<sigh> Someday, this is going to come back and haunt me, but...

All things being equal, (ship type, figs, cash, turns, last sector, obviously you can't have the same player or ship numbers...) If you have a situation where the fed limit is 5, and you have 6 people who park in there, the person who parked 3rd will get towed the most often.
1st person there 16.6%
2nd person there 16.6%
3rd person there 18.2%
4th person there 16.6%
5th person there 16.6%
6th person there 15.4%
Note how the person who parks LAST is the LEAST likely to get towed.

These numbers were based off of 1600 externs run in 8 200 extern batches. each batch was within +-0.5% of each other. If it were totally random, these numbers should all be 16.6%. (If you take things a bit further, and you subtract 16.6 from 18.2, you are left with 1.6, or 10% of the base chance. I have used some rounding in the above numbers, but when you carry it out to 3 places, it's the same thing, which makes me think it's not a coincidence. Especially when you take 16.6 and subtract 15.4 and you end up with .8, which is 1/2 of 1.6.)

I have more data (with different amounts of figs, etc...) that shows some interesting trends, but I can't rule out a statistical glitch in those cases.
FYI

Wow Traitor. You have more time and patience than I do. I would never have gone to all that work. Thanks for that information. Think i will add it to my Tradewars articles. [:)]

quote:Originally posted by Traitor

<sigh> Someday, this is going to come back and haunt me, but...

All things being equal, (ship type, figs, cash, turns, last sector, obviously you can't have the same player or ship numbers...) If you have a situation where the fed limit is 5, and you have 6 people who park in there, the person who parked 3rd will get towed the most often.
1st person there 16.6%
2nd person there 16.6%
3rd person there 18.2%
4th person there 16.6%
5th person there 16.6%
6th person there 15.4%
Note how the person who parks LAST is the LEAST likely to get towed.

These numbers were based off of 1600 externs run in 8 200 extern batches. each batch was within +-0.5% of each other. If it were totally random, these numbers should all be 16.6%. (If you take things a bit further, and you subtract 16.6 from 18.2, you are left with 1.6, or 10% of the base chance. I have used some rounding in the above numbers, but when you carry it out to 3 places, it's the same thing, which makes me think it's not a coincidence. Especially when you take 16.6 and subtract 15.4 and you end up with .8, which is 1/2 of 1.6.)

I have more data (with different amounts of figs, etc...) that shows some interesting trends, but I can't rule out a statistical glitch in those cases.
FYI

Thanks for the info, didn't know that.

I should also mention that the experience and alignment of those test dummies above were 0 and 0. The above test I ran a couple of years ago under ZOC. (I was trying to figure out how to pry someone out of fed. We knew where he was parking, and it was fun trying to always put him in the sweet spot. We played the odds, and luck was on our side. We got him four times in five days. He stopped logging in after that. [:D])

The actual testing takes less time than you might think. The above one only took about 20 hours, mostly unatended, counting writing the scripts. I use a LOT of scripts to test things like this. I can't thank Eleq enough for telling me about ZOC way back when, and Xide for making TWX. Without those two tools, these tests like this wouldn't be possible in a reasonable amount of time.

TW is filled with patterns. There are similar 10% of 10%, and 5% of 10% patterns all over the place. It's an important factor for port pricing too. And if you pay attention, you will see it happens in combat also. (Like when you are in a merf and trying to kill a single def fig. The % chance that you will loose a specific amount of figs follows a similar pattern.) And in planet collisions. (SIMILAR, not the same tho. I have a higher chance of getting the planets I want to collide to collide than if I just left it to chance. [;)]) After a while, the patterns tend to leap out at you, and you end up predicting the results of your own tests. This thread has got me curious about towing again, so I'll probably run a few more tests and see if I can pin down those glitches.

quote:Originally posted by Traitor

<sigh> Someday, this is going to come back and haunt me, but...

All things being equal, (ship type, figs, cash, turns, last sector, obviously you can't have the same player or ship numbers...) If you have a situation where the fed limit is 5, and you have 6 people who park in there, the person who parked 3rd will get towed the most often.
1st person there 16.6%
2nd person there 16.6%
3rd person there 18.2%
4th person there 16.6%
5th person there 16.6%
6th person there 15.4%
Note how the person who parks LAST is the LEAST likely to get towed.

These numbers were based off of 1600 externs run in 8 200 extern batches. each batch was within +-0.5% of each other. If it were totally random, these numbers should all be 16.6%. (If you take things a bit further, and you subtract 16.6 from 18.2, you are left with 1.6, or 10% of the base chance. I have used some rounding in the above numbers, but when you carry it out to 3 places, it's the same thing, which makes me think it's not a coincidence. Especially when you take 16.6 and subtract 15.4 and you end up with .8, which is 1/2 of 1.6.)

I have more data (with different amounts of figs, etc...) that shows some interesting trends, but I can't rule out a statistical glitch in those cases.
FYI


NICE WORK!