Choose another port or find a bunch of eq and sell it to the port to bump up it's level. Not sure if Ram's is smart enough to use that tho. Keep in mind that if you sell off to the port to bump up it's level you'll increase the amount on-port, which'll decrease it's offer, so you'll make a little less money. Stay at the lowest level you need (5k? 10k? don't remember) so that your haggles will happen near 0% and you can fix that tho.

Frankly I think I'd just use another port.

Cheers Sing,
I tried selling Equ before i posted and it never worked.
That was my best port too lol

I cant believe it wont work with a maxed port... that sucks!!

Just a reminder...old twx rounded down in many cases where the new twx rounds up. So a steal factor of 22 instead of 21 might work better in scripts written for old twx.

Nod, eventually SDT ports get worn out because of constant upgrades over time. Always have backup ports ready. You could write a teamSDT that checks for eq on hand first, but you'd still have the problem of putting EQ on the port (ptrade to it I guess).

Actually, the price improvement due to experience is constant above 1k.

+EP+

When I tested it I found very small increases, but it could've just been due to the randomness... still it seemed to increase.

May be superstition on the jinxed ship, but it sure seems like I get into a bust cycle and see the same ship number a lot. Granted 50:50 on which ship and like the bust rates longer term stealing might even out the number of times I see the same ship number, but still.........

I had never tried tried EP's script so I thought it might be good to see and I am guessing that I am not doing something right. I am using twx 2.04 final....

Loading script: EP_Beta06i.cts

Fighters: 1 (yours) [Defensive]
Script load error: File is not a compiled TWX script


Script terminated: EP_Beta06i.cts

               / Promethius / Enigma / Wolfen /

"A man who has no skills can be taught, a man who has no honor has nothing."

nod. It's called the law of averages and "regression to the mean." It's where you have a 1 in 50 chance, or 2% chance, of busting per port. Over the course of any given subset of the run your odds may be higher, but over larger and larger amounts of runs you will experience the "regression to the mean" where your odds more closely regress to match the average numbers. Oftentimes gamblers will experience either the leading or trailing edge of this and call it good luck or bad luck accordingly, and then find themselves in a reversal later on. Hence the saying "quit while you're ahead." This whole thing leads to a lot of funky superstitions, "systems" and unproven assertions.

But sometimes there are ways to manipulate the odds themselves. An example would be blackjack. Back in the days of a single or double deck shoe you could effectively count cards and bet higher when the odds were in your favor, betting only the minimum when the odds were against you. This led to a lot of changes in the way the game is played, but the strategy still works in smaller games. There was a time when a disciplined skilled player could make a lot of money doing this. Similar strategies are used in today's hedge funds, hence their ability to bring in a 20% ROI despite the risk.

I've wondered whether or not there was a way to improve your busting odds in a similar way. I've tried to track a lot of variables, MCIC, ship number, player number, scores, timing between steals, etc. I've never found any reliable correlations I could prove, but that doesn't mean they don't exist. I do however have some unproven suspicions that I've used across games... and I have tended to get better results (altho it also correlates to my game experience too, so it's not something I can isolate) than I used to.

One test I've thought about doing, but never bothered to... if you find yourself busting a lot in a particular ship, you can replace that ship and continue testing... track your busts and see if, over 1000 busts, you get better or worse results (did it take more, less or equal steals to reach 1000 busts than before?). You could even do this with a worldSST script by simply tracking the results and swapping out the ship w/ the most busts. Of course if you ran worldSST for 1000 ports... would the 2 ships be about even on busts or would 1 ship have far more than the other?

Then again there's another aspect to this... small numbers don't always match the odds. If you flip a coin once, you still have a 50:50 chance of a heads or tails, even if one side is weighted. Why? Because on a single flip you still only have 2 possible outcomes, and 1 result between them. How often do you reach 1000 busts in a game? By then do you even care? I can't imagine a turns game reaching a high confidence interval within a time frame that matters, and while an unlim might be able to do it quicker... odds are everyone will be dead by then anyway. It's not exactly like your average 7 day truce player is going to be concerned over an extra 1b credits after running worldSST the entire time.

So for academic curiosity... it's interesting. For practicality? I dunno. But maybe in a turns game, if you bust a lot in a particular ship, it'd be smarter to swap it out with another ship number or switch to another port. Could be worth a try.

Hey Prom,
I just downloaded TWX 2.04 final and the latest haggle from GrimyTrader.com, and the script works fine. I'm happy to help you troubleshoot on ICQ if you'd like.

Sing, if you saw an improvement in offers with Exp over 1k, it was due to one of the other factors changing or the random variance.

Speaking of randomness... I believe that is exactly what busts are governed by. Let me qualify my opinion by stating that I've never trended busts. However, I researched port offers for months, even after breaking the trade and haggle algorithms, trying to pin down the behavior of the factor that I refer to as baseVariance. Kind of vague, but every other factor was related to Exp, MCIC, day of week, percentage, or some other concrete stat. And yet there is this one random factor that has a small range of possiblities, and I can determine which of the possible values the port has selected to use based on it's first offer, yet I hoped to be able to determine ahead of time which possibility it would select. I went so far as to create a script that ported, over bidded to cancel the trade, checked the server's time, and repeated. It could do this about 3 times in a server second (as in CT), with no other trade factors changing, yet the baseVariance factor was genuinely random and changed even between offers within the same server second. It was at this point that I concluded... you guessed it... it was random. So being convinced that TW really does operate on some random factors, it's easy for me to believe that busts are simply 1:50, independent of <insert favorite superstition>.

+EP+

Sing,
Let me know how the ship number thing goes. I know in my head that it's random. But in my heart, I think that there are cursed ship numbers.