stated earlier, my free time is usually greatly limited, so if my participation in this thread suddenly ends, it is not because I'm rude or have lost interest.

Bills beat the Patriots or that the Ravens beat the Jets

Being European, these references mean nothing to me, regretfully.

Bogus. Not only do sports have extremely minimal elements of chance, but that chance is up to decisions made by the player. Also, "skill" is not fixed.

In my experience, results of sporting events are very often basically a dice roll. Otherwise the bookies would run out of business, no?

Aside from the logical flaw in something being impossible in theory and then minimally possible in practice, I see nothing wrong here so far.

Take a competition that has absolutely 0 luck factor: shooting 3 point shots indoors with identical lighting and no wind.

There can be no such thing as 0 luck factor. Consider a broader definition of "luck" here. If there was a 0 luck factor the shoot off will always end the same. If it doesn't, then there must be something that affects the outcome. Maybe one guy had one beer too many the night before, maybe he got a hickup at a crucial time. There can be a number of reasons, one cannot account for all possibilities. But the poor player will almost always lose, even if a number of things go his way. But just because it is possible that a fluke may happen, it doesn't mean it will happen. Probability factor may be so low, so both players die of old age before that, or perhaps win the lottery and find something better to do. This is what I meant when I wrote "theoretically impossible and practically almost so". I see no logical flaw here.

How large an impact does this issue have on the jumps?

Difference between +1,5 & -1,5 wind velocity would probably affect the result with at least a foot, effect is huge. Has to do with air friction during the run up phase and even more so during the jump phase. I presume everyone has biked with and against the wind. The difference is monumental.

Basically this point is true but has 0 relevance to the merits of events

I had the impression that you object to events in competitions because they may create an unfair advantage. My point was that being lucky is not an unfair advantage. It is not unfair. It is just luck.

By the way, uniques and religions are not truly chance-based unto themselves and constitute bad examples.

I play random civs and from that point of view which civ and uniques I end up with is a dice roll. Religion spread may affect your strategy for the rest of the game, and one can very rarely controll wich one spreads to one first. Sure, it's not completely random, but very close to it.

Such a shame that we've already demonstrated times where adding an additional chance element such as events DOES make a difference. Believing otherwise is one's choice, but it's proven wrong.

What is "additional"? Where do you draw the line? Monty, Nappy, Shaka and you on a small continent, does that make a difference? Four spies in a raw failing and as a result city defences stay - is that an additional chance element? Is 3 spies failing additional? My point is that there are so many other things that can go wrong, that one more has no significant effect. Ok, you got the Bermuda triangle. But maybe on turn 35 you decided to go for "copper city" instead of "horse city". How can one possibly be sure that the bermuda triangle event would still have happened, had one gone with a different option earlier on? One may argue that in fact it was the decission to go for "copper city" early on that was the "game breaker". Which of course is cowcrap. No one can ever demonstrate that a certain event has a specific impact, because the event does not stand alone, but is a result of every action in the game thus far.

starcraft 2 matches. Or matches between organized teams in Gears of War 3

Again. Regretfully, these titles mean nothing to me (that time issue again).

Try bridge. You'll never look back. Beats any strategy game I have tried. Even chess, which I used to play competitively. Ironic, isnt't it? I think random events should be on and at the same time reccomend a game where chance is intentionally drawn to a minimum.

 

[TheMeInTeam]

In my experience, results of sporting events are very often basically a dice roll. Otherwise the bookies would run out of business, no?

Bookies will very frequently adjust the "line" of the game to what they believe makes it a coin flip. For example, if a team is highly favored they'll make it so that someone has to bet that team wins by, say, 2 goals in soccer or 10 points in basketball instead of simply winning. Where they set the line is where they expect the game to become a coin flip, but they adjust the line based on how people bet too. Basically sports bets come down to what people believe will happen vs what actually happens, with those who guess correctly winning money and those guessing incorrectly losing money.

Would a 0 random-outcome sport put bookies out of a job? @#%$ no . They'd still find a way to allow betting, I promise.

There can be no such thing as 0 luck factor. Consider a broader definition of "luck" here. If there was a 0 luck factor the shoot off will always end the same.

No, this is where you are wrong. You are using a definition of luck that is not an actual definition of luck. Sometimes, people simply make mistakes of their own volition or are forced by opponents to guess.

Maybe one guy had one beer too many the night before

How is this luck? Did the RNG COMPEL him to drink beer the previous night? I don't think so . Player choices that don't allow them to perform at top form (including skipping game prep) are not luck factors.

maybe he got a hickup at a crucial time.

Really? A hiccup? That's the best we can get for a random chance factor in my example?

But the poor player will almost always lose, even if a number of things go his way.

This is where you are wrong. In practice 3 point shots are pretty difficult; some of the best shooters in the pro leagues can hit less than 50% of them sometimes. PRO shooters. Now, some college players will occasionally hit well over 50%. Does that make them better shooters? No, because all time the pro hits more.

