Gaming Do you think FFTA2 hit chance is a fake?

[jabjab]

You flipped a coin 10 times. And you got 10 heads. The chance of getting 10 heads in a row is 1/1024 ~= 0.1%. Now you're going to flip the same coin again, but you need to bet first. Are you going to bet on heads, or tails? And why?

heads - you didn't give enough info so it could be double headed or weighted and if its fair just as well to bet on heads as tails.

If it were double-headed, the chance of getting a heads 10 times in a row would be 100%
If it were weighted, the chance of getting a heads 10 times in a row wouldn't be 1/1024 due to a different chance.
Since coins are generally two-sided(I've yet to see oen that's not), it is fair to assume that it is a fair coin with a different value on each face (e.g. heads or tails).


As for FFTA2, I think the percentages are... weird... The CPU has missed on a 99% one before... I was like =O And you can easily hit people from the front when it would have been different in another FFT game.

ah but thats assumes that the statistics are based on facts and not assumptions.
Also coins have 3 sides if you include the edge but good luck getting one landing there.
In the same way the analysis of the stats in game may be based on assumptions.
Games dont just use the displayed 'hitrate' there can be many other factors such as internal capping, character levels, special modifiers and whatever else they decide is relevant - the displays may just be a 'base chance' which is then modified.
That said id be annoyed if 3 99% displayed chance attacks failed in a row.

 

[JBates]

And you can easily hit people from the front when it would have been different in another FFT game.

I'm pretty sure they changed that. Now hitting from the side increases the damage you deal, and the back increases it a bit more. Someone correct me if I'm wrong though.

 

[ivykinz]

The hit rates could also be based on what the actual hit rate would be not taking the opponents stats into consideration.
Like in theory you can hit target X 99% from the front if it was a 0 stat dummy.
But if like the target has an evasion rate of maybe 10, then in reality the actual hit rate would be like 60% without you actually knowing it.

i disagree, because when your target is sleeping, your hit % is significantly higher than normal.

 

[Hyperlisk_]

And you can easily hit people from the front when it would have been different in another FFT game.

I'm pretty sure they changed that. Now hitting from the side increases the damage you deal, and the back increases it a bit more. Someone correct me if I'm wrong though.

Yeah, I noticed that too. I find it kind of annoying however, being used to the other FFT games...

 

[deathfisaro]

Yup so some people got the coin thing right. Regardless of your previous outcome (10 heads in a row) the next flip is still 50/50. So it doesn't matter which side you bet on.

And missing 3 of 99% in a row is the same. On your first attack, you have 1% chance of missing. If you've missed that, the next attack still has 1% chance of missing. And so on.
Saying missing 3 of 99% in a row is 0.0001% is only applicable if you have attacked only three of 99% in the game and all of them were missed.

It's gonna get a tiny bit technical, but you can read on if you don't mind.

Okay, so I'm a slow player, but I throw more than 300 attacks per 2 hour periods. (I actually counted how many attacks I did in one battle and measured the battle time. Pumped up to 2 hours and got over 300)
Over 300 attacks as in over 300 calculations, not 300 actions. (For example, if you cast Fire on group of enemies that are standing +, that's 5 calculations from one action)
I spend most of my time reading quest descriptions, dialogues, changing equips and such. So if you're battle oriented fast reader, you could throw 300 per hour.

Okay, so let's say I've performed 300 attacks of which all of them were 99%. (Unlikely but for the sake of proving the point) Expected number of hits would be 297 (300 * 0.99).

The chance of hitting 297 out of 300 is (300! / 297! * 3!) 0.99 ^ 297 * 0.01 ^ 3.
Don't let the formula get in your head, but if you may have notice that the order of hits and misses are totally irrelevant.


This means, as long as you're missing 3 out of 300, 3 misses in a row and 1 miss every other 99 is exactly the same.
The only difference is, you immediately notice when you get 3 misses in a row and think the game is so cheating. When you miss 1 every ~100, you think that's very natural because 99% isn't exactly 100%.

