Please upgrade your browser for the best possible experience.

Chrome Firefox Internet Explorer
×

RE- Proc Rate


LauKensai's Avatar


LauKensai
03.17.2014 , 08:53 PM | #1
Proc rate for reverse engineering not 20 percent.

Less than 1 hour ago crafted...
5 aim lvl 15 armorings
5 cunning lvl 15 armorings
5 willpower lvl 15 armorings
5 strength lvl 15 armorings

Got ...

8, 7, 6, and 5 with crits...

RE the 7, no proc
RE the 6, proc on number 6
RE the 5, no proc
RE the 8, no proc

That's 1 proc in 26. The math on that comes to 3.8 percent....a far far cry from 20.

I'm not asking for extra procs on crafting. I'm not looking to get rich quick on the GTN by upping the crit rate on crafting. Everything is fine as is for MAKING items. But for RE you obviously intended 20 percent. 3.8 percent is not 20 percent. It's the difference between RE'ing something 5 times, or 26.3 times.

On RE number 5 the crafting mat should proc. That's 20 percent. That's what you advertise. Please fix.

StrikePrice's Avatar


StrikePrice
03.17.2014 , 09:47 PM | #2
ib4t "your sample size is not big enough".

There's a huge thread on this. It's obviously a bug. Neither the community nor the dev team will acknowledge.

psandak's Avatar


psandak
03.18.2014 , 07:59 AM | #3
Quote: Originally Posted by LauKensai View Post
Proc rate for reverse engineering not 20 percent.

Less than 1 hour ago crafted...
5 aim lvl 15 armorings
5 cunning lvl 15 armorings
5 willpower lvl 15 armorings
5 strength lvl 15 armorings

Got ...

8, 7, 6, and 5 with crits...

RE the 7, no proc
RE the 6, proc on number 6
RE the 5, no proc
RE the 8, no proc

That's 1 proc in 26. The math on that comes to 3.8 percent....a far far cry from 20.

I'm not asking for extra procs on crafting. I'm not looking to get rich quick on the GTN by upping the crit rate on crafting. Everything is fine as is for MAKING items. But for RE you obviously intended 20 percent. 3.8 percent is not 20 percent. It's the difference between RE'ing something 5 times, or 26.3 times.

On RE number 5 the crafting mat should proc. That's 20 percent. That's what you advertise. Please fix.
So basically you had one 0 for 12 streak and one 0 for 13 streak with a single success in between. The chance of 12 consecutive failures is (80%^12 or) 6.872%, better than 1 in 20. The chance of 13 failures is (80%^13 or) 5.498%, still better than 1 in 20. Neither is beyond the realm of probability.

I did a run of 1500 back in October when that long thread was started, and my results were about 18%, well within a reasonable margin of error for even my small sample.
The Ortzid Legacy on The Harbinger
Car'beerd (Guardian) Dalkery (Scoundrel) Blairnah (Sage) Ulunnes (Commando)
Zaggema (Sin) Daellia (Merc) Carrud (Marauder) Quith'edi'ata (Operative)


high end schematics * best crew skills * crew skill money making * Probability 101
"You can't just ask customers what they want and then try to give it to them. By the time you get it built, they'll want something else" Steve Jobs

Grayseven's Avatar


Grayseven
03.18.2014 , 08:20 AM | #4
Quote: Originally Posted by StrikePrice View Post
ib4t "your sample size is not big enough".

There's a huge thread on this. It's obviously a bug. Neither the community nor the dev team will acknowledge.
It isn't "obviously a bug", just people who fail to realize that 20% chance does not mean that you should get 1 critical every 5 RE's.

If you are a gamer and have access to a D10, try a little experiment. Pick any two numbers on that die and start rolling. Keep track of how many rolls you do and keep track of the times you land on one of those two numbers. I can guarantee that you'll run into long streaks where neither of those numbers is hit and that the 20% chance you have to hit one of those two numbers will not be the percentage of times you hit those numbers for small sample sizes.

Now, if you devote 10 minutes a day doing that and keep track of all the rolls and hits and do that for a few months, you'll start seeing the percentage of hits getting close to 20% and the longer you do it to closer to 20% you'll get.

The only thing wrong is that you had a couple of streaks of bad luck. We've all had it. I just had Tano craft 5 Mk-9 kits while I was offline a week ago and when I logged back in he crit'd on all 5 kits that I queued up. I've never seen that before and don't expect to see it again but I'm not blaming the RNG.
"50 Grades of Shae", a heart-warming novel about a Mandalorian that delivers beat-downs and assigns grades to her victims.

wainot-keel's Avatar


wainot-keel
03.18.2014 , 09:21 AM | #5
Quote: Originally Posted by StrikePrice View Post
ib4t "your sample size is not big enough".

There's a huge thread on this. It's obviously a bug. Neither the community nor the dev team will acknowledge.
I hate to say it, but yes. It is a small sample. And even when it was bugged/stealth nerfed, we cannot reasonably say so with only few cases

Similar case with crits on crew missions.
Since 2.4, I've been keeping track of the results of the "Unsliceable?" mission, using four +2 slicing companions and two others without bonus.