However, if you made them do 10 sets of 10 3 point shots, I'm betting the college guy would hit more out of 10 at least once or twice. On the same day, with the same extenuating circumstances. You can't wiggle out of this one, there is no random chance here, just variable levels of performance from the same guy. This example is not unique, either, it is typical of competition between closely matched players.

This is what I meant when I wrote "theoretically impossible and practically almost so". I see no logical flaw here.

The only logical flaw was in what was probably a typo (given that you changed wording this time), I was just giving you some grief over it .

I had the impression that you object to events in competitions because they may create an unfair advantage. My point was that being lucky is not an unfair advantage. It is not unfair. It is just luck.

Not exactly. If you read up to one of my recent posts you'll see why I'm actually arguing against events. The luck might be "fair", but it nevertheless has an adverse impact on whether player skill determines the outcome. Strictly speaking, that can still be fair, but it is bad for competition. It's also a barrier to competition in Hall of Fame (since players who can game spam can get better event draws through sheer #games played).

I play random civs and from that point of view which civ and uniques I end up with is a dice roll.

. These things don't HAVE to be random. If you make them deliberately, they are I suppose.

Religion spread may affect your strategy for the rest of the game, and one can very rarely controll wich one spreads to one first.

Get open borders. In bts 3.19 most religion founders will ship missionaries to you if you get OB with them early. I know Ramesses Rules has trouble with this, but he doesn't have BTS. I have pretty much never struggled (or extremely rarely) to get the religion of the AI I wanted into my lands, assuming they were anywhere near me early on.

No one can ever demonstrate that a certain event has a specific impact, because the event does not stand alone, but is a result of every action in the game thus far.

Killing you on turn 30 is pretty decisive . Bermuda triangle sinking 10000's of hammers when you're en-route to intercept culture win is game-breaking (none of this early game decision crap; bermuda triangle happens because events are on and it trips. There is no strategic link between settling a city t35 and the event, don't dodge realities). Strong events like tower shields don't win games vs the AI by themselves, but can EASILY be a few turns advantage in HoF, and that can cause the difference. Try breaking a choke when your opponent gets shock on prats or something in MP, too.

I'm not saying events can't be fun to people because obviously they are to some. My point is that they shouldn't be on in competitions, and your bridge example is a good example of a reason why, as are most competitive sports which are notably LACKING anything comparable.

 

[Utenlok]

Sports outcomes are extremely vulnerable to chance. Your QB gets hurt on first play of the game, lucky bounce of the ball, etc. That is why many major sports play "best of" playoffs to try to reduce luck getting the lesser team through on a fluke win.

 

[strijder20]

TMIT, surely the RNG which made the player whose culture win you were trying to intercept go culture instead of space is random too. I do not see how you can complain about random events in a game based on a RNG. In completely predictable games like chess, yes, there randomness is completely out of place, but here?
Say, the RNG decides to give you no early strategic resources. Shaka is next door. You are in trouble.
The RNG decides to roll an artist instead of a GS to bulb Education in your NE city when you are desperately trying to get Liberalism before Mansa Musa. You suffer a severe setback.
The RNG spreads a jungle on your gem tile in your capital's BFC which you were about to mine. You also suffer a severe set back, or do not get a nice boost, as grassland gem tiles are quite powerful early in the game.
The RNG decides to kill 3 more cavalries then it should have killed on average. You completely optimalised production, trained the maximum amount of cavalries and attacked the right person on the right moment on the right place, but because of those three cavalries, you will not be able to completely wipe out the civ, which will get rifling when they still have 5 cities left. Of course, they upgrade all their protective muskets to rifles. Great.
But : the more randomness there is in the game, the more likely it is it cancels out itself. Imagine you and me are playing a coin-tossing game. I, however, cheated and took an unfair coin, which will show heads 66% of the times. When it shows heads, I win, when it shows tails, you win. If we throw once, I will win 66% of the times, you will win 33% of the times.
When, however, we toss the coin 3 times, and look who has won the most times, I will win 20/27 of the times, you will win 7/27 of the times. It has shifted slightly in my favour.
So in most games, including these, the more random factors you add, the more likely you will get an 'average' result. I do agree with you, however, that some ridiculously overpowered events like the Bermuda Triangle should be removed.

 

[TheMeInTeam]

TMIT, surely the RNG which made the player whose culture win you were trying to intercept go culture instead of space is random too. I do not see how you can complain about random events in a game based on a RNG. In completely predictable games like chess, yes, there randomness is completely out of place, but here?

It's degrees of impact, please address the points I put forth already.

Say, the RNG decides to give you no early strategic resources. Shaka is next door. You are in trouble.

Shaka techs very, very slowly, just FYI. Again, the existence of other RNG factors is not a true counter-argument to any of my points.

But : the more randomness there is in the game, the more likely it is it cancels out itself.