"The law of large numbers" is a mathematical law that describes how random outcomes of fixed percentage get close to that percentage as the number of tries get bigger.

If you threw 50 of 99% attacks and missed 3, that doesn't seem right at first. But as you throw more attacks (as in hundreds and thousands more times, because you're likely to play for dozens of hours =P) the cumulative hit rate becomes closer and closer to 99%.


Sorry about that big wall of math. I just wanted to throw a BIG rant and it happened to be about FFTA2 and its hit percentage.

 

[Syao4]

Yes the hit rate is messed up.
Missing with 95%? That's stupid as hell.
Those useless misses can screw you..

 

[Jackreyes]

Yup so some people got the coin thing right. Regardless of your previous outcome (10 heads in a row) the next flip is still 50/50. So it doesn't matter which side you bet on.

And missing 3 of 99% in a row is the same. On your first attack, you have 1% chance of missing. If you've missed that, the next attack still has 1% chance of missing. And so on.
Saying missing 3 of 99% in a row is 0.0001% is only applicable if you have attacked only three of 99% in the game and all of them were missed.

It's gonna get a tiny bit technical, but you can read on if you don't mind.

Okay, so I'm a slow player, but I throw more than 300 attacks per 2 hour periods. (I actually counted how many attacks I did in one battle and measured the battle time. Pumped up to 2 hours and got over 300)
Over 300 attacks as in over 300 calculations, not 300 actions. (For example, if you cast Fire on group of enemies that are standing +, that's 5 calculations from one action)
I spend most of my time reading quest descriptions, dialogues, changing equips and such. So if you're battle oriented fast reader, you could throw 300 per hour.

Okay, so let's say I've performed 300 attacks of which all of them were 99%. (Unlikely but for the sake of proving the point) Expected number of hits would be 297 (300 * 0.99).

The chance of hitting 297 out of 300 is (300! / 297! * 3!) 0.99 ^ 297 * 0.01 ^ 3.
Don't let the formula get in your head, but if you may have notice that the order of hits and misses are totally irrelevant.


This means, as long as you're missing 3 out of 300, 3 misses in a row and 1 miss every other 99 is exactly the same.
The only difference is, you immediately notice when you get 3 misses in a row and think the game is so cheating. When you miss 1 every ~100, you think that's very natural because 99% isn't exactly 100%.

"The law of large numbers" is a mathematical law that describes how random outcomes of fixed percentage get close to that percentage as the number of tries get bigger.

If you threw 50 of 99% attacks and missed 3, that doesn't seem right at first. But as you throw more attacks (as in hundreds and thousands more times, because you're likely to play for dozens of hours =P) the cumulative hit rate becomes closer and closer to 99%.


Sorry about that big wall of math. I just wanted to throw a BIG rant and it happened to be about FFTA2 and its hit percentage.

If you get 3 misses in 50 attacks this is the equivalent of rolling 2 dice and the sum being 12 on 8 occasions in 50 rolls, so you think that the chance of getting a sum of 12 is more likely than it actually is.

If you would like to actually see this happening then I suggest you take a look at http://nrich.maths.org/public/viewer.php?obj_id=6033
Try spinning the spinner a few times and you'll see that you get a few results and it may make it seem like getting some sums is impossible,
however if you choose Run 50,000 you'll see that the graph evens out to the actual odds of getting each total. (Except the ones that are impossible)

So like deathfisaro says if you base the odds on very few tests then it may seem extraordinary however so long as you dilute these results with hundreds more, the odds will even out.

 

[Hyperlisk_]

Try spinning the spinner a few times and you'll see that you get a few results and it may make it seem like getting some sums is impossible,
however if you choose Run 50,000 you'll see that the graph evens out to the actual odds of getting each total. (Except the ones that are impossible)

That's what he means by the Law of Large Numbers. As the number of trials approaches infinity, the proportion of successes approaches it's true value.