My "first" 550 missions or so (from 2.4 to 2.5 ) were giving me a crit rate of almost 24% (for companions, at 10k affection, that mission crit chance is, theoretically, 20%). Then, ouch, the next 350 missions ( from 2.5 to 2.6) that crit rate dropped to 18%.... That was certainly frustrating....
But now, with almost 1100 missions, if I consider the overall crit rate since I started, I have 21, 87% crit rate. Considering those 1100 missions were evenly ran by four companions at 22 %chance and two at 20%, I'd say I'm getting pretty much what I'm supposed to get in terms of crits....

The sample size is very important and 60 is definitely not big enough....

MaximilianPower's Avatar


MaximilianPower
03.18.2014 , 02:03 PM | #6
Quote: Originally Posted by StrikePrice View Post
ib4t "your sample size is not big enough".

There's a huge thread on this. It's obviously a bug. Neither the community nor the dev team will acknowledge.
Because you're wrong. No offense, but if you really think OP's figures represent a useful sample size, then your understanding of statistics is severely limited. It's been explained many times, in many different ways, in this thread and elsewhere. You absolutely cannot draw any conclusions based on the kind of evidence presented here.

I'll say it again to be perfectly clear: there is nothing wrong with RE rates.
Hey, Barsen'thor Bill
What did you kill
Barsen'thor Bill

Zuckerkorn Zuckernaut Spiderzuck Bøbløblaw EddieBrøck SylvesterMcbean Zuck-srw'i
TøbiasFünke MaebyFünke MaximilianPower MaxPower-srw
Jedi Covenant / Shadowlands / Harbinger / The Ebon Hawk / Bastion / Begeren Colony /The Red Eclipse / Progenitor


Shadowlands Reconnaisance Wing / Retrocide / Imperial Entanglements

psandak's Avatar


psandak
03.18.2014 , 03:37 PM | #7
In response to the OP's statistics I have generated statistics of my own. This morning I started leveling armstech on my gunslinger (the start of a process to "optimize" my crafting crew skills). I only used materials I have been saving so I was short on some investigation materials (resulting in an inability to attempt many blue to purple REs) and am only crafting barrels. But so far my results are polar opposite of the OP's; my rate is far better than 20%....it is currently 26.875%. REing 160 items I have had 117 failures and 43 successes. Not to mention a current streak of FIVE successes in a row (FYI my longest single item failure streak was 14. I also had two 7s, two 8s, and a 9)

For the most part I have been REing as the units are crafted, and stopping production of an item once I get the better schematic (this has resulted in a number of "leftovers" from crit crafting). In a few cases, I had to craft items just to get to the next trained schematic; these crafts are not counted in my RE rate calculations because there was no schematic to be gotten from them.

Just for confirmation here are two images of when I started and where I am now:

http://i86.photobucket.com/albums/k9..._43_879680.jpg
http://i86.photobucket.com/albums/k9..._33_222116.jpg
The Ortzid Legacy on The Harbinger
Car'beerd (Guardian) Dalkery (Scoundrel) Blairnah (Sage) Ulunnes (Commando)
Zaggema (Sin) Daellia (Merc) Carrud (Marauder) Quith'edi'ata (Operative)


high end schematics * best crew skills * crew skill money making * Probability 101
"You can't just ask customers what they want and then try to give it to them. By the time you get it built, they'll want something else" Steve Jobs

Kerensk's Avatar


Kerensk
03.18.2014 , 05:10 PM | #8
People just don't understand statistics.
If you flip a coin 100 times and get heads 100 times you're not "due to get tails" because past flips don't physically affect the coin to increase its chances.
Your next flip is still 50/50.

The chance of a long losing streak is small (1 in 20,like they already said) but your next RE still has a 4 to one chance to fail, and the odds will be 4 to 1 against you each time.
The Harbinger, Piett Legacy
“It's an older quote but it checks out. I was about to clear it. ”

psandak's Avatar


psandak
03.18.2014 , 05:59 PM | #9
Just finished for the afternoon. Got my armstech to 410 (did not have grade 9 materials on the character and am breaking for food ). Check out my results at: https://drive.google.com/file/d/0By5...it?usp=sharing

My sample is still small but in 213 total crafts destined for RE, resulting in 263 total items to be REed, I got 58 successes with 24 "leftovers" (units not REed for schematic because I got the schematic prior to using them for RE), for a success rate of 24.27%

As an added bonus I also tracked my crafting crit rate and got 19.77% (no legacy of crafting perk)
The Ortzid Legacy on The Harbinger
Car'beerd (Guardian) Dalkery (Scoundrel) Blairnah (Sage) Ulunnes (Commando)
Zaggema (Sin) Daellia (Merc) Carrud (Marauder) Quith'edi'ata (Operative)


high end schematics * best crew skills * crew skill money making * Probability 101
"You can't just ask customers what they want and then try to give it to them. By the time you get it built, they'll want something else" Steve Jobs

StrikePrice's Avatar


StrikePrice
03.18.2014 , 07:48 PM | #10
Quote: Originally Posted by Grayseven View Post
It isn't "obviously a bug", just people who fail to realize that 20% chance does not mean that you should get 1 critical every 5 RE's.
I'm just going to keep saying it b/c I think it needs to be investigated.