How about no? Read R_Rolo1's posts here re: events canceling out. Or use logic. Say you have x y or 1 2 3. Possible distributions:

xx
xy
yy

Or

111
112
113
122
123
222
223
333

What do you have? While the odds of extreme outliers decrease, the odds of having negative and positive outcomes truly "even out" are actually reduced (IE 2/8 is less than 1/3); you are still likely to either get more positive or more negative outcomes in a given playthrough. Minimizing the #factors that can be game-deciding independent of strategy is a good thing, not a bad thing...especially in settings where people have time to spam games until they get outlier results favoring them...but also in settings where everyone plays the same start (IE a small difference between equal players can mean the outcome, and now you have a higher incidence of a small difference).

By the way, in *competitive settings*, either everybody is spawning next to shaka, or nobody is. In MP you rarely have trouble with anything but land distribution, which can be ironed out with map script + balanced resources or a map made by a 3rd party. In XOTM everyone plays the same start. In HoF, you get to pick your opponents and even your start, meaning that the limiting factor is #tries. Why do you want to FORCIBLY ADD #tries to that equation? It doesn't make sense.

Hopefully this demonstrates why

So in most games, including these, the more random factors you add, the more likely you will get an 'average' result.

Is logically false.

 

[strijder20]

It's degrees of impact, please address the points I put forth already.



Shaka techs very, very slowly, just FYI. Again, the existence of other RNG factors is not a true counter-argument to any of my points.



How about no? Read R_Rolo1's posts here re: events canceling out. Or use logic. Say you have x y or 1 2 3. Possible distributions:

xx
xy
yy

Or

111
112
113
122
123
222
223
333

What do you have? While the odds of extreme outliers decrease, the odds of having negative and positive outcomes truly "even out" are actually reduced (IE 2/8 is less than 1/3); you are still likely to either get more positive or more negative outcomes in a given playthrough. Minimizing the #factors that can be game-deciding independent of strategy is a good thing, not a bad thing...especially in settings where people have time to spam games until they get outlier results favoring them...but also in settings where everyone plays the same start (IE a small difference between equal players can mean the outcome, and now you have a higher incidence of a small difference).

By the way, in *competitive settings*, either everybody is spawning next to shaka, or nobody is. In MP you rarely have trouble with anything but land distribution, which can be ironed out with map script + balanced resources or a map made by a 3rd party. In XOTM everyone plays the same start. In HoF, you get to pick your opponents and even your start, meaning that the limiting factor is #tries. Why do you want to FORCIBLY ADD #tries to that equation? It doesn't make sense.

Hopefully this demonstrates why



Is logically false.

Let me be clear - I do NOT want to forcibly add #tries as a must-do for a high score. It just clears away all skill level - I might pop AH, HBR, Pottery and The Wheel from four huts, clearly boosting my chances of winning.
The more amount of random events there are, the more extreme the extreme value gets, true. But the more amount of random events, and random factors in general, there are, the less chance you will get an extreme value. But indeed, with #plays everything can be achieved .
So, although I'm still convinced if you compare one single game played with another single game played, you will get more equal results if random events are turned on - after all, how much do we get the Bermuda triangle and the like - but you are right about the fact that they encourage huge amount of tries to get one extremely good result.

According to this nonsense definition, Rolo's previous definitions he assumed based on your story are accurate. You really are asserting that 999 good events and 1 catastrophic one are fine, and that 999 bad events and one game winner are examples of "averaging out". Also, because you use averaging out as a defense of events, you are asserting that such an event pattern would be fine.

However, there is nothing "average" about those outcomes, despite that they fit your definition. If you use a wrong definition as a core basis for an entire argument, that does not help the credibility of the argument much...

I think AJ is pointing at the fact that in general, over a large amount of games, and looking at those games as a group, events won't do you much harm or help, if added together. While you are saying that in a single game, it is very possible to get harmed or helped by events a lot. Two different things.
Of course, looking to a group of events is not exactly what we should do here - it is one single game which get submitted for HoF, not all the other tries too.

 

[krikav]

I googled pita to see what it meant in English that was related to food and it was bread.

I don't know why you think TMIT is/was a vegetarian, and I don't know why would vegetarians not eat something made out of wheat. Unless that is that your veg didn't mean vegetarian.

And no, I wasn't talking about bread.

Mixed Pita up with PETA!

 

[logotet]

@Ataxerxes
I’m not sure i understand your bridge reference. I was talking about minimizing the chance factor by transporting played hands unaltered to other tables. I believe it’s called duplicate bridge, but am not sure since English is not my first language. Another reference I made was to the IMP scoring system where it is common to achieve better score with a lesser hand.


@TMIT
I am actually quite amazed that you contest that “sports have extremely minimal elements of chance”. Speechless really. It doesn’t matter what the spread is, bookies will always lose money if they offer odds on 0 random events. The whole idea to betting on sports is that there are no any “sure things”. It’s a bookmakers’ general practice to remove games from the list if a “fix” is suspected, by which I mean an illegal agreement between the sides to achieve a particular result.

