GOLFMK8
GOLFMK7
GOLFMK6
GOLFMKV

Important PSA regarding improperly built TPC20 Turbos

manu97

Autocross Champion
Location
Chicago
Car(s)
MK7 R
You have some logs we can look at? For science...
...where are the 4 failed turbos logs?

for science







also i have no horse in the race, I'm on an XL, but your conclusions wouldn't pass the sniff test in a high school stats class
 

ZuMBLe

Autocross Champion
Location
NY
Car(s)
Alltrack 6MT
...where are the 4 failed turbos logs?

for science







also i have no horse in the race, I'm on an XL, but your conclusions wouldn't pass the sniff test in a high school stats class
There is one example in in the first post. Did you see the first post?

I asked for his log, and there are examples of my log in the first post. What doesn't pass your high stats class? I think your logic if flawed. Then you say you have no horse in the race, but obviously you are trying to start shit because you thought your comment was real smart. Oops.
 

manu97

Autocross Champion
Location
Chicago
Car(s)
MK7 R
There is one example in in the first post. Did you see the first post?

I asked for his log, and there are examples of my log in the first post. What doesn't pass your high stats class? I think your logic if flawed. Then you say you have no horse in the race, but obviously you are trying to start shit because you thought your comment was real smart. Oops.
Lmao what doesn't pass high school stats class?


Ok, first things first, you've got sampling bias. If you want to extrapolate, ever, you need a truly kinda random sample set. Yours isn't.

Even then, you need a sufficiently large sample size, and nobody would think that N=6 is sufficient to assume normality.

There's ya sniff test. Oops.
 

ZuMBLe

Autocross Champion
Location
NY
Car(s)
Alltrack 6MT
Lmao what doesn't pass high school stats class?


Ok, first things first, you've got sampling bias. If you want to extrapolate, ever, you need a truly random sample set. Yours isn't.

Even then, you need a sufficiently large sample size, and nobody would think that N=6 is sufficient to assume normality.

There's ya sniff test. Oops.
I explained the "bias" MANY MANY times now. It's not a bias if that's all you have to work with. Is that really hard to understand?

If you don't want to extrapolate, that's fine, nobody is forcing you to. It's funny how the numbers are above 60% for the small sample we have. Totally unplanned. Carry on.
 

manu97

Autocross Champion
Location
Chicago
Car(s)
MK7 R
I explained the "bias" MANY MANY times now. It's not a bias if that's all you have to work with. Is that really hard to understand?
...it's still bias...even if it's all you have to work with...you must acknowledge the bias.


The bias doesn't just get ignored because you don't have access to randomized data sets lmao
 

ZuMBLe

Autocross Champion
Location
NY
Car(s)
Alltrack 6MT
...it's still bias...even if it's all you have to work with...you must acknowledge the bias.


The bias doesn't just get ignored because you don't have access to randomized data sets lmao

Umm.. I said 6 turbos, 4 are bad is 66%. Did I say 60% of TPC turbos are definitely bad? No. I said I wouldn't be surprised if 60% were bad. And look, "randomly" we came upon that number. Huh.

I think you're making assumptions and injecting them as facts. We should stop. I'll agree to disagree with you.
 

scrllock

Autocross Champion
Location
MI
for what its worth, i have a tpc770, and i know of 2 others that are doing just fine, one of them, a member here is all over instagram whos been beating the dog shit out of a tpc20 and now a tpc770.

so there is 3 that are not blown up.

shrug.
the people who use one working product as evidence of... something? are hilarious.
 

jay745

What Would Glenn Danzig Do
Location
Slightly Outside Chicago
Car(s)
Mk6 racecar, Tacoma
I explained the "bias" MANY MANY times now. It's not a bias if that's all you have to work with. Is that really hard to understand?
You're literally saying your data is garbage which is what I've been saying all day. Hence the talking in circles. Glad you're finally starting to understand though
 

manu97

Autocross Champion
Location
Chicago
Car(s)
MK7 R
Umm.. I said 6 turbos, 4 are bad is 66%. Did I say 60% of TPC turbos are definitely bad? No. I said I wouldn't be surprised if 60% were bad. And look, "randomly" we came upon that number. Huh.

I think you're making assumptions and injecting them as facts. We should stop. I'll agree to disagree with you.

But since we don't have TPC's numbers, we can only work with what we have. We have to extrapolate.

1684275614784.png


You kinda did...
 

ZuMBLe

Autocross Champion
Location
NY
Car(s)
Alltrack 6MT
You're literally saying your data is garbage which is what I've been saying all day. Hence the talking in circles. Glad you're finally starting to understand though

I keep saying data set is not great, but we have nothing else to work on. This has been my stance all along. Please show me where I've said otherwise. You're making things up in your mind and trying to make it fact. If you give me the real numbers, we can work with that. We need the following:

1. Number of TPC turbos sold.
2. Logs of complaints (email and phone)
3. Number of turbos warrantied properly (cuz I've seen one returned with a HUGE crack in the turbine housing, kinda funny).

Without the above, we can look the small sample set numbers which are real, and draw our own conclusions.
 
Top