I actually liked statistics. I took AP Statistics this past year. I think I totally aced the AP Test. Waiting for my results though~

 

[triple-sevenz]

i dont think the game is cheating me but i really hate it when i miss but the enemy hits me with a lower hit chance.

 

[Treflex]

I don't think it's actually fake, but at times it seems to get a bit cruel. I was doing one of the bonga bugle quests, the one with all the ghosts and about 30 minutes in to the battle I just had my tinker, gladiator, and Luso left and I kept casting Red Spring for haste, about 6-7 times, and it landed on the enemies every time >.< I still won though, somehow.

 

[Frederica Bernkastel]

I think its fake.

 

[rustinr]

I don't think it's actually fake, but at times it seems to get a bit cruel. I was doing one of the bonga bugle quests, the one with all the ghosts and about 30 minutes in to the battle I just had my tinker, gladiator, and Luso left and I kept casting Red Spring for haste, about 6-7 times, and it landed on the enemies every time >.< I still won though, somehow.

The only thing cruel there is having a tinker in your party

 

[GDragon]

I thinks its fake, i mean i missed with 97% -__-

 

[ivykinz]

birth control has a 98% chance of preventing pregnancy and look how many bastard children there are.

 

[DeMoN]

I definitely think it's fake for Looting with a thief, but you need to cheat to give your character infinite moves to see for yourself.

 

[Tenkaichi]

birth control has a 98% chance of preventing pregnancy and look how many bastard children there are.

I'm not sure if I'm supposed to feel bad that I laughed at that, but it's so true.

Now pertaining to the topic; it seems like FFTA2 for me has been falling with the second option (I have my doubts). I don't particularly think that it's working against me as much as possible, but at times it seems somewhat random (if that). I personally thought that FFTA (GBA) was pretty messed up in regards to missing a lot more often than I thought I should, but the system they reworked feels more stable to me, even though I haven't played too much of it yet. I had noticed the difference in damage from the front vs. side vs. back, and it makes me feel like I'm not as cheated for making an effort to get behind the enemy - IIRC there was only damage difference if you attacked from the front vs. the back (for greater hit chance + some damage) but this works out well, I think. I'll come back for another comment if I think it deserves with more playing.

 

[rafcor]

And you can easily hit people from the front when it would have been different in another FFT game.

I'm pretty sure they changed that. Now hitting from the side increases the damage you deal, and the back increases it a bit more. Someone correct me if I'm wrong though.

Yeah, I noticed that too. I find it kind of annoying however, being used to the other FFT games...

the game tells you that, don't you read the story line!!!
and height also decrease/increase the attacks

 

[Tenkaichi]

And you can easily hit people from the front when it would have been different in another FFT game.

I'm pretty sure they changed that. Now hitting from the side increases the damage you deal, and the back increases it a bit more. Someone correct me if I'm wrong though.

Yeah, I noticed that too. I find it kind of annoying however, being used to the other FFT games...

the game tells you that, don't you read the story line!!!
and height also decrease/increase the attacks

In FFTA I think it didn't play quite as big a role as it does in FFTA2. FFT the emphasis is more on the back and the height (somewhat? can't remember), but not the side.

Some people also don't read signs in front of their own faces, but ah, that's a discussion that I have at work all the time...

 

[deathfisaro]

99% isn't a big percentage.

If 99% of the people will live through today to see tomorrow, that means 65 MILLION people are dying overnight.

More in-context with the game:
If you've played FFTA2 for 20 hours, that means you're likely to have missed 99% attacks more than 50 times. Yes fifty, not fifteen.

Well, that's only if the game is NOT cheating.

If the game's cheating you'd miss 30 of 99% attacks every hour or something.

 

[Kaos]

I thinks its fake, i mean i missed with 97% -__-

You had a 3% chance to miss. You missed. Where's the cheating?