I don’t think I quite follow your example with the basketball player shooting 3 pointers. First you talk about a “poor player” and a “good player”, then about a pro and a college player. Quite a difference there. Then you lost me claiming that this competitive example had “absolutely no luck factor”. Do you really believe that?

I generally agree that “an additional chance element such as events DOES make a difference”. But we are not talking about a laboratory experiment here, we are talking about a game that typically has many thousand random decisions during it’s course. It’s quite possible for instance for one to get lucky and conquer a neighbouring capital early on with two warriors vs. an archer defender. More often though one will need about four warriors to get the job done. Producing four warriors takes considerably longer time though and delays a much needed worker. So a couple of random battle results can have a rather huge impact on the game, decisive even. Now how is that better than a random event? The Vedic Aryans “killing you on turn 30 is pretty decisive” too, I agree. But what is the difference between the two? Both are definitely random based. Why should one be less acceptable than the other? Besides, by turn 30 one could have had 3-5 warriors defending ones capital and made the Vedic Aryans bite the dust. Yes, I almost always go worker first, but then I accept the risk of being run over. It’s a logical choice, because chance of the Aryans is rather small. Anyway, all these are repetitions of earlier discussion in the thread, I guess my point is that events are merely one more random element in a game that already is heavily influenced by random elements. That is why I don’t think your parallel with hockey teams randomly benching players is relevant. Hockey is not a game based on a variety of random calculations.

I agree with you that “luck has impact” on the outcome of a competition, but is it really “adverse impact”? I don’t really understand your beef with some lucky HoF entries. To me the highest score is like a world record – a really exceptional achievement, that you must work very hard to beat. Yes, some people choose to spend a lot of time replaying games. It’s only natural that their payoff should be larger, their investment is larger too. What is the basis of the claim that lucky high scores discourage competition? More logical would be the opposite – a reasoning along the lines of “hey, I know I’m not the best player out there, but under right circumstances I could win”. The element of luck in fact encourages competition, making even less skilful players/athletes feel they may have an outside chance.

Possible distributions:
111
112
113
122
123
222
223
333

A little bit off topic, but there is something erroneous about you 1 2 3 distribution example. It looks like you are talking about multicombinations and in that case there are some possibilities missing. The total number of possibilities should be (5*4*3)/(3*2*1)=10, which means you’re missing two. The most common outcome (6 out of 10) is to have two of the same element and another one (112 or 223), the second most common outcome is to have three of the same (3 out of 10, i.e. 111 or 222). Finally the least probable outcome is to have one of each (1 out of 10, i.e. 123). It is rather unclear why you single out the combinations 122 and 123. I must be missing something.

 

[TheMeInTeam]

It is rather unclear why you single out the combinations 122 and 123. I must be missing something.

They are the 2 closest to the center of the distribution. I also didn't want to make the post needlessly long. Strictly speaking, even those outcomes are slightly imbalanced but they're as close as you're getting in that model.

To me the highest score is like a world record – a really exceptional achievement, that you must work very hard to beat.

To me and many good competitors the HoF is like a display of who played enough games to get a lucky outcome. I'd rather see the best play get rewarded, not the luckiest play. HoF already takes steps to reduce #tries in other facets. The allowance of map finder and choosing oppositions but also essentially mandatory huts/events is completely mind boggling. They serve counter-purposes.

Warrior rush vs aryan doesn't hold water by the way. Try rushing anyone halfway decent with warriors in MP and see what happens. However sudden aryans would kill most players. Why the differentiation? I'll leave that as a mystery for you to figure out, but they are NOT even remotely comparable .

I don’t think I quite follow your example with the basketball player shooting 3 pointers. First you talk about a “poor player” and a “good player”, then about a pro and a college player. Quite a difference there. Then you lost me claiming that this competitive example had “absolutely no luck factor”. Do you really believe that?

Relative to a top pro, a college player is poor. (He is also obviously better than most people, but same can be said for top civ players vs decent ones vs general public...). The previous assertion is that the pro would win a 3 point shoot-off 100% of the time, which is both baseless and ridiculous.

I also like how people insist this little example has a luck factor to it, and then refuse to provide an example of luck affecting the outcome . SURELY, you can come up with something better than drinking alcohol the previous night, which is an obvious player choice?

I think AJ is pointing at the fact that in general, over a large amount of games, and looking at those games as a group, events won't do you much harm or help, if added together. While you are saying that in a single game, it is very possible to get harmed or helped by events a lot. Two different things.

Every single competition in civ IV I've seen competes at the single-game level where the outcome of 1 game is extremely meaningful (even the challenger series, which scores points based on 10 games, gives a tremendous point value for ranking highly on an individual game). Using an argument that events don't have much net effect over a large # of games is completely borked and irrelevant in terms of competitions...but that is the only facet of civ IV I've argued events need to be off. What's the point of that stance then? None.

So, although I'm still convinced if you compare one single game played with another single game played, you will get more equal results if random events are turned on - after all, how much do we get the Bermuda triangle and the like - but you are right about the fact that they encourage huge amount of tries to get one extremely good result.

Not quite. You're missing key components of the competitions!

HoF: Game spam just goes on until the outliers are realized.
BOTM: Everyone starts with exactly the same map, leaving the opportunities for other bad "random" draws very limited, especially given that everyone starts with the same map seed.
MP: We see settings like mirror, balanced resources, or maps edited by 3rd parties to be fair. Not seeing how the addition of events makes these games more fair on average.

Competitions already have rules and efforts to maximize the impact of skill on the game; and yet ignorance (IE not knowing the implications) has allowed events to creep into these formats at times.

 

[strijder20]

Every single competition in civ IV I've seen competes at the single-game level where the outcome of 1 game is extremely meaningful (even the challenger series, which scores points based on 10 games, gives a tremendous point value for ranking highly on an individual game). Using an argument that events don't have much net effect over a large # of games is completely borked and irrelevant in terms of competitions...but that is the only facet of civ IV I've argued events need to be off. What's the point of that stance then? None.

I think we have a bit of a misunderstanding here. Yes, it is irrelevant - but I'm not talking about competitions here. I'm just saying that, over a large amount of games, the amount of points you will gain/lose by random events will not be much - but in a single game, the difference can be huge.
If I play 10 games, and in one game my neighbour gets taken out by Vedic Aryans, then I might have these scores :
200K, and nine times 100K.
Without events, that would have been 10 times 100K. A difference of 10%.
If we single out that game, though, then there is a difference of 100%.
So what I'm saying is that if in a competition where you have to submit a game, but can only try once (how they would do that I have no clue about), events are potentially much more game-deciding than if you had to submit 10 games (and were only allowed to try once a game).

Not quite. You're missing key components of the competitions!

HoF: Game spam just goes on until the outliers are realized.
BOTM: Everyone starts with exactly the same map, leaving the opportunities for other bad "random" draws very limited, especially given that everyone starts with the same map seed.
MP: We see settings like mirror, balanced resources, or maps edited by 3rd parties to be fair. Not seeing how the addition of events makes these games more fair on average.

Competitions already have rules and efforts to maximize the impact of skill on the game; and yet ignorance (IE not knowing the implications) has allowed events to creep into these formats at times.

In MP mirror, those events obviously increase the amount of randomness - because it has already been tuned down to combat odds and GP odds. But if you already have a lot of random factors, another amount of random factors thrown in will increase the extrme values, but make the result closer to the average on general. So, if we both would play one game, without retrying, and say we are exactly of the same skill level, then on average our results would be closer to each other with events on than with them turned off.
Mind you, I'm talking about one game, one try - not one game submitted, but possible hundreds of tries.

 

[logotet]

I'll leave that as a mystery for you to figure out, but they are NOT even remotely comparable

Magnificient argument, I'm convinced.

I also like how people insist this little example has a luck factor to it, and then refuse to provide an example of luck affecting the outcome

Another great argument. If there was no randomness to shooting 3 pointers the result would always be the same. Do you really want me to start say things like tripping over shoelace, hiccup and so on? If you find it mundane, imagine a meteor obliterating the site of the experiment.

I don't really care about the HoF issue and agree it is a list of who got most lucky circumstances. It is still puzzling what you have against it - it's quite clear what it represents, so why do u care about it? Make an alternate list and look at that instead. Then the person who got a decent score while on the receiving end of both Aryans and Bermuda triangle event can make yet another list and look at that. To each his/hers own.

 

[TheMeInTeam]

I think we have a bit of a misunderstanding here. Yes, it is irrelevant - but I'm not talking about competitions here.

I am only arguing against events in competitive settings. Outside of those it is player preference. If you're only arguing in favor of events in more casual settings, we're not actually opposed here .

(how they would do that I have no clue about)

HoF mod has some good stuff there. It's used in both BOTM and HoF. Obviously in MP you can just run 1 session!

Magnificient argument, I'm convinced.

Ok fine. I didn't want to walk you through basic gameplay, but since you insist:

1. Build warriors ----> power graph increase. Yes, on both the power graph (and anybody near you should be focused) and on the demo screen (you can always tell when someone builds a warrior so early because it adds 2000 power). By simply checking that screen once per turn, even on a blazing timer, you can easily determine whether someone is investing in warriors rather than a worker and act accordingly.

Of course, vedic aryans don't even show up on the power graph and are "constructed" instantly.

2. Travel time. Warriors produced have to travel tiles to reach an opposing capitol, often 10+. Even for tight spawns the travel time is easily the difference of a warrior, but often several.

Vedic aryans in this timeframe have to travel at most 3 tiles. In comparing to early warriors, they've essentially teleported to you.

3. Map setup. You can easily use map size + #players + map type to get a reasonable estimate of the threat of a warrior rush. Warrior rush threat on an islands map is nil, on crowded pangaea very real.

Of course, Aryans can kill you even on islands.

4. Archers are stronger than warriors.

I really didn't want to walk people through this, but whatever. This should be crystal clear evidence of the differences between warrior rush and aryans, and it's large enough that they aren't remotely comparable.

