Royal Challenge

[jkwongfl_7919855]

What most of you guys are forgetting is that the randomness in the game is technically pseudo-random and you generally cannot tell pseudo-random from truly random results. You cannot generate true randomness with a computer and pseudo-random results are obtained by plugged some sort of "seed" into a formula. The seed is usually determined by time or some other constantly changing number that the game keeps track of. So concepts such as diminishing probability do not always work all that well with computer generated randomness.

And if this is true, then this computer-generated "pseudo-randomness" is as close as we can get to true randomness in a virtual game. So we can't complain if this is the best we get.

[jkwongfl_7919855]

Thanks for letting me know I'm stupid, jkwong. I understand statistics. However, I was wondering if I wasn't aware of something. I understand that I could do this 1000 times and never get Herja, but it is very unlikely. It is very unlikely, though possible, that 11 attempts fail. A very unlikely probability hinted that I may not have been aware of something, such as having to recruit all preceeding heroes before getting to the next one, which is why I asked.

Lol I wasn't letting you know you were stupid, I was just trying to get some others to see the point. I actually think you are more logical than most other Tynoners. And although there are undoubtedly some flaws in the game, I am very sure that the recruit chances are not rigged nor are there any special prerequisites as you suggested.

And PS, I go by jkwongfl Or Twy.

[mattp169]

no you dont have to have all the previous heroes to get the next one. I thought it was an average number of wins that worked but that seems to be untrue. Most people I have talked to have recruited pompey in 15 tries or less, yet it took me 50 wins to get him. Bryn took me 30 like a bunch of others, General A took like 60 or so like a bunch of others. But I got herja in like 2 tries, lance in about 10, yet I hear people who have hit purple heroes 70 times and won without a recruit. But since EVERYTHING else is predetermined in this game, I suspect this is as well.

[g346007_1906957]

I have a feeling something is wrong with the probabilities as well. I've found a few things I've been particularly unlucky in (10 star chances in a row failing at 45%, 30 misses at a 10% recruit)
A 10% chance could mean 90% for 1 out of 10 people and 1-2% for the other 9. Unless we have some data showing distributions are somewhat normal, it might be best to not be too hopeful... tynon likes to be tricky with implied probabilities.

(In this case, failing the first 10 or so will actually indicate a less than 10% probability of recruiting since it is very likely you are one of the 'unlucky 9'. Purely made up example though)

[Rockstar]

Just keep on trying and you will get what you want

[DrthCaedus]

Looking at the set of all flips, the probability of having 99 consecutive heads followed by a tails would be very unlikely, but looking at the single flip by itself the probability of flipping a tails would just be 1/2. It's a matter of viewpoint, you could say; looking at the big picture differs from focusing on specific events. The in-game probabilities operate on a single-event probability; that is, each attempt has its own chances. The game is not engineered so that after 100 tries you'll have 20 recruits. We players tend to think in terms of consecutive-events, taking into account previous failings and expecting a likelier favorable outcome in the future. Tynon doesn't do that.

We can both agree, though, that there is nothing wrong with the recruit chances. It's just that some people are "unfortunate" in their endeavors.

The concept of diminishing probability still applies in this situation, even if you look at it like a single-event. the odds of failing to recruit herja, for example, are 4/5. so you have an 80% chance of failing to recruit her each attempt. Attempt twice in a row the odds for each attempt are the same single-event odds, 4/5 and 4/5. but your overall odds of failing both times are 16/25 or 64%. Attempt a third time 4/5, 4/5, and 4/5 odds each time, but overall odds are 64/125 or 52%. For each single-event there is a set probability of recruitment, however, there are also the odds of achieving the same result over and over again. The more often you achieve the same result under the same conditions, the less likely it is for that same result to occur again. This is why there is such an unlikely probability you would flip heads 99 times in a row (roughly 1 in 633,825,300,000,000,000,000,000,000,000). So the odds of the OP not recruiting herja in 11 attempts would be: 4,194,304/48,828,125 or 8.5%. And it continues to decrease from there. The fact is that Tynon continually proves that results must be pre-determined because they fail to comply with the odds they project far to often, and they are too consistently achieving outlier statistical results.

[Rumbleguts]

The concept of diminishing probability still applies in this situation, even if you look at it like a single-event. the odds of failing to recruit herja, for example, are 4/5. so you have an 80% chance of failing to recruit her each attempt. Attempt twice in a row the odds for each attempt are the same single-event odds, 4/5 and 4/5. but your overall odds of failing both times are 16/25 or 64%. Attempt a third time 4/5, 4/5, and 4/5 odds each time, but overall odds are 64/125 or 52%. For each single-event there is a set probability of recruitment, however, there are also the odds of achieving the same result over and over again. The more often you achieve the same result under the same conditions, the less likely it is for that same result to occur again. This is why there is such an unlikely probability you would flip heads 99 times in a row (roughly 1 in 6,338,253,000,00,000,000,000,000,000,000). So the odds of the OP not recruiting herja in 11 attempts would be: 4,194,304/48,828,125 or 8.5%. And it continues to decrease from there. The fact is that Tynon continually proves that results must be pre-determined because they fail to comply with the odds they project far to often, and they are too consistently achieving outlier statistical results.

A very nice analysis. However, you don't appear to be taking into consideration the spread of possibilities across all the servers. Yes the original poster may only have that 8.5% chance of failing to recruit on their last attempt, but others have recruited in very few attempts, I myself got Lance du Lac in 3 attempts. We would have to have results from every player on every server (and accurate reports too, not just, "I seem to remember" results) to say that there is an error in the probability generator. And we don't have that kind of a data set.

[DrthCaedus]

A very nice analysis. However, you don't appear to be taking into consideration the spread of possibilities across all the servers. Yes the original poster may only have that 8.5% chance of failing to recruit on their last attempt, but others have recruited in very few attempts, I myself got Lance du Lac in 3 attempts. We would have to have results from every player on every server (and accurate reports too, not just, "I seem to remember" results) to say that there is an error in the probability generator. And we don't have that kind of a data set.

Very true, and I'd love to see that data so us players would feel less frustrated with our results. A large amount of players come to the forums and report it taking much longer than the prescribed probability to recruit heroes, achieve stars, get a successful sainting, etc.... However, this is just a small sampling of the total amount of players within Tynon, and they might not be an accurate representation of overall results, but there is no data to refute it. Also, you can use the forums much like Gallup uses a poll. A small cross-section of the overall amount of players within Tynon post in the forums, and could be construed to represent statistically the overall outcomes, opinions, and experiences throughout the entire player-base.

Personally I would just love to see Tynon release some actual figures to show that they're predicted probabilities are an accurate and true representation of overall success and failure throughout the game.

[Rumbleguts]

Very true DrthCaedus. And one thing I always remember when thinking about problems reported on forums like this is people are far more likely to complain about a perceived problem then to say when things are going good. That creates a lot of bias.