Then again, wouldn't an experienced player who understands the full implications of events enough to use heavy arguments in their favor for competitive settings know something like this? SURELY, players arguing in favor of usage of events in competitive settings have some actual knowledge of the topic of discussion, right?

I don't really care about the HoF issue and agree it is a list of who got most lucky circumstances. It is still puzzling what you have against it - it's quite clear what it represents, so why do u care about it? Make an alternate list and look at that instead. Then the person who got a decent score while on the receiving end of both Aryans and Bermuda triangle event can make yet another list and look at that. To each his/hers own.

Yes, let's dilute the competition as much as possible. Actually, I'd rather not do that, because it undermines actual competition.

Another great argument. If there was no randomness to shooting 3 pointers the result would always be the same.

What is the basis for this assertion?

None!

What credibility does an argument with no basis have?

None!

We go on to get 2 more examples that aren't random at all and a 3rd which would prevent the competition altogether. Nevermind the complete refusal to acknowledge the VERY REAL possibility that with no extraneous factors, the player who is worse, on average, simply hits more shots out of 10 this time. No shoelace issues, no wind, no sun in the eyes, no tooth fairy stompings, nothing. Same conditions, same spots to make the shot attempts, I guarantee the guy who hits less shots on average would take some series. So once again, we have a concrete example of a competition where the lesser player wins on occasion, that is completely and utterly independent from luck in the vast majority of cases. That can be extrapolated into the rest of competition by the way.

And it completely annihilates the assertion that "without random elements, the better player/team will always win".

That assertion has been PROVEN wrong, unless you can explain all non-random variance in those competitions. If you can't, the entire basis of random-ness being necessary for different outcomes in competitions is undermined and that argument is now extremely weak.

 

[logotet]

Nevermind the complete refusal to acknowledge the VERY REAL possibility that with no extraneous factors, the player who is worse, on average, simply hits more shots out of 10 this time. No shoelace issues, no wind, no sun in the eyes, no tooth fairy stompings, nothing. Same conditions, same spots to make the shot attempts, I guarantee the guy who hits less shots on average would take some series. So once again, we have a concrete example of a competition where the lesser player wins on occasion, that is completely and utterly independent from luck in the vast majority of cases. That can be extrapolated into the rest of competition by the way.

I see. No extraneus factors. You mean that eliminating "luck" completely is a basic condition. Well, of course, if luck is not present per definition, it cannot possibly matter. However, I know not of any situation in the physical world, where eliminating all axtraneus factors is possible. There is no practical model that takes into account all possibilities. We cannot command the basketball players to not blink, sneeze etc. ridiculous activities during the shoot off. Same imput equals same result. 2+2=4 2+2+x may equal 4, but also may not. Call x luck or a combined value for extraneus factors if you will. If the result differs, the input must differ. Therefore the difference cannot be found within the known factors, the 2+2, but within the unknown, the x. You say that x can be made to not matter, which is rather unorhodox.

It's amusing though that someone can "guarantee" the outcome of an impossible situation, which in itself constitutes a "concrete example".

Ok fine. I didn't want to walk you through basic gameplay, but since you insist:

1. Build warriors ----> power graph increase. Yes, on both the power graph (and anybody near you should be focused) and on the demo screen (you can always tell when someone builds a warrior so early because it adds 2000 power). By simply checking that screen once per turn, even on a blazing timer, you can easily determine whether someone is investing in warriors rather than a worker and act accordingly.

Of course, vedic aryans don't even show up on the power graph and are "constructed" instantly.

2. Travel time. Warriors produced have to travel tiles to reach an opposing capitol, often 10+. Even for tight spawns the travel time is easily the difference of a warrior, but often several.

Vedic aryans in this timeframe have to travel at most 3 tiles. In comparing to early warriors, they've essentially teleported to you.

3. Map setup. You can easily use map size + #players + map type to get a reasonable estimate of the threat of a warrior rush. Warrior rush threat on an islands map is nil, on crowded pangaea very real.

Of course, Aryans can kill you even on islands.

4. Archers are stronger than warriors.

I really didn't want to walk people through this, but whatever. This should be crystal clear evidence of the differences between warrior rush and aryans, and it's large enough that they aren't remotely comparable.

Then again, wouldn't an experienced player who understands the full implications of events enough to use heavy arguments in their favor for competitive settings know something like this? SURELY, players arguing in favor of usage of events in competitive settings have some actual knowledge of the topic of discussion, right?

I must be unfocused or something, but what are you talking about? I said that one can luck out and early on conquer another capital defended by an archer with just 2 warriors. Then again u can have bad luck with the Aryans and get your bottom kicked. Both events may have a decisive impact on the game and both depend on random factors. My point was that since none require specific skills, they should be treated equally competitionwise. What is your point in what you write? I see the sarcasm in point 4 but honestly have no clue what it is about. Yes, archers are stronger than warriors. Has anyone claimed otherwise?

 

[civac]

We go on to get 2 more examples that aren't random at all and a 3rd which would prevent the competition altogether. Nevermind the complete refusal to acknowledge the VERY REAL possibility that with no extraneous factors, the player who is worse, on average, simply hits more shots out of 10 this time. No shoelace issues, no wind, no sun in the eyes, no tooth fairy stompings, nothing. Same conditions, same spots to make the shot attempts, I guarantee the guy who hits less shots on average would take some series. So once again, we have a concrete example of a competition where the lesser player wins on occasion, that is completely and utterly independent from luck in the vast majority of cases. That can be extrapolated into the rest of competition by the way.

And it completely annihilates the assertion that "without random elements, the better player/team will always win".

That assertion has been PROVEN wrong, unless you can explain all non-random variance in those competitions. If you can't, the entire basis of random-ness being necessary for different outcomes in competitions is undermined and that argument is now extremely weak.

Your basketball analogy is confused. In particular, the statement that there is no randomness in your basketball example is meaningless. (I'm not supposing that it is wrong but that discussing its truth value makes no sense.) Consider a basketball player preparing for a 3 pointer. I suspect that given complete information on the initial conditions and sufficient computation power we could determine whether he hits or not. We could also likely calculate which of two players wins a contest involving a series of 3 point shots. (Whether there is actual randomness in the real world is subject of ongoing debate.) So in that sense the basketball example is not random. However, this has nothing to do with how the term randomness is usually used.

The basis of a statement like "He has a 50% chance to hit." is actually a mathematical model of the occurence. For the basketball example we might construct this model by considering the average 3 point hit percentage of both players and assuming that the probalbility of hitting any future shots is equal to these averages. In fact you used this very line of reasoning in an earlier post. Using our model, we may compute approximate winning chances for both players.

The key point here is that we are talking about a model, a mathematical tool used to give approximate outcomes in situations which are way to hard to predict precisely due to lack of information and computing power. In that sense every sports contest can be viewed as a sequence of random events. In fact, this viewpoint has been very successful in the past. So much so that it is widely used in coaching, tactics discussion and betting.

The situation in Civ 4 is somewhat different. For instance, a battle between warriors in Civ 4 is already a model, namely of two groups of people hitting each other with clubs.

 

[Ataxerxes]

We can never eliminate the luck factor, true. For games where results are compared, we just try to minimize it so people can see how other players did in the same circumstances. In a game with the same starting conditions, one player gets the Vedic Aryans and is eliminated. The second gets an early slave revolt, slowing his early expansion. The next gets Tower Shields. The strategy can change due to events and then what one player does isn't what the other does, even though both were right. HOF games need luck anyway, so I don't know about them. I'm thinking of something like Nobles Club.

@logotet. Sorry I confused you with bridge. You were thinking of IMP scoring, where luck is somewhat reduced. I was thinking of Matchpoints, which has about the least luck (although it's still prevalent). Matchpoints is the more common form of competitive bridge in the U.S. - not sure about other countries.

 

[TheMeInTeam]

It's amusing though that someone can "guarantee" the outcome of an impossible situation

Impossible situation? You mean impossible like the blanket assertion that something will always end the same way without a luck factor? You can no more prove that assertion than I can there are no extraneous factors. What, am I the only one not allowed to use that kind of tactic? I didn't start its usage though .

I must be unfocused or something, but what are you talking about? I said that one can luck out and early on conquer another capital defended by an archer with just 2 warriors. Then again u can have bad luck with the Aryans and get your bottom kicked. Both events may have a decisive impact on the game and both depend on random factors. My point was that since none require specific skills, they should be treated equally competitionwise. What is your point in what you write? I see the sarcasm in point 4 but honestly have no clue what it is about. Yes, archers are stronger than warriors. Has anyone claimed otherwise?

*COUGH* *COUGH* *COUGH* *COUGH* *COUGH* *COUGH* *COUGH*:

Producing four warriors takes considerably longer time though and delays a much needed worker. So a couple of random battle results can have a rather huge impact on the game, decisive even. Now how is that better than a random event? The Vedic Aryans “killing you on turn 30 is pretty decisive” too, I agree. But what is the difference between the two? Both are definitely random based. Why should one be less acceptable than the other? Besides, by turn 30 one could have had 3-5 warriors defending ones capital and made the Vedic Aryans bite the dust.

Aside from the fact that it is poor play to keep 3 warriors in the capitol in every non-event case that early on (aside from human warrior rushes that you've already concretely identified), hopefully you can see where the response came from, and why aryans have NOTHING in common with early warrior play. Not that the original assertion was relevant anyway.

Your basketball analogy is confused. In particular, the statement that there is no randomness in your basketball example is meaningless.

The statement that randomness is required for different outcomes is meaningless. Gee, aren't these kinds of arguments fun?

The key point here is that we are talking about a model, a mathematical tool used to give approximate outcomes in situations which are way to hard to predict precisely due to lack of information and computing power. In that sense every sports contest can be viewed as a sequence of random events.

Nonsense. It's like saying going into work in the morning is a random event. In that sense, everything is random, but such a definition is worthless because it encompasses absolutely everything.

Now, tell me what is random about a 3 point shot from the same place, without going esoteric. The guy takes the shot, and if he uses the same force at the same angle from the same spot from the same height he's going to hit the shot every single time. That doesn't happen often, though, because it's difficult to do it. If you can't prove there's chance here and I can't prove there isn't, it's a hung argument and to be honest, not a particularly relevant one. The initial assertion that "all outcome variance between two players is chance" itself is unprovable, and as a result using it as the basis for an argument is questionable.

You know what is amusing? The 5 point summary I gave against events earlier on...well...hasn't even been addressed since. That, by itself, is very telling. Derailing me from that has been an effective diversion, but it still hasn't given any remotely decent counter-points to the assertion that "events should be off in competitive settings", nor have those points been refuted as of yet. Rather than using an impossible assertion as the sole basis of an argument (and then arguing over my own impossible assertion despite it being a copy cat tactic), maybe we can get a little bit more on topic and show how events can possibly be a good thing for competitions that I listed?

 

[civac]

You have a point that the probability stuff is not central to the question if there should be random events in competitive Civ games. Moreover, I actually agree with you there should not. They add a bit of flavour and excitement I suppose and some people may think this more important than the disadvantage of adding more randomness. It wouldn't be a big deal if Civ 4's random events were better balanced. But I would still advocate leaving them off.

Your basketball analogy kinda piqued my interest. That's why I answered in the first place. I put my response rgarding that part of your post in a spoiler. I agree it's a side issue. Feel free to ignore it if you will.

Spoiler :

I'm sorry to say that you completely missed my point. A 3 point shot from the same position with the same force and angle will always have the same result, yes. But humans are unable to control position, angle or force to sufficient degree. So sometimes they will hit and sometimes not even given the exact same outside conditions. From a physical viewpoint this is not random since, of course, there is no such thing. (Assume we are both determinists.) However, in practice it may as well be because you lack the information to predict it. You can do quite well by assuming it is random and assigning probabilities based on experience. That is the model I was talking about.

Your way to work can also be considered as a sequence of random events. I don't understand why you think this is a refutation to my point. You can even make meaningful estimates about, say, your arrival time provided you have solid statistical information on your previous experiences. It just is generally not very interesting to do so.

For the basketball example not only it is much more interesting to have some estimation on whether a player will hit a 3 point attempt but the kind of model I talked about is actually used all the time in practice. Example: A manager has two players one hitting 3 pointers in 48% of cases and another hitting for 40%. He will likely tell his players to create chances for the first player to shoot while the second is to block or go closer to the basket. This is meaningful information gained from the statistical model. It is important that we are talking about the model here not physical reality. From a physical viewpoint saying player A has a 48% chance to hit makes no sense at all. It is certain whether he will hit or not on a given attempt. We are merely unable to predict it due to our limitations.

 

[LegionSteve]

The basketball thing piqued my interest too. It seems like people are coming at it from two different angles.

Some people are looking at it from the point of view of the Civ player as the basketball coach, and the players as the random events. The coach knows that player A has 65% chance of making a 3 point shot, and player B has 55% chance. From the coach's point of view if he tries to guess who will win over 10 shots there is an element of luck involved.

If I understand TMIT's argument, player A and player B are the Civ players. There will still be some variance in who will win over 10 shots but it is not luck, because the 55% and 65% are based on the players' skill. In this context, the random events are for example: as either player takes his shot there is an earthquake/naked woman on the court/bee stings him in the ass, i.e. nonsense.

Completely agree on the principle of not adding purely luck based things to competitive Civ games. However I don't play Civ competitively so I don't really care.

For me it wasn't the game-endingly ridiculous things that made me turn events off. It was the whack-a-mole tedious ones: pay money to stop something lame happening ... rebuild that pasture ... rebuild that mine ... pay money to stop something lame happening ... rebuild that pasture ... rebuild that mine ... bored now

 

[TheMeInTeam]

If I understand TMIT's argument, player A and player B are the Civ players. There will still be some variance in who will win over 10 shots but it is not luck, because the 55% and 65% are based on the players' skill. In this context, the random events are for example: as either player takes his shot there is an earthquake/naked woman on the court/bee stings him in the ass, i.e. nonsense.

Yes, I am going from the perspective as a player in basketball, much as I am going from the perspective of a player in civ IV. When you consider another person's activities outside of your control, I guess you can factor it as an element of chance, even though strictly speaking the outcome is centered on the decision-making of the other individual.

 

[Utenlok]

Keep in mind that 40% on 3 pointers puts you in the top 20 all time in the NBA. When you consider that 20% to 30% is solid on 3's, a run of misses or makes happens a lot more often.

As far as random events go, can't you guys just agree that no matter what stats or "facts" you try to use that it is an opinion and you just